diff --git a/.gitattributes b/.gitattributes index 180d44765d9c8ab95a37ac0856000963487dff15..daba8bcc4eeb300a3eec44dad01db2d1c791f362 100644 --- a/.gitattributes +++ b/.gitattributes @@ -391,3 +391,4 @@ tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cudnn/lib/ .venv/lib/python3.11/site-packages/numpy/lib/__pycache__/function_base.cpython-311.pyc filter=lfs diff=lfs merge=lfs -text .venv/lib/python3.11/site-packages/numpy/lib/__pycache__/npyio.cpython-311.pyc filter=lfs diff=lfs merge=lfs -text .venv/lib/python3.11/site-packages/numpy/lib/tests/__pycache__/test_function_base.cpython-311.pyc filter=lfs diff=lfs merge=lfs -text +.venv/lib/python3.11/site-packages/mistral_common/data/mistral_instruct_tokenizer_240323.model.v3 filter=lfs diff=lfs merge=lfs -text diff --git a/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/INSTALLER b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/INSTALLER new file mode 100644 index 0000000000000000000000000000000000000000..a1b589e38a32041e49332e5e81c2d363dc418d68 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/INSTALLER @@ -0,0 +1 @@ +pip diff --git a/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/METADATA b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/METADATA new file mode 100644 index 0000000000000000000000000000000000000000..68b17eb3007fcab16e09b0f659810c01bd9ea891 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/METADATA @@ -0,0 +1,138 @@ +Metadata-Version: 2.3 +Name: cryptography +Version: 43.0.3 +Classifier: Development Status :: 5 - Production/Stable +Classifier: Intended Audience :: Developers +Classifier: License :: OSI Approved :: Apache Software License +Classifier: License :: OSI Approved :: BSD License +Classifier: Natural Language :: English +Classifier: Operating System :: MacOS :: MacOS X +Classifier: Operating System :: POSIX +Classifier: Operating System :: POSIX :: BSD +Classifier: Operating System :: POSIX :: Linux +Classifier: Operating System :: Microsoft :: Windows +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3 :: Only +Classifier: Programming Language :: Python :: 3.7 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Programming Language :: Python :: 3.12 +Classifier: Programming Language :: Python :: Implementation :: CPython +Classifier: Programming Language :: Python :: Implementation :: PyPy +Classifier: Topic :: Security :: Cryptography +Requires-Dist: cffi >=1.12 ; platform_python_implementation != 'PyPy' +Requires-Dist: bcrypt >=3.1.5 ; extra == 'ssh' +Requires-Dist: nox ; extra == 'nox' +Requires-Dist: cryptography-vectors ==43.0.3 ; extra == 'test' +Requires-Dist: pytest >=6.2.0 ; extra == 'test' +Requires-Dist: pytest-benchmark ; extra == 'test' +Requires-Dist: pytest-cov ; extra == 'test' +Requires-Dist: pytest-xdist ; extra == 'test' +Requires-Dist: pretend ; extra == 'test' +Requires-Dist: certifi ; extra == 'test' +Requires-Dist: pytest-randomly ; extra == 'test-randomorder' +Requires-Dist: sphinx >=5.3.0 ; extra == 'docs' +Requires-Dist: sphinx-rtd-theme >=1.1.1 ; extra == 'docs' +Requires-Dist: pyenchant >=1.6.11 ; extra == 'docstest' +Requires-Dist: readme-renderer ; extra == 'docstest' +Requires-Dist: sphinxcontrib-spelling >=4.0.1 ; extra == 'docstest' +Requires-Dist: build ; extra == 'sdist' +Requires-Dist: ruff ; extra == 'pep8test' +Requires-Dist: mypy ; extra == 'pep8test' +Requires-Dist: check-sdist ; extra == 'pep8test' +Requires-Dist: click ; extra == 'pep8test' +Provides-Extra: ssh +Provides-Extra: nox +Provides-Extra: test +Provides-Extra: test-randomorder +Provides-Extra: docs +Provides-Extra: docstest +Provides-Extra: sdist +Provides-Extra: pep8test +License-File: LICENSE +License-File: LICENSE.APACHE +License-File: LICENSE.BSD +Summary: cryptography is a package which provides cryptographic recipes and primitives to Python developers. +Author: The cryptography developers +Author-email: The Python Cryptographic Authority and individual contributors +License: Apache-2.0 OR BSD-3-Clause +Requires-Python: >=3.7 +Description-Content-Type: text/x-rst; charset=UTF-8 +Project-URL: homepage, https://github.com/pyca/cryptography +Project-URL: documentation, https://cryptography.io/ +Project-URL: source, https://github.com/pyca/cryptography/ +Project-URL: issues, https://github.com/pyca/cryptography/issues +Project-URL: changelog, https://cryptography.io/en/latest/changelog/ + +pyca/cryptography +================= + +.. image:: https://img.shields.io/pypi/v/cryptography.svg + :target: https://pypi.org/project/cryptography/ + :alt: Latest Version + +.. image:: https://readthedocs.org/projects/cryptography/badge/?version=latest + :target: https://cryptography.io + :alt: Latest Docs + +.. image:: https://github.com/pyca/cryptography/workflows/CI/badge.svg?branch=main + :target: https://github.com/pyca/cryptography/actions?query=workflow%3ACI+branch%3Amain + + +``cryptography`` is a package which provides cryptographic recipes and +primitives to Python developers. Our goal is for it to be your "cryptographic +standard library". It supports Python 3.7+ and PyPy3 7.3.11+. + +``cryptography`` includes both high level recipes and low level interfaces to +common cryptographic algorithms such as symmetric ciphers, message digests, and +key derivation functions. For example, to encrypt something with +``cryptography``'s high level symmetric encryption recipe: + +.. code-block:: pycon + + >>> from cryptography.fernet import Fernet + >>> # Put this somewhere safe! + >>> key = Fernet.generate_key() + >>> f = Fernet(key) + >>> token = f.encrypt(b"A really secret message. Not for prying eyes.") + >>> token + b'...' + >>> f.decrypt(token) + b'A really secret message. Not for prying eyes.' + +You can find more information in the `documentation`_. + +You can install ``cryptography`` with: + +.. code-block:: console + + $ pip install cryptography + +For full details see `the installation documentation`_. + +Discussion +~~~~~~~~~~ + +If you run into bugs, you can file them in our `issue tracker`_. + +We maintain a `cryptography-dev`_ mailing list for development discussion. + +You can also join ``#pyca`` on ``irc.libera.chat`` to ask questions or get +involved. + +Security +~~~~~~~~ + +Need to report a security issue? Please consult our `security reporting`_ +documentation. + + +.. _`documentation`: https://cryptography.io/ +.. _`the installation documentation`: https://cryptography.io/en/latest/installation/ +.. _`issue tracker`: https://github.com/pyca/cryptography/issues +.. _`cryptography-dev`: https://mail.python.org/mailman/listinfo/cryptography-dev +.. _`security reporting`: https://cryptography.io/en/latest/security/ + diff --git a/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/RECORD b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/RECORD new file mode 100644 index 0000000000000000000000000000000000000000..2cbb6e07a7811e38ab9dffcd765a0f629443e582 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/RECORD @@ -0,0 +1,173 @@ +cryptography-43.0.3.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4 +cryptography-43.0.3.dist-info/METADATA,sha256=6zbg5CUehHnvNpZEQHVe8ivt1BG6h6k_cm-o5bsOZLA,5440 +cryptography-43.0.3.dist-info/RECORD,, +cryptography-43.0.3.dist-info/WHEEL,sha256=5SNCVD9cb88a-xAIrDHIo1CvpgNriOYcNgb4b8rPcOw,107 +cryptography-43.0.3.dist-info/license_files/LICENSE,sha256=Pgx8CRqUi4JTO6mP18u0BDLW8amsv4X1ki0vmak65rs,197 +cryptography-43.0.3.dist-info/license_files/LICENSE.APACHE,sha256=qsc7MUj20dcRHbyjIJn2jSbGRMaBOuHk8F9leaomY_4,11360 +cryptography-43.0.3.dist-info/license_files/LICENSE.BSD,sha256=YCxMdILeZHndLpeTzaJ15eY9dz2s0eymiSMqtwCPtPs,1532 +cryptography/__about__.py,sha256=-FkHKD9mSuEfH37wsSKnQzJZmL5zUAUTpB5OeUQjPE0,445 +cryptography/__init__.py,sha256=mthuUrTd4FROCpUYrTIqhjz6s6T9djAZrV7nZ1oMm2o,364 +cryptography/__pycache__/__about__.cpython-311.pyc,, +cryptography/__pycache__/__init__.cpython-311.pyc,, +cryptography/__pycache__/exceptions.cpython-311.pyc,, +cryptography/__pycache__/fernet.cpython-311.pyc,, +cryptography/__pycache__/utils.cpython-311.pyc,, +cryptography/exceptions.py,sha256=835EWILc2fwxw-gyFMriciC2SqhViETB10LBSytnDIc,1087 +cryptography/fernet.py,sha256=aPj82w-Z_1GBXUtWRUsZdVbMwRo5Mbjj0wkA9wG4rkw,6696 +cryptography/hazmat/__init__.py,sha256=5IwrLWrVp0AjEr_4FdWG_V057NSJGY_W4egNNsuct0g,455 +cryptography/hazmat/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/__pycache__/_oid.cpython-311.pyc,, +cryptography/hazmat/_oid.py,sha256=e9yLmxtdQtuL94ztQv3SGtt_ea1Mx6aUwGftJsP6EXk,15201 +cryptography/hazmat/backends/__init__.py,sha256=O5jvKFQdZnXhKeqJ-HtulaEL9Ni7mr1mDzZY5kHlYhI,361 +cryptography/hazmat/backends/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/backends/openssl/__init__.py,sha256=p3jmJfnCag9iE5sdMrN6VvVEu55u46xaS_IjoI0SrmA,305 +cryptography/hazmat/backends/openssl/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/backends/openssl/__pycache__/backend.cpython-311.pyc,, +cryptography/hazmat/backends/openssl/backend.py,sha256=pUXUbugLwMm2Gls-h5U5fw2RvepaNjEvnao6CTmL1xQ,9648 +cryptography/hazmat/bindings/__init__.py,sha256=s9oKCQ2ycFdXoERdS1imafueSkBsL9kvbyfghaauZ9Y,180 +cryptography/hazmat/bindings/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/bindings/_rust.abi3.so,sha256=QrghdFa6x-vG5lFvGVfv-slqoy0UA9a8eHmtp9hzGCk,10862344 +cryptography/hazmat/bindings/_rust/__init__.pyi,sha256=wb1OT76lG19vjq97_q2MM3qdJlQhyloXfVbKFDmRse4,737 +cryptography/hazmat/bindings/_rust/_openssl.pyi,sha256=mpNJLuYLbCVrd5i33FBTmWwL_55Dw7JPkSLlSX9Q7oI,230 +cryptography/hazmat/bindings/_rust/asn1.pyi,sha256=BrGjC8J6nwuS-r3EVcdXJB8ndotfY9mbQYOfpbPG0HA,354 +cryptography/hazmat/bindings/_rust/exceptions.pyi,sha256=exXr2xw_0pB1kk93cYbM3MohbzoUkjOms1ZMUi0uQZE,640 +cryptography/hazmat/bindings/_rust/ocsp.pyi,sha256=R-xJ-XmJZ1lOk-fWHHvRnP3QNTCFnKv-l3xlNWfLVt4,868 +cryptography/hazmat/bindings/_rust/openssl/__init__.pyi,sha256=Lvn250QMdPyeF-hoBF6rkQgHLBJxVauXCb8i8uYTomQ,1368 +cryptography/hazmat/bindings/_rust/openssl/aead.pyi,sha256=i0gA3jUQ4rkJXTGGZrq-AuY-VQLN31lyDeWuDZ0zJYw,2553 +cryptography/hazmat/bindings/_rust/openssl/ciphers.pyi,sha256=iK0ZhQ-WyCQbjaraaFgK6q4PpD-7Rf5RDHkFD3YEW_g,1301 +cryptography/hazmat/bindings/_rust/openssl/cmac.pyi,sha256=nPH0X57RYpsAkRowVpjQiHE566ThUTx7YXrsadmrmHk,564 +cryptography/hazmat/bindings/_rust/openssl/dh.pyi,sha256=Z3TC-G04-THtSdAOPLM1h2G7ml5bda1ElZUcn5wpuhk,1564 +cryptography/hazmat/bindings/_rust/openssl/dsa.pyi,sha256=qBtkgj2albt2qFcnZ9UDrhzoNhCVO7HTby5VSf1EXMI,1299 +cryptography/hazmat/bindings/_rust/openssl/ec.pyi,sha256=zJy0pRa5n-_p2dm45PxECB_-B6SVZyNKfjxFDpPqT38,1691 +cryptography/hazmat/bindings/_rust/openssl/ed25519.pyi,sha256=OJsrblS2nHptZctva-pAKFL5q8yPEAkhmjPZpJ6TA94,493 +cryptography/hazmat/bindings/_rust/openssl/ed448.pyi,sha256=SkPHK2HdbYN02TVQEUOgW3iTdiEY7HBE4DijpdkAzmk,475 +cryptography/hazmat/bindings/_rust/openssl/hashes.pyi,sha256=J8HoN0GdtPcjRAfNHr5Elva_nkmQfq63L75_z9dd8Uc,573 +cryptography/hazmat/bindings/_rust/openssl/hmac.pyi,sha256=ZmLJ73pmxcZFC1XosWEiXMRYtvJJor3ZLdCQOJu85Cw,662 +cryptography/hazmat/bindings/_rust/openssl/kdf.pyi,sha256=wPS5c7NLspM2632II0I4iH1RSxZvSRtBOVqmpyQATfk,544 +cryptography/hazmat/bindings/_rust/openssl/keys.pyi,sha256=JSrlGNaW49ZCZ1hcb-YJdS1EAbsMwRbVEcLL0P9OApA,872 +cryptography/hazmat/bindings/_rust/openssl/poly1305.pyi,sha256=9iogF7Q4i81IkOS-IMXp6HvxFF_3cNy_ucrAjVQnn14,540 +cryptography/hazmat/bindings/_rust/openssl/rsa.pyi,sha256=2OQCNSXkxgc-3uw1xiCCloIQTV6p9_kK79Yu0rhZgPc,1364 +cryptography/hazmat/bindings/_rust/openssl/x25519.pyi,sha256=2BKdbrddM_9SMUpdvHKGhb9MNjURCarPxccbUDzHeoA,484 +cryptography/hazmat/bindings/_rust/openssl/x448.pyi,sha256=AoRMWNvCJTiH5L-lkIkCdPlrPLUdJvvfXpIvf1GmxpM,466 +cryptography/hazmat/bindings/_rust/pkcs12.pyi,sha256=afhB_6M8xI1MIE5vxkaDF1jSxA48ib1--NiOxtf6boM,1394 +cryptography/hazmat/bindings/_rust/pkcs7.pyi,sha256=QCmuA0IgDr4iOecUOXgUUeh3BAjJx8ubjz__EnNbyGY,972 +cryptography/hazmat/bindings/_rust/test_support.pyi,sha256=Xo1Gd7bh9rU4HuIS4pm9UwCY6IS1gInvFwmhABLOVO4,936 +cryptography/hazmat/bindings/_rust/x509.pyi,sha256=WLrGmqmFss8dXKhlG_J9nVhoCcodR72xJdCoxEuBtjY,3551 +cryptography/hazmat/bindings/openssl/__init__.py,sha256=s9oKCQ2ycFdXoERdS1imafueSkBsL9kvbyfghaauZ9Y,180 +cryptography/hazmat/bindings/openssl/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/bindings/openssl/__pycache__/_conditional.cpython-311.pyc,, +cryptography/hazmat/bindings/openssl/__pycache__/binding.cpython-311.pyc,, +cryptography/hazmat/bindings/openssl/_conditional.py,sha256=dkGKGU-22uR2ZKeOOwaSxEJCGaafgUjb2romWcu03QE,5163 +cryptography/hazmat/bindings/openssl/binding.py,sha256=e1gnFAZBPrkJ3CsiZV-ug6kaPdNTAEROaUFiFrUh71M,4042 +cryptography/hazmat/decrepit/__init__.py,sha256=wHCbWfaefa-fk6THSw9th9fJUsStJo7245wfFBqmduA,216 +cryptography/hazmat/decrepit/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/decrepit/ciphers/__init__.py,sha256=wHCbWfaefa-fk6THSw9th9fJUsStJo7245wfFBqmduA,216 +cryptography/hazmat/decrepit/ciphers/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/decrepit/ciphers/__pycache__/algorithms.cpython-311.pyc,, +cryptography/hazmat/decrepit/ciphers/algorithms.py,sha256=HWA4PKDS2w4D2dQoRerpLRU7Kntt5vJeJC7j--AlZVU,2520 +cryptography/hazmat/primitives/__init__.py,sha256=s9oKCQ2ycFdXoERdS1imafueSkBsL9kvbyfghaauZ9Y,180 +cryptography/hazmat/primitives/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/_asymmetric.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/_cipheralgorithm.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/_serialization.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/cmac.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/constant_time.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/hashes.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/hmac.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/keywrap.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/padding.cpython-311.pyc,, +cryptography/hazmat/primitives/__pycache__/poly1305.cpython-311.pyc,, +cryptography/hazmat/primitives/_asymmetric.py,sha256=RhgcouUB6HTiFDBrR1LxqkMjpUxIiNvQ1r_zJjRG6qQ,532 +cryptography/hazmat/primitives/_cipheralgorithm.py,sha256=gKa0WrLz6K4fqhnGbfBYKDSxgLxsPU0uj_EK2UT47W4,1495 +cryptography/hazmat/primitives/_serialization.py,sha256=qrozc8fw2WZSbjk3DAlSl3ResxpauwJ74ZgGoUL-mj0,5142 +cryptography/hazmat/primitives/asymmetric/__init__.py,sha256=s9oKCQ2ycFdXoERdS1imafueSkBsL9kvbyfghaauZ9Y,180 +cryptography/hazmat/primitives/asymmetric/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/dh.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/dsa.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/ec.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/ed25519.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/ed448.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/padding.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/rsa.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/types.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/utils.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/x25519.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/__pycache__/x448.cpython-311.pyc,, +cryptography/hazmat/primitives/asymmetric/dh.py,sha256=OOCjMClH1Bf14Sy7jAdwzEeCxFPb8XUe2qePbExvXwc,3420 +cryptography/hazmat/primitives/asymmetric/dsa.py,sha256=xBwdf0pZOgvqjUKcO7Q0L3NxwalYj0SJDUqThemhSmI,3945 +cryptography/hazmat/primitives/asymmetric/ec.py,sha256=lwZmtAwi3PM8lsY1MsNaby_bVi--49OCxwE_1yqKC-A,10428 +cryptography/hazmat/primitives/asymmetric/ed25519.py,sha256=kl63fg7myuMjNTmMoVFeH6iVr0x5FkjNmggxIRTloJk,3423 +cryptography/hazmat/primitives/asymmetric/ed448.py,sha256=2UzEDzzfkPn83UFVFlMZfIMbAixxY09WmQyrwinWTn8,3456 +cryptography/hazmat/primitives/asymmetric/padding.py,sha256=eZcvUqVLbe3u48SunLdeniaPlV4-k6pwBl67OW4jSy8,2885 +cryptography/hazmat/primitives/asymmetric/rsa.py,sha256=nW_Ko7PID9UBJF10GVJOc_1L00ymFsfZDUJYtM5kfGQ,7637 +cryptography/hazmat/primitives/asymmetric/types.py,sha256=LnsOJym-wmPUJ7Knu_7bCNU3kIiELCd6krOaW_JU08I,2996 +cryptography/hazmat/primitives/asymmetric/utils.py,sha256=DPTs6T4F-UhwzFQTh-1fSEpQzazH2jf2xpIro3ItF4o,790 +cryptography/hazmat/primitives/asymmetric/x25519.py,sha256=VGYuRdIYuVBtizpFdNWd2bTrT10JRa1admQdBr08xz8,3341 +cryptography/hazmat/primitives/asymmetric/x448.py,sha256=GKKJBqYLr03VewMF18bXIM941aaWcZIQ4rC02GLLEmw,3374 +cryptography/hazmat/primitives/ciphers/__init__.py,sha256=eyEXmjk6_CZXaOPYDr7vAYGXr29QvzgWL2-4CSolLFs,680 +cryptography/hazmat/primitives/ciphers/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/primitives/ciphers/__pycache__/aead.cpython-311.pyc,, +cryptography/hazmat/primitives/ciphers/__pycache__/algorithms.cpython-311.pyc,, +cryptography/hazmat/primitives/ciphers/__pycache__/base.cpython-311.pyc,, +cryptography/hazmat/primitives/ciphers/__pycache__/modes.cpython-311.pyc,, +cryptography/hazmat/primitives/ciphers/aead.py,sha256=Fzlyx7w8KYQakzDp1zWgJnIr62zgZrgVh1u2h4exB54,634 +cryptography/hazmat/primitives/ciphers/algorithms.py,sha256=QvBMDmphRZfNmykij58L5eDkd_2NnCzIpJpyX2QwMxc,4223 +cryptography/hazmat/primitives/ciphers/base.py,sha256=tg-XNaKUyETBi7ounGDEL1_ICn-s4FF9LR7moV58blI,4211 +cryptography/hazmat/primitives/ciphers/modes.py,sha256=BFpxEGSaxoeZjrQ4sqpyPDvKClrqfDKIBv7kYtFURhE,8192 +cryptography/hazmat/primitives/cmac.py,sha256=sz_s6H_cYnOvx-VNWdIKhRhe3Ymp8z8J0D3CBqOX3gg,338 +cryptography/hazmat/primitives/constant_time.py,sha256=xdunWT0nf8OvKdcqUhhlFKayGp4_PgVJRU2W1wLSr_A,422 +cryptography/hazmat/primitives/hashes.py,sha256=EvDIJBhj83Z7f-oHbsA0TzZLFSDV_Yv8hQRdM4o8FD0,5091 +cryptography/hazmat/primitives/hmac.py,sha256=RpB3z9z5skirCQrm7zQbtnp9pLMnAjrlTUvKqF5aDDc,423 +cryptography/hazmat/primitives/kdf/__init__.py,sha256=4XibZnrYq4hh5xBjWiIXzaYW6FKx8hPbVaa_cB9zS64,750 +cryptography/hazmat/primitives/kdf/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/primitives/kdf/__pycache__/concatkdf.cpython-311.pyc,, +cryptography/hazmat/primitives/kdf/__pycache__/hkdf.cpython-311.pyc,, +cryptography/hazmat/primitives/kdf/__pycache__/kbkdf.cpython-311.pyc,, +cryptography/hazmat/primitives/kdf/__pycache__/pbkdf2.cpython-311.pyc,, +cryptography/hazmat/primitives/kdf/__pycache__/scrypt.cpython-311.pyc,, +cryptography/hazmat/primitives/kdf/__pycache__/x963kdf.cpython-311.pyc,, +cryptography/hazmat/primitives/kdf/concatkdf.py,sha256=bcn4NGXse-EsFl7nlU83e5ilop7TSHcX-CJJS107W80,3686 +cryptography/hazmat/primitives/kdf/hkdf.py,sha256=uhN5L87w4JvtAqQcPh_Ji2TPSc18IDThpaYJiHOWy3A,3015 +cryptography/hazmat/primitives/kdf/kbkdf.py,sha256=eSuLK1sATkamgCAit794jLr7sDNlu5X0USdcWhwJdmk,9146 +cryptography/hazmat/primitives/kdf/pbkdf2.py,sha256=Xj3YIeX30h2BUaoJAtOo1RMXV_em0-eCG0PU_0FHJzM,1950 +cryptography/hazmat/primitives/kdf/scrypt.py,sha256=4QONhjxA_ZtuQtQ7QV3FnbB8ftrFnM52B4HPfV7hFys,2354 +cryptography/hazmat/primitives/kdf/x963kdf.py,sha256=wCpWmwQjZ2vAu2rlk3R_PX0nINl8WGXYBmlyMOC5iPw,1992 +cryptography/hazmat/primitives/keywrap.py,sha256=XV4Pj2fqSeD-RqZVvY2cA3j5_7RwJSFygYuLfk2ujCo,5650 +cryptography/hazmat/primitives/padding.py,sha256=QUq0n-EAgEan9aQzuTsiJYGKbWiK1nSHkcYjDF1L1ok,5518 +cryptography/hazmat/primitives/poly1305.py,sha256=P5EPQV-RB_FJPahpg01u0Ts4S_PnAmsroxIGXbGeRRo,355 +cryptography/hazmat/primitives/serialization/__init__.py,sha256=jyNx_7NcOEbVRBY4nP9ks0IVXBafbcYnTK27vafPLW8,1653 +cryptography/hazmat/primitives/serialization/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/primitives/serialization/__pycache__/base.cpython-311.pyc,, +cryptography/hazmat/primitives/serialization/__pycache__/pkcs12.cpython-311.pyc,, +cryptography/hazmat/primitives/serialization/__pycache__/pkcs7.cpython-311.pyc,, +cryptography/hazmat/primitives/serialization/__pycache__/ssh.cpython-311.pyc,, +cryptography/hazmat/primitives/serialization/base.py,sha256=ikq5MJIwp_oUnjiaBco_PmQwOTYuGi-XkYUYHKy8Vo0,615 +cryptography/hazmat/primitives/serialization/pkcs12.py,sha256=7vVXbiP7qhhvKAHJT_M8-LBZdbpOwrpWRHWxNrNqzXE,4492 +cryptography/hazmat/primitives/serialization/pkcs7.py,sha256=CNzcsuDMyEFMe3EUii4NfJlQzmakB2hLlfRFYObnHRs,11141 +cryptography/hazmat/primitives/serialization/ssh.py,sha256=VKscMrVdYK5B9PQISjjdRMglRvqa_L3sDNm5vdjVHJY,51915 +cryptography/hazmat/primitives/twofactor/__init__.py,sha256=tmMZGB-g4IU1r7lIFqASU019zr0uPp_wEBYcwdDCKCA,258 +cryptography/hazmat/primitives/twofactor/__pycache__/__init__.cpython-311.pyc,, +cryptography/hazmat/primitives/twofactor/__pycache__/hotp.cpython-311.pyc,, +cryptography/hazmat/primitives/twofactor/__pycache__/totp.cpython-311.pyc,, +cryptography/hazmat/primitives/twofactor/hotp.py,sha256=l1YdRMIhfPIuHKkA66keBDHhNbnBAlh6-O44P-OHIK8,2976 +cryptography/hazmat/primitives/twofactor/totp.py,sha256=v0y0xKwtYrP83ypOo5Ofd441RJLOkaFfjmp554jo5F0,1450 +cryptography/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +cryptography/utils.py,sha256=Rp7ppg4XIBVVzNQ6XngGndwkICJoYp6FoFOOgTWLJ7g,3925 +cryptography/x509/__init__.py,sha256=uGdiViR7KFnWGoJFVUStt-e_ufomWc87RQBGAZ7dT-4,7980 +cryptography/x509/__pycache__/__init__.cpython-311.pyc,, +cryptography/x509/__pycache__/base.cpython-311.pyc,, +cryptography/x509/__pycache__/certificate_transparency.cpython-311.pyc,, +cryptography/x509/__pycache__/extensions.cpython-311.pyc,, +cryptography/x509/__pycache__/general_name.cpython-311.pyc,, +cryptography/x509/__pycache__/name.cpython-311.pyc,, +cryptography/x509/__pycache__/ocsp.cpython-311.pyc,, +cryptography/x509/__pycache__/oid.cpython-311.pyc,, +cryptography/x509/__pycache__/verification.cpython-311.pyc,, +cryptography/x509/base.py,sha256=3NbbUn9wPruhmoPO7Cl3trc3SrqV2OFIBBE0P2l05mg,37081 +cryptography/x509/certificate_transparency.py,sha256=6HvzAD0dlSQVxy6tnDhGj0-pisp1MaJ9bxQNRr92inI,2261 +cryptography/x509/extensions.py,sha256=R70KkJ_c5NQ6Kx7Rho0sGJ0Rh-bOuBHjVOFSQGRAFCs,67370 +cryptography/x509/general_name.py,sha256=sP_rV11Qlpsk4x3XXGJY_Mv0Q_s9dtjeLckHsjpLQoQ,7836 +cryptography/x509/name.py,sha256=MYCxCSTQTpzhjxFPZaANqJ9fGrhESH73vPkoay8HSWM,14830 +cryptography/x509/ocsp.py,sha256=P6A02msz5pe-IkUFpvxezHvnEHGvPdXiD3S0wsuf4-I,20003 +cryptography/x509/oid.py,sha256=X8EbhkRTLrGuv9vHZSGqPd9zpvRVsonU_joWAL5LLY8,885 +cryptography/x509/verification.py,sha256=alfx3VaTSb2bMz7_7s788oL90vzgHwBjVINssdz0Gv0,796 diff --git a/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/WHEEL b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/WHEEL new file mode 100644 index 0000000000000000000000000000000000000000..b3e268a945016b375135efee3ff578960543bbee --- /dev/null +++ b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/WHEEL @@ -0,0 +1,5 @@ +Wheel-Version: 1.0 +Generator: maturin (1.7.0) +Root-Is-Purelib: false +Tag: cp39-abi3-manylinux_2_28_x86_64 + diff --git a/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE new file mode 100644 index 0000000000000000000000000000000000000000..b11f379efe1504d235b4d2d42685ba5dc6af6e9f --- /dev/null +++ b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE @@ -0,0 +1,3 @@ +This software is made available under the terms of *either* of the licenses +found in LICENSE.APACHE or LICENSE.BSD. Contributions to cryptography are made +under the terms of *both* these licenses. diff --git a/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE.APACHE b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE.APACHE new file mode 100644 index 0000000000000000000000000000000000000000..62589edd12a37dd28b6b6fed1e2d728ac9f05c8d --- /dev/null +++ b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE.APACHE @@ -0,0 +1,202 @@ + + Apache License + Version 2.0, January 2004 + https://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + https://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE.BSD b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE.BSD new file mode 100644 index 0000000000000000000000000000000000000000..ec1a29d34d6e419411c75523408aca72f705345c --- /dev/null +++ b/.venv/lib/python3.11/site-packages/cryptography-43.0.3.dist-info/license_files/LICENSE.BSD @@ -0,0 +1,27 @@ +Copyright (c) Individual contributors. +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + 1. Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + 3. Neither the name of PyCA Cryptography nor the names of its contributors + may be used to endorse or promote products derived from this software + without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND +ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED +WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR +ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES +(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; +LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON +ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/INSTALLER b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/INSTALLER new file mode 100644 index 0000000000000000000000000000000000000000..a1b589e38a32041e49332e5e81c2d363dc418d68 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/INSTALLER @@ -0,0 +1 @@ +pip diff --git a/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/METADATA b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/METADATA new file mode 100644 index 0000000000000000000000000000000000000000..d9536b6bf5399723e870d5381cf2e9f1fe14486a --- /dev/null +++ b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/METADATA @@ -0,0 +1,176 @@ +Metadata-Version: 2.3 +Name: jsonschema +Version: 4.23.0 +Summary: An implementation of JSON Schema validation for Python +Project-URL: Homepage, https://github.com/python-jsonschema/jsonschema +Project-URL: Documentation, https://python-jsonschema.readthedocs.io/ +Project-URL: Issues, https://github.com/python-jsonschema/jsonschema/issues/ +Project-URL: Funding, https://github.com/sponsors/Julian +Project-URL: Tidelift, https://tidelift.com/subscription/pkg/pypi-jsonschema?utm_source=pypi-jsonschema&utm_medium=referral&utm_campaign=pypi-link +Project-URL: Changelog, https://github.com/python-jsonschema/jsonschema/blob/main/CHANGELOG.rst +Project-URL: Source, https://github.com/python-jsonschema/jsonschema +Author-email: Julian Berman +License: MIT +License-File: COPYING +Keywords: data validation,json,json schema,jsonschema,validation +Classifier: Development Status :: 5 - Production/Stable +Classifier: Intended Audience :: Developers +Classifier: License :: OSI Approved :: MIT License +Classifier: Operating System :: OS Independent +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Programming Language :: Python :: 3.12 +Classifier: Programming Language :: Python :: 3.13 +Classifier: Programming Language :: Python :: Implementation :: CPython +Classifier: Programming Language :: Python :: Implementation :: PyPy +Classifier: Topic :: File Formats :: JSON +Classifier: Topic :: File Formats :: JSON :: JSON Schema +Requires-Python: >=3.8 +Requires-Dist: attrs>=22.2.0 +Requires-Dist: importlib-resources>=1.4.0; python_version < '3.9' +Requires-Dist: jsonschema-specifications>=2023.03.6 +Requires-Dist: pkgutil-resolve-name>=1.3.10; python_version < '3.9' +Requires-Dist: referencing>=0.28.4 +Requires-Dist: rpds-py>=0.7.1 +Provides-Extra: format +Requires-Dist: fqdn; extra == 'format' +Requires-Dist: idna; extra == 'format' +Requires-Dist: isoduration; extra == 'format' +Requires-Dist: jsonpointer>1.13; extra == 'format' +Requires-Dist: rfc3339-validator; extra == 'format' +Requires-Dist: rfc3987; extra == 'format' +Requires-Dist: uri-template; extra == 'format' +Requires-Dist: webcolors>=1.11; extra == 'format' +Provides-Extra: format-nongpl +Requires-Dist: fqdn; extra == 'format-nongpl' +Requires-Dist: idna; extra == 'format-nongpl' +Requires-Dist: isoduration; extra == 'format-nongpl' +Requires-Dist: jsonpointer>1.13; extra == 'format-nongpl' +Requires-Dist: rfc3339-validator; extra == 'format-nongpl' +Requires-Dist: rfc3986-validator>0.1.0; extra == 'format-nongpl' +Requires-Dist: uri-template; extra == 'format-nongpl' +Requires-Dist: webcolors>=24.6.0; extra == 'format-nongpl' +Description-Content-Type: text/x-rst + +========== +jsonschema +========== + +|PyPI| |Pythons| |CI| |ReadTheDocs| |Precommit| |Zenodo| + +.. |PyPI| image:: https://img.shields.io/pypi/v/jsonschema.svg + :alt: PyPI version + :target: https://pypi.org/project/jsonschema/ + +.. |Pythons| image:: https://img.shields.io/pypi/pyversions/jsonschema.svg + :alt: Supported Python versions + :target: https://pypi.org/project/jsonschema/ + +.. |CI| image:: https://github.com/python-jsonschema/jsonschema/workflows/CI/badge.svg + :alt: Build status + :target: https://github.com/python-jsonschema/jsonschema/actions?query=workflow%3ACI + +.. |ReadTheDocs| image:: https://readthedocs.org/projects/python-jsonschema/badge/?version=stable&style=flat + :alt: ReadTheDocs status + :target: https://python-jsonschema.readthedocs.io/en/stable/ + +.. |Precommit| image:: https://results.pre-commit.ci/badge/github/python-jsonschema/jsonschema/main.svg + :alt: pre-commit.ci status + :target: https://results.pre-commit.ci/latest/github/python-jsonschema/jsonschema/main + +.. |Zenodo| image:: https://zenodo.org/badge/3072629.svg + :alt: Zenodo DOI + :target: https://zenodo.org/badge/latestdoi/3072629 + + +``jsonschema`` is an implementation of the `JSON Schema `_ specification for Python. + +.. code:: python + + >>> from jsonschema import validate + + >>> # A sample schema, like what we'd get from json.load() + >>> schema = { + ... "type" : "object", + ... "properties" : { + ... "price" : {"type" : "number"}, + ... "name" : {"type" : "string"}, + ... }, + ... } + + >>> # If no exception is raised by validate(), the instance is valid. + >>> validate(instance={"name" : "Eggs", "price" : 34.99}, schema=schema) + + >>> validate( + ... instance={"name" : "Eggs", "price" : "Invalid"}, schema=schema, + ... ) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + ValidationError: 'Invalid' is not of type 'number' + +It can also be used from the command line by installing `check-jsonschema `_. + +Features +-------- + +* Full support for `Draft 2020-12 `_, `Draft 2019-09 `_, `Draft 7 `_, `Draft 6 `_, `Draft 4 `_ and `Draft 3 `_ + +* `Lazy validation `_ that can iteratively report *all* validation errors. + +* `Programmatic querying `_ of which properties or items failed validation. + + +Installation +------------ + +``jsonschema`` is available on `PyPI `_. You can install using `pip `_: + +.. code:: bash + + $ pip install jsonschema + + +Extras +====== + +Two extras are available when installing the package, both currently related to ``format`` validation: + + * ``format`` + * ``format-nongpl`` + +They can be used when installing in order to include additional dependencies, e.g.: + +.. code:: bash + + $ pip install jsonschema'[format]' + +Be aware that the mere presence of these dependencies – or even the specification of ``format`` checks in a schema – do *not* activate format checks (as per the specification). +Please read the `format validation documentation `_ for further details. + +About +----- + +I'm Julian Berman. + +``jsonschema`` is on `GitHub `_. + +Get in touch, via GitHub or otherwise, if you've got something to contribute, it'd be most welcome! + +You can also generally find me on Libera (nick: ``Julian``) in various channels, including ``#python``. + +If you feel overwhelmingly grateful, you can also `sponsor me `_. + +And for companies who appreciate ``jsonschema`` and its continued support and growth, ``jsonschema`` is also now supportable via `TideLift `_. + + +Release Information +------------------- + +v4.23.0 +======= + +* Do not reorder dictionaries (schemas, instances) that are printed as part of validation errors. +* Declare support for Py3.13 diff --git a/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/RECORD b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/RECORD new file mode 100644 index 0000000000000000000000000000000000000000..1cd1baaead3227bbc4bcf57d8cdf76e253ed58a3 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/RECORD @@ -0,0 +1,76 @@ +../../../bin/jsonschema,sha256=XHo009U0gdoLuMe818hXZdVGUS_4zdLM86D9zqPvDvA,231 +jsonschema-4.23.0.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4 +jsonschema-4.23.0.dist-info/METADATA,sha256=Hd96gAfdO0v5RpFeT25qjyo7PvhASy56F4Jw3FUUTlo,7906 +jsonschema-4.23.0.dist-info/RECORD,, +jsonschema-4.23.0.dist-info/WHEEL,sha256=1yFddiXMmvYK7QYTqtRNtX66WJ0Mz8PYEiEUoOUUxRY,87 +jsonschema-4.23.0.dist-info/entry_points.txt,sha256=vO7rX4Fs_xIVJy2pnAtKgTSxfpnozAVQ0DjCmpMxnWE,51 +jsonschema-4.23.0.dist-info/licenses/COPYING,sha256=T5KgFaE8TRoEC-8BiqE0MLTxvHO0Gxa7hGw0Z2bedDk,1057 +jsonschema/__init__.py,sha256=LkPwscySlJ9lTOp7ZB1M7jQ8mbG7-bYG41iBwbZ-o9s,3941 +jsonschema/__main__.py,sha256=iLsZf2upUB3ilBKTlMnyK-HHt2Cnnfkwwxi_c6gLvSA,115 +jsonschema/__pycache__/__init__.cpython-311.pyc,, +jsonschema/__pycache__/__main__.cpython-311.pyc,, +jsonschema/__pycache__/_format.cpython-311.pyc,, +jsonschema/__pycache__/_keywords.cpython-311.pyc,, +jsonschema/__pycache__/_legacy_keywords.cpython-311.pyc,, +jsonschema/__pycache__/_types.cpython-311.pyc,, +jsonschema/__pycache__/_typing.cpython-311.pyc,, +jsonschema/__pycache__/_utils.cpython-311.pyc,, +jsonschema/__pycache__/cli.cpython-311.pyc,, +jsonschema/__pycache__/exceptions.cpython-311.pyc,, +jsonschema/__pycache__/protocols.cpython-311.pyc,, +jsonschema/__pycache__/validators.cpython-311.pyc,, +jsonschema/_format.py,sha256=F_MA52IkrhOIxDqD8x-01bH37mG5nh0kyNrWUSLtWb8,14591 +jsonschema/_keywords.py,sha256=r8_DrqAfn6QLwQnmXEggveiSU-UaIL2p2nuPINelfFc,14949 +jsonschema/_legacy_keywords.py,sha256=2tWuwRPWbYS7EAl8wBIC_rabGuv1J4dfYLqNEPpShhA,15191 +jsonschema/_types.py,sha256=HQ5QD_oL85zF1FSW2v-5rvfYF0967HJdxSR88kzw2mY,5367 +jsonschema/_typing.py,sha256=NZhPhkBOn9INYZk8G69rDeuRamztgXCMLh10z9cfT6g,610 +jsonschema/_utils.py,sha256=ODga3vrJ6K2wMGxerpgn4ipc9q7ZSqBsvwKU4embLEE,10642 +jsonschema/benchmarks/__init__.py,sha256=A0sQrxDBVHSyQ-8ru3L11hMXf3q9gVuB9x_YgHb4R9M,70 +jsonschema/benchmarks/__pycache__/__init__.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/const_vs_enum.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/contains.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/issue232.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/json_schema_test_suite.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/nested_schemas.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/subcomponents.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/unused_registry.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/useless_applicator_schemas.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/useless_keywords.cpython-311.pyc,, +jsonschema/benchmarks/__pycache__/validator_creation.cpython-311.pyc,, +jsonschema/benchmarks/const_vs_enum.py,sha256=DVFi3WDqBalZFOibnjpX1uTSr3Rxa2cPgFcowd7Ukrs,830 +jsonschema/benchmarks/contains.py,sha256=gexQoUrCOwECofbt19BeosQZ7WFL6PDdkX49DWwBlOg,786 +jsonschema/benchmarks/issue232.py,sha256=3LLYLIlBGQnVuyyo2iAv-xky5P6PRFHANx4-zIIQOoE,521 +jsonschema/benchmarks/issue232/issue.json,sha256=eaPOZjMRu5u8RpKrsA9uk7ucPZS5tkKG4D_hkOTQ3Hk,117105 +jsonschema/benchmarks/json_schema_test_suite.py,sha256=PvfabpUYcF4_7csYDTcTauED8rnFEGYbdY5RqTXD08s,320 +jsonschema/benchmarks/nested_schemas.py,sha256=mo07dx-CIgmSOI62CNs4g5xu1FzHklLBpkQoDxWYcKs,1892 +jsonschema/benchmarks/subcomponents.py,sha256=fEyiMzsWeK2pd7DEGCuuY-vzGunwhHczRBWEnBRLKIo,1113 +jsonschema/benchmarks/unused_registry.py,sha256=hwRwONc9cefPtYzkoX_TYRO3GyUojriv0-YQaK3vnj0,940 +jsonschema/benchmarks/useless_applicator_schemas.py,sha256=EVm5-EtOEFoLP_Vt2j4SrCwlx05NhPqNuZQ6LIMP1Dc,3342 +jsonschema/benchmarks/useless_keywords.py,sha256=bj_zKr1oVctFlqyZaObCsYTgFjiiNgPzC0hr1Y868mE,867 +jsonschema/benchmarks/validator_creation.py,sha256=UkUQlLAnussnr_KdCIdad6xx2pXxQLmYtsXoiirKeWQ,285 +jsonschema/cli.py,sha256=SGy9JPg02mgXhNxugU8iXhYNivfSjBhKTNAgV90ty-M,8551 +jsonschema/exceptions.py,sha256=RxE2T5xxgg_B6ttR8a3lCbZyh29RUtFe4oZKMoHPBAE,15035 +jsonschema/protocols.py,sha256=7mpZxO1gfRNMCGXwldwsSN3nEugVfIVyKZ_HZgN1vSw,7174 +jsonschema/tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +jsonschema/tests/__pycache__/__init__.cpython-311.pyc,, +jsonschema/tests/__pycache__/_suite.cpython-311.pyc,, +jsonschema/tests/__pycache__/fuzz_validate.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_cli.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_deprecations.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_exceptions.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_format.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_jsonschema_test_suite.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_types.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_utils.cpython-311.pyc,, +jsonschema/tests/__pycache__/test_validators.cpython-311.pyc,, +jsonschema/tests/_suite.py,sha256=QAfBj34zMbJQ5_JJ2ogpiTlw9hQ6Is43dvo_bpS0EdM,8156 +jsonschema/tests/fuzz_validate.py,sha256=fUA7yTJIihaCwJplkUehZeyB84HcXEcqtY5oPJXIO7I,1114 +jsonschema/tests/test_cli.py,sha256=uFMu2YbIfbSDCnykhLL4-VR3-jg1tvQLJn2Bliwp_Bw,28587 +jsonschema/tests/test_deprecations.py,sha256=9VxOCfWzMG1Tg4OD8riU_Znd6HDOQZkepzVgxsdUdU8,15760 +jsonschema/tests/test_exceptions.py,sha256=JgC-E1ZFZK2puVBp35WFRnG8CNOiSWLYtyLjh9IvFKI,22591 +jsonschema/tests/test_format.py,sha256=eVm5SMaWF2lOPO28bPAwNvkiQvHCQKy-MnuAgEchfEc,3188 +jsonschema/tests/test_jsonschema_test_suite.py,sha256=a2saPs2Cwwg0sdRdu-uJ8goSXLbqrS-pC48QJy0K4DE,8674 +jsonschema/tests/test_types.py,sha256=cF51KTDmdsx06MrIc4fXKt0X9fIsVgw5uhT8CamVa8U,6977 +jsonschema/tests/test_utils.py,sha256=sao74o1PyYMxBfqweokQN48CFSS6yhJk5FkCfMJ5PsI,4163 +jsonschema/tests/test_validators.py,sha256=eiaigsZMzHYYsniQ1UPygaS56a1d-_7-9NC4wVXAhzs,87975 +jsonschema/validators.py,sha256=H31FwHdyB7LP5eunxdBrZ9E57hpvozfnRlZaOYy45jU,47045 diff --git a/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/WHEEL b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/WHEEL new file mode 100644 index 0000000000000000000000000000000000000000..cdd68a497cdfa8d3f2b837225beacef711b85047 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/WHEEL @@ -0,0 +1,4 @@ +Wheel-Version: 1.0 +Generator: hatchling 1.25.0 +Root-Is-Purelib: true +Tag: py3-none-any diff --git a/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/entry_points.txt b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/entry_points.txt new file mode 100644 index 0000000000000000000000000000000000000000..eecef9d8fad351de6b048532776160696a796ef9 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/entry_points.txt @@ -0,0 +1,2 @@ +[console_scripts] +jsonschema = jsonschema.cli:main diff --git a/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/licenses/COPYING b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/licenses/COPYING new file mode 100644 index 0000000000000000000000000000000000000000..af9cfbdb134f42e5205ecbad597421d778826481 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/jsonschema-4.23.0.dist-info/licenses/COPYING @@ -0,0 +1,19 @@ +Copyright (c) 2013 Julian Berman + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in +all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN +THE SOFTWARE. diff --git a/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/__init__.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..79f8aacff4a37fef07345e46225d506fb40b0db2 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/__init__.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/base.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/base.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..2140867f054a5509465b71b61b63f1bdd53e475e Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/base.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/exceptions.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/exceptions.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..cf9a52d8b77852b209d00a5314753c56340a7c20 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/exceptions.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/multimodal.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/multimodal.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..855c0f74827e2388ae46e72842edc5990bdeff69 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/__pycache__/multimodal.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/data/mistral_instruct_tokenizer_240323.model.v3 b/.venv/lib/python3.11/site-packages/mistral_common/data/mistral_instruct_tokenizer_240323.model.v3 new file mode 100644 index 0000000000000000000000000000000000000000..ed1cacdb9dfea8a9369109388816f04656e1720e --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/data/mistral_instruct_tokenizer_240323.model.v3 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9addc8bdce5988448ae81b729336f43a81262160ae8da760674badab9d4c7d33 +size 587591 diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__init__.py b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/__init__.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..4729ea8ce22148458675d7778755cb665ec456d7 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/__init__.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/request.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/request.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..e9f97ed243e8d99e8d20ddb2619d5dfc2188cbc1 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/request.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/response.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/response.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..205e81ee2021be15fe0ce3bc25e49e1dd3f55522 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/__pycache__/response.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/request.py b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/request.py new file mode 100644 index 0000000000000000000000000000000000000000..d965495a181a4c6041fc043381a45c9adcc2d5bb --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/request.py @@ -0,0 +1,11 @@ +from typing import List, Optional, Union + +from pydantic import Field + +from mistral_common.base import MistralBase + + +class EmbeddingRequest(MistralBase): + input: Union[str, List[str]] = Field(description="Text to embed.") + model: str = Field(description="ID of the model to use.") + encoding_format: Optional[str] = Field(default="float", description="The format to return the embeddings in.") diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/response.py b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/response.py new file mode 100644 index 0000000000000000000000000000000000000000..d9ac56361d2d9146f2db14f344f2cd94c08a4350 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/protocol/embedding/response.py @@ -0,0 +1,21 @@ +from typing import List + +from pydantic import Field + +from mistral_common.base import MistralBase +from mistral_common.protocol.base import UsageInfo +from mistral_common.protocol.utils import random_uuid + + +class EmbeddingObject(MistralBase): + object: str = Field(default="embedding", description="The type of the object returned.") + embedding: List[float] = Field(description="The type of the object returned.") + index: int = Field(description="The index of the embedding in the input text.") + + +class EmbeddingResponse(MistralBase): + id: str = Field(default_factory=lambda: f"embd-{random_uuid()}") + object: str = Field(default="list", description="The type of the object returned.") + data: List[EmbeddingObject] = Field(description="List of embeddings.") + model: str = Field(description="The model used to generate the embeddings.") + usage: UsageInfo diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/__init__.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..30448a275179bb986f63ae1177f27a50ffc09d2b Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/__init__.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/messages.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/messages.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..0df65274b323b915b44cf32365dbe2a38addc67d Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/messages.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/response.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/response.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..f3ccbc63291ea8cdf6d84eff891c55f3201478c4 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/protocol/instruct/__pycache__/response.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/__init__.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/__pycache__/__init__.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..235f13880b3c5c62817288c93c91bd0c76f4c861 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/__pycache__/__init__.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__init__.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__pycache__/__init__.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..8314e75032b130912fd767213983acc6886b5740 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__pycache__/__init__.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__pycache__/request.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__pycache__/request.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..d7ad368f9519832308ec53c559f9ed30d9e33f90 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/__pycache__/request.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/request.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/request.py new file mode 100644 index 0000000000000000000000000000000000000000..95ebdb345d36b8982c7d5acf0c020546f782fdf1 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/tokens/instruct/request.py @@ -0,0 +1,25 @@ +from typing import Generic, List, Optional + +from mistral_common.base import MistralBase +from mistral_common.protocol.instruct.messages import ChatMessageType +from mistral_common.protocol.instruct.tool_calls import ToolType + + +class FIMRequest(MistralBase): + """ + A valid Fill in the Middle completion request to be tokenized + """ + + prompt: str + suffix: Optional[str] = None + + +class InstructRequest(MistralBase, Generic[ChatMessageType, ToolType]): + """ + A valid request to be tokenized + """ + + messages: List[ChatMessageType] + system_prompt: Optional[str] = None + available_tools: Optional[List[ToolType]] = None + truncate_at_max_tokens: Optional[int] = None diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__init__.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/__init__.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..531479c5c1f588b03af29e88247c32b2c3a2ecc8 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/__init__.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/base.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/base.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..0a6c67db008e65430b0f75997b50016273e10785 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/base.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/mistral.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/mistral.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..5316aeeffdc914213739213a6fa542d15205dcdc Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/mistral.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/multimodal.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/multimodal.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..b738ed600565591d95bd72d71202d42f3c8bad01 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/multimodal.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/sentencepiece.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/sentencepiece.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..6416402d5028493a6d7ab424b6fe27b3f5bb5e44 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/sentencepiece.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/tekken.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/tekken.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..e90f551a1adfc26496fa8c9837a0f59721606d07 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/tekken.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/utils.cpython-311.pyc b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/utils.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..142120cc009c4dea690a15d61fa9ad06c6fa139c Binary files /dev/null and b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/__pycache__/utils.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/base.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/base.py new file mode 100644 index 0000000000000000000000000000000000000000..63b60406f5c30b34320137fdfd57b7f4a020a797 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/base.py @@ -0,0 +1,200 @@ +from abc import ABC, abstractmethod +from dataclasses import dataclass +from enum import Enum +from typing import Generic, List, Optional, Protocol, Tuple, TypeVar, Union + +import numpy as np +from pydantic import ConfigDict + +from mistral_common.base import MistralBase +from mistral_common.protocol.instruct.messages import ( + AssistantMessageType, + ContentChunk, + ImageChunk, + ImageURLChunk, + UserMessage, +) +from mistral_common.protocol.instruct.tool_calls import Tool +from mistral_common.tokens.instruct.request import FIMRequest, InstructRequest + + +class SpecialTokens(str, Enum): + bos = "" + eos = "" + begin_inst = "[INST]" + end_inst = "[/INST]" + begin_tools = "[AVAILABLE_TOOLS]" + end_tools = "[/AVAILABLE_TOOLS]" + begin_tool_results = "[TOOL_RESULTS]" + end_tool_results = "[/TOOL_RESULTS]" + tool_calls = "[TOOL_CALLS]" + img = "[IMG]" + img_break = "[IMG_BREAK]" + img_end = "[IMG_END]" + prefix = "[PREFIX]" + middle = "[MIDDLE]" + suffix = "[SUFFIX]" + begin_system = "[SYSTEM_PROMPT]" + end_system = "[/SYSTEM_PROMPT]" + begin_tool_content = "[TOOL_CONTENT]" + + +class TokenizerVersion(str, Enum): + v1 = "v1" # vocab_size = 32000 + v2 = "v2" # vocab_size = 32768 with special control tokens [INST], [\INST] + v3 = "v3" # vocab_size = 32768 (spm) OR 128000 (tekken) with improved function calling + v7 = "v7" # vocab_size = 32768 (spm) or 128000 (tekken) with improved system prompt and function calling + + +class Tokenized(MistralBase): + """ + A tokenized InstructRequest + """ + + model_config = ConfigDict(arbitrary_types_allowed=True) + tokens: List[int] + text: Optional[str] = None + prefix_ids: Optional[List[int]] = None + images: List[np.ndarray] = [] + + +class Tokenizer(ABC): + @property + @abstractmethod + def n_words(self) -> int: + """Vocabulary size""" + + @abstractmethod + def vocab(self) -> List[str]: + """All tokens in the vocabulary as strings""" + + @abstractmethod + def id_to_piece(self, token_id: int) -> str: + """Convert a token id to the token str""" + + @property + @abstractmethod + def bos_id(self) -> int: + """id of the Beginning of String token""" + + @property + @abstractmethod + def eos_id(self) -> int: + """id of the End of String token""" + + @property + @abstractmethod + def pad_id(self) -> int: + """id of the Pad token""" + + @property + @abstractmethod + def unk_id(self) -> int: + """id of the Unk token""" + + @abstractmethod + def encode(self, s: str, bos: bool, eos: bool) -> List[int]: + """String to token ids""" + + @abstractmethod + def decode(self, t: List[int]) -> str: + """Token ids to string""" + + @abstractmethod + def get_control_token(self, s: str) -> int: + """Get the id of a control token""" + + @property + @abstractmethod + def version(self) -> TokenizerVersion: + """Get the version of the tokenizer""" + + @abstractmethod + def to_string(self, tokens: List[int]) -> str: + """Convert token ids to string""" + + +InstructRequestType = TypeVar("InstructRequestType", bound=InstructRequest) +FIMRequestType = TypeVar("FIMRequestType", bound=FIMRequest) +TokenizedType = TypeVar("TokenizedType", bound=Tokenized) + + +@dataclass +class ImageEncoding: + tokens: List[int] + image: np.ndarray + + +@dataclass +class SpecialImageIDs: + img: int + img_break: int + img_end: int + + @staticmethod + def from_tokenizer(tokenizer: "Tokenizer") -> "SpecialImageIDs": + return SpecialImageIDs( + img=tokenizer.get_control_token(SpecialTokens.img.value), + img_break=tokenizer.get_control_token(SpecialTokens.img_break.value), + img_end=tokenizer.get_control_token(SpecialTokens.img_end.value), + ) + + +class MultiModalEncoder(Protocol): + def __call__(self, content: Union[ImageChunk, ImageURLChunk]) -> ImageEncoding: + """ + Encode the given content. + + Args: + content (ChunkContent): The content to be encoded. + + Returns: + ImageEncoding: The encoded image content. + """ + ... + + @property + def image_token(self) -> int: + ... + + +class InstructTokenizer(Generic[InstructRequestType, FIMRequestType, TokenizedType, AssistantMessageType]): + tokenizer: Tokenizer + mm_encoder: Optional[MultiModalEncoder] + + def __init__(self, tokenizer: Tokenizer, mm_encoder: Optional[MultiModalEncoder]) -> None: + """Init from tokenizer""" + + @abstractmethod + def encode_instruct(self, request: InstructRequestType) -> TokenizedType: + """Instruct request to Tokenized object""" + + @abstractmethod + def decode(self, tokens: List[int]) -> str: + """Convert token ids to string""" + + @abstractmethod + def encode_fim(self, request: FIMRequestType) -> TokenizedType: + """FIM request to Tokenized object""" + + @abstractmethod + def encode_user_message( + self, + message: UserMessage, + available_tools: Optional[List[Tool]], + is_last: bool, + is_first: bool, + system_prompt: Optional[str] = None, + force_img_first: bool = False, + ) -> Tuple[List[int], List[np.ndarray]]: + ... + + @abstractmethod + def encode_user_content( + self, + content: Union[str, List[ContentChunk]], + is_last: bool, + system_prompt: Optional[str] = None, + force_img_first: bool = False, + ) -> Tuple[List[int], List[np.ndarray]]: + ... diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/mistral.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/mistral.py new file mode 100644 index 0000000000000000000000000000000000000000..c54141232f5334085e00044b99def9ccd4980c22 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/mistral.py @@ -0,0 +1,251 @@ +import warnings +from pathlib import Path +from typing import Callable, Dict, Generic, List, Optional, Union + +from mistral_common.exceptions import ( + TokenizerException, +) +from mistral_common.protocol.instruct.messages import ( + UATS, + AssistantMessageType, + SystemMessageType, + ToolMessageType, + UserMessageType, +) +from mistral_common.protocol.instruct.normalize import InstructRequestNormalizer, normalizer_for_tokenizer_version +from mistral_common.protocol.instruct.request import ChatCompletionRequest +from mistral_common.protocol.instruct.validator import ( + MistralRequestValidator, + MistralRequestValidatorV3, + ValidationMode, +) +from mistral_common.tokens.instruct.request import FIMRequest +from mistral_common.tokens.tokenizers.base import ( + InstructRequest, + InstructRequestType, + InstructTokenizer, + SpecialTokens, + TokenizedType, + TokenizerVersion, +) +from mistral_common.tokens.tokenizers.multimodal import ( + ImageEncoder, + MultimodalConfig, + MultiModalEncoder, + SpecialImageIDs, +) +from mistral_common.tokens.tokenizers.sentencepiece import ( + InstructTokenizerV1, + InstructTokenizerV2, + InstructTokenizerV3, + InstructTokenizerV7, + SentencePieceTokenizer, + get_mm_config, + is_sentencepiece, +) +from mistral_common.tokens.tokenizers.tekken import Tekkenizer, is_tekken + + +def load_mm_encoder( + mm_config: MultimodalConfig, tokenizer: Union[Tekkenizer, SentencePieceTokenizer] +) -> MultiModalEncoder: + special_ids = SpecialImageIDs( + img=tokenizer.get_control_token(SpecialTokens.img.value), + img_break=tokenizer.get_control_token(SpecialTokens.img_break.value), + img_end=tokenizer.get_control_token(SpecialTokens.img_end.value), + ) + return ImageEncoder(mm_config, special_ids) + + +class MistralTokenizer( + Generic[UserMessageType, AssistantMessageType, ToolMessageType, SystemMessageType, TokenizedType] +): + def __init__( + self, + instruct_tokenizer: InstructTokenizer[InstructRequest, FIMRequest, TokenizedType, AssistantMessageType], + validator: MistralRequestValidator[UserMessageType, AssistantMessageType, ToolMessageType, SystemMessageType], + request_normalizer: InstructRequestNormalizer[ + UserMessageType, AssistantMessageType, ToolMessageType, SystemMessageType, InstructRequestType + ], + ): + self._chat_completion_request_validator = validator + self._instruct_request_normalizer = request_normalizer + self.instruct_tokenizer = instruct_tokenizer + + @classmethod + def _data_path(cls) -> Path: + return Path(__file__).parents[2] / "data" + + @classmethod + def v1(cls) -> "MistralTokenizer": + """open 7B x 8x7B + embed""" + return cls.from_file(str(cls._data_path() / "tokenizer.model.v1"), mode=ValidationMode.test) + + @classmethod + def v2(cls) -> "MistralTokenizer": + """mistral-small // mistral-large""" + return cls.from_file( + str(cls._data_path() / "mistral_instruct_tokenizer_240216.model.v2"), mode=ValidationMode.test + ) + + @classmethod + def v3(cls, is_tekken: bool = False, is_mm: bool = False) -> "MistralTokenizer": + """open-mixtral-8x22B""" + if is_tekken and is_mm: + tokenizer_name = "tekken_240911.json" + elif is_tekken and not is_mm: + tokenizer_name = "tekken_240718.json" + elif not is_tekken and is_mm: + raise ValueError("Multimodal tokenizer is currently only supported for tekken") + else: + tokenizer_name = "mistral_instruct_tokenizer_240323.model.v3" + + return cls.from_file(str(cls._data_path() / tokenizer_name), mode=ValidationMode.test) + + @classmethod + def v7(cls, is_mm: bool = False) -> "MistralTokenizer": + """mistral-large 2.1""" + if is_mm: + return cls.from_file( + str(cls._data_path() / "mistral_instruct_tokenizer_241114.model.v7m1"), mode=ValidationMode.test + ) + else: + return cls.from_file( + str(cls._data_path() / "mistral_instruct_tokenizer_241114.model.v7"), mode=ValidationMode.test + ) + + @classmethod + def from_model(cls, model: str, strict: bool = False) -> "MistralTokenizer": + model_name_to_tokenizer_cls: Dict[str, Callable[[], MistralTokenizer]] = { + "ministral-8b-2410": lambda: MistralTokenizer.v3(is_tekken=True), + "mistral-tiny-2312": MistralTokenizer.v2, + "open-mistral-nemo-2407": lambda: MistralTokenizer.v3(is_tekken=True), + "mistral-tiny-2407": MistralTokenizer.v3, + "mistral-small-2312": MistralTokenizer.v2, + "open-mixtral-8x22b-2404": MistralTokenizer.v3, + "mistral-small-2402": MistralTokenizer.v2, + "mistral-small-2409": lambda: MistralTokenizer.v3(is_tekken=True), + "mistral-medium-2312": MistralTokenizer.v1, + "mistral-large-2402": MistralTokenizer.v2, + "mistral-large-2407": MistralTokenizer.v3, + "mistral-large-2411": MistralTokenizer.v7, + "pixtral-large-2411": lambda: MistralTokenizer.v7(is_mm=True), + "codestral-2405": MistralTokenizer.v3, + "codestral-mamba-2407": MistralTokenizer.v3, + "pixtral-12b-2409": lambda: MistralTokenizer.v3(is_tekken=True, is_mm=True), + # The following are deprecated - only left for backward comp. Delete in >= 1.6.0 + "open-mistral-7b": MistralTokenizer.v1, + "open-mixtral-8x7b": MistralTokenizer.v1, + "mistral-embed": MistralTokenizer.v1, + "mistral-small-v1": MistralTokenizer.v2, + "mistral-large-v1": MistralTokenizer.v2, + "mistral-small": MistralTokenizer.v3, + "mistral-large": MistralTokenizer.v3, + "open-mixtral-8x22b": MistralTokenizer.v3, + "codestral-22b": MistralTokenizer.v3, + "mistral-nemo": lambda: MistralTokenizer.v3(is_tekken=True), + "pixtral": lambda: MistralTokenizer.v3(is_tekken=True, is_mm=True), + "pixtral-large": lambda: MistralTokenizer.v7(is_mm=True), + } + + if not strict: + warnings.warn( + "Calling `MistralTokenizer.from_model(..., strict=False)` is deprecated as it can lead to incorrect " + "tokenizers. It is strongly recommended to use MistralTokenizer.from_model(..., strict=True)` " + "which will become the default in `mistral_common=1.6.0`." + "If you are using `mistral_common` for open-sourced model weights, we recommend using " + "`MistralTokenizer.from_file('')` instead.", + FutureWarning, + ) + + # TODO(Delete this code in mistral_common >= 1.6.0 + # Prefix search the model name mapping + for model_name, tokenizer_cls in model_name_to_tokenizer_cls.items(): + if model_name in model.lower(): + return tokenizer_cls() + + if model not in model_name_to_tokenizer_cls: + raise TokenizerException(f"Unrecognized model: {model}") + + return model_name_to_tokenizer_cls[model]() + + @classmethod + def from_file( + cls, + tokenizer_filename: str, + mode: ValidationMode = ValidationMode.test, + ) -> "MistralTokenizer": + """ + Depending on which model we are loading, tokenization and validation might be different. 💩 + """ + tokenizer: Union[SentencePieceTokenizer, Tekkenizer] + + if is_tekken(tokenizer_filename): + tokenizer = Tekkenizer.from_file(tokenizer_filename) + mm_config = tokenizer.multimodal + elif is_sentencepiece(tokenizer_filename): + tokenizer = SentencePieceTokenizer(tokenizer_filename) + mm_config = get_mm_config(tokenizer_filename) + else: + raise TokenizerException(f"Unrecognized tokenizer file: {tokenizer_filename}") + + mm_encoder = load_mm_encoder(mm_config, tokenizer) if mm_config is not None else None + + request_normalizer = normalizer_for_tokenizer_version(tokenizer.version) + + if tokenizer.version == TokenizerVersion.v1: + assert mm_encoder is None, "Tokenizer version needs to be >= v3" + return MistralTokenizer( + InstructTokenizerV1(tokenizer), + validator=MistralRequestValidator(mode=mode), + request_normalizer=request_normalizer, + ) + elif tokenizer.version == TokenizerVersion.v2: + assert mm_encoder is None, "Tokenizer version needs to be >= v3" + return MistralTokenizer( + InstructTokenizerV2(tokenizer), + validator=MistralRequestValidator(mode=mode), + request_normalizer=request_normalizer, + ) + elif tokenizer.version == TokenizerVersion.v3: + return MistralTokenizer( + InstructTokenizerV3(tokenizer, mm_encoder=mm_encoder), + validator=MistralRequestValidatorV3(mode=mode), + request_normalizer=request_normalizer, + ) + elif tokenizer.version == TokenizerVersion.v7: + return MistralTokenizer( + InstructTokenizerV7(tokenizer, mm_encoder=mm_encoder), + validator=MistralRequestValidatorV3(mode=mode), + request_normalizer=request_normalizer, + ) + else: + raise TokenizerException(f"Unrecognized tokenizer filename: {tokenizer_filename}") + + raise TokenizerException(f"Unrecognized tokenizer version: {tokenizer.version}") + + def encode_chat_completion( + self, request: ChatCompletionRequest[UATS], max_model_input_len: Optional[int] = None + ) -> TokenizedType: + validated_request = self._chat_completion_request_validator.validate_request(request) + + if max_model_input_len is None and request.truncate_for_context_length: + # the max_model_input_len arg should not be optionnal ; + # but this function is used in many small scripts that have no use + # for truncation, and don't provide the max model len + raise TokenizerException( + "encoding a chat completion request with truncation, but no max model len was provided", + ) + + instruct_request = self._instruct_request_normalizer.from_chat_completion_request(validated_request) + + if request.truncate_for_context_length: + instruct_request.truncate_at_max_tokens = max_model_input_len + + return self.instruct_tokenizer.encode_instruct(instruct_request) + + def encode_fim(self, request: FIMRequest) -> TokenizedType: + return self.instruct_tokenizer.encode_fim(request) + + def decode(self, tokens: List[int]) -> str: + return self.instruct_tokenizer.decode(tokens) diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/multimodal.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/multimodal.py new file mode 100644 index 0000000000000000000000000000000000000000..c7542daff83a6c6f4a21b7375cddccfb077d9e94 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/multimodal.py @@ -0,0 +1,172 @@ +import base64 +import logging +from dataclasses import dataclass +from enum import Enum +from io import BytesIO +from typing import Tuple, Union + +import numpy as np +from PIL import Image + +from mistral_common.multimodal import SerializableImage, download_image +from mistral_common.protocol.instruct.messages import ImageChunk, ImageURLChunk +from mistral_common.tokens.tokenizers.base import ( + ImageEncoding, + MultiModalEncoder, + SpecialImageIDs, +) + +logger = logging.getLogger(__name__) + + +_cv2_installed: bool +try: + import cv2 + + _cv2_installed = True +except ImportError: + _cv2_installed = False +except Exception as e: + # cv2 has lots of import problems: https://github.com/opencv/opencv-python/issues/884 + # for better UX, let's simply skip all errors that might arise from import for now + logger.warning( + f"Warning: Your installation of OpenCV appears to be broken: {e}." + "Please follow the instructions at https://github.com/opencv/opencv-python/issues/884 " + "to correct your environment. The import of cv2 has been skipped." + ) + + +def is_cv2_installed() -> bool: + return _cv2_installed + + +def image_from_chunk(chunk: Union[ImageURLChunk, ImageChunk]) -> SerializableImage: + """Get a serializable image from a chunk.""" + if isinstance(chunk, ImageChunk): + return chunk.image + if chunk.get_url().startswith("data:image"): + data = chunk.get_url().split(",")[1] + image_data = base64.b64decode(data) + return Image.open(BytesIO(image_data)) + if chunk.get_url().startswith("http"): + return download_image(chunk.get_url()) + + raise RuntimeError(f"Unsupported image url scheme {chunk.get_url()}") + + +DATASET_MEAN = (0.48145466, 0.4578275, 0.40821073) # RGB +DATASET_STD = (0.26862954, 0.26130258, 0.27577711) # RGB + + +# only relevant for spm +class MultiModalVersion(str, Enum): + m1 = "m1" + + @property + def config(self) -> "MultimodalConfig": + if self.name == "m1": + return MultimodalConfig(16, 1024) + + raise NotImplementedError(f"{self.name}") + + +@dataclass +class MultimodalConfig: + image_patch_size: int + max_image_size: int + + +def _convert_to_rgb(image: Image.Image) -> Image.Image: + """ + Convert a PIL image to RGB. + We ensure transparent background becomes white. + """ + if image.mode == "RGB": + return image + if image.mode != "RGBA": + image = image.convert("RGBA") + white_bg: Image.Image = Image.new("RGBA", image.size, "WHITE") + white_bg.paste(image, (0, 0), image) + return white_bg.convert("RGB") + + +def normalize( + np_image: np.ndarray, + mean: Tuple[float, float, float], + std: Tuple[float, float, float], +) -> np.ndarray: + """ + Normalize a tensor image with mean and standard deviation. + + Args: + image (np.ndarray): Image to be normalized. + mean (tuple[float, float, float]): Mean for each channel. + std (tuple[float, float, float]): Standard deviation for each channel. + + Returns: + np.ndarray: Normalized image with shape (C, H, W). + """ + np_image = np_image / 255.0 + + assert len(np_image.shape) == 3, f"{np_image.shape=}" + assert np_image.shape[2] == len(mean) == len(std), f"{np_image.shape=}, {mean=}, {std=}" + + np_image = (np_image - mean) / std + + return np_image.transpose(2, 0, 1) + + +def transform_image(image: Image.Image, new_size: Tuple[int, int]) -> np.ndarray: + if not is_cv2_installed(): + raise ImportError("OpenCV is required for this function. Install it with 'pip install mistral_common[opencv]'") + + np_image = cv2.resize(np.array(_convert_to_rgb(image), dtype=np.float32), new_size, interpolation=cv2.INTER_CUBIC) + return normalize(np_image, DATASET_MEAN, DATASET_STD) + + +class ImageEncoder(MultiModalEncoder): + def __init__(self, mm_config: MultimodalConfig, special_ids: SpecialImageIDs) -> None: + self.mm_config = mm_config + self.special_ids = special_ids + + def _image_to_num_tokens(self, img: Image.Image) -> Tuple[int, int]: + w: Union[int, float] + h: Union[int, float] + + w, h = img.size + ratio = max(h / self.mm_config.max_image_size, w / self.mm_config.max_image_size) + if ratio > 1: + w = round(w / ratio) + h = round(h / ratio) + + width_tokens = (w - 1) // self.mm_config.image_patch_size + 1 + height_tokens = (h - 1) // self.mm_config.image_patch_size + 1 + + return width_tokens, height_tokens + + def __call__(self, content: Union[ImageChunk, ImageURLChunk]) -> ImageEncoding: + """ + Converts ImageChunks to numpy image arrays and image token ids + + Args: + image (ImageChunk, ImageURLChunk): ImageChunk to be converted + + Returns: + ImageEncoding containing image token ids and processed image in numpy format + """ + image = image_from_chunk(content) + w, h = self._image_to_num_tokens(image) + assert w > 0 + assert h > 0 + image_tokens = ([self.special_ids.img] * w + [self.special_ids.img_break]) * h + image_tokens[-1] = self.special_ids.img_end + new_image_size = ( + w * self.mm_config.image_patch_size, + h * self.mm_config.image_patch_size, + ) + processed_image = transform_image(image, new_image_size) + return ImageEncoding(tokens=image_tokens, image=processed_image) + + @property + def image_token(self) -> int: + return self.special_ids.img diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/sentencepiece.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/sentencepiece.py new file mode 100644 index 0000000000000000000000000000000000000000..6b92411778adf55750444365df8f912b0df92a9f --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/sentencepiece.py @@ -0,0 +1,672 @@ +import json +import logging +import os +from abc import abstractmethod +from functools import cached_property +from pathlib import Path +from typing import Any, Dict, Generic, List, Optional, Set, Tuple, Union + +import numpy as np +from sentencepiece import SentencePieceProcessor + +from mistral_common.exceptions import TokenizerException +from mistral_common.protocol.instruct.messages import ( + AssistantMessage, + AssistantMessageType, + ContentChunk, + SystemMessage, + TextChunk, + ToolMessage, + UserMessage, +) +from mistral_common.protocol.instruct.tool_calls import Tool, ToolCall +from mistral_common.tokens.instruct.request import FIMRequest, InstructRequest +from mistral_common.tokens.tokenizers.base import ( + FIMRequestType, + InstructRequestType, + InstructTokenizer, + SpecialTokens, + Tokenized, + TokenizedType, + Tokenizer, + TokenizerVersion, +) +from mistral_common.tokens.tokenizers.multimodal import MultimodalConfig, MultiModalEncoder, MultiModalVersion + + +def is_sentencepiece(path: Union[str, Path]) -> bool: + if isinstance(path, str): + path = Path(path) + + instruct_versions = list(TokenizerVersion.__members__) + mm_versions = list(MultiModalVersion.__members__) + [""] # allow no mm version + suffixes = [f".model.{v}{m}" for v in instruct_versions for m in mm_versions] + [".model"] + + return path.is_file() and any(path.name.endswith(suffix) for suffix in suffixes) + + +def get_spm_version(tokenizer_filename: str, raise_deprecated: bool = False) -> TokenizerVersion: + _version_str = tokenizer_filename.split(".")[-1].split("m")[0] + if _version_str == "model": + if raise_deprecated: + raise TokenizerException(f"Make sure to rename your tokenizer file to end with {tokenizer_filename}.v1.") + + # tokenizer.model => tokenizer.model.v1 + return TokenizerVersion("v1") + + if _version_str not in TokenizerVersion.__members__: + raise TokenizerException(f"Unrecognized tokenizer filename: {tokenizer_filename}") + + return TokenizerVersion(_version_str) + + +def get_mm_config(tokenizer_filename: str) -> Optional[MultimodalConfig]: + _version_str = tokenizer_filename.split(".")[-1] + if "m" not in _version_str: + return None + + _mm_version_str = "m" + _version_str.split("m")[-1] + + if _mm_version_str not in MultiModalVersion.__members__: + raise TokenizerException(f"Unrecognized tokenizer filename: {tokenizer_filename}") + + return MultiModalVersion(_mm_version_str).config + + +class SentencePieceTokenizer(Tokenizer): + def __init__(self, model_path: str, tokenizer_version: Optional[TokenizerVersion] = None) -> None: + self._logger = logging.getLogger(self.__class__.__name__) + # reload tokenizer + assert os.path.isfile(model_path), model_path + self._model = SentencePieceProcessor(model_file=model_path) + + assert self._model.vocab_size() == self._model.get_piece_size() + self._vocab = [self._model.id_to_piece(i) for i in range(self.n_words)] + + self._version: TokenizerVersion = tokenizer_version or get_spm_version(model_path, raise_deprecated=False) + + super().__init__() + + @property + def version(self) -> TokenizerVersion: + return self._version + + def get_control_token(self, s: str) -> int: + return self._model.piece_to_id(s) # type: ignore + + @property + def n_words(self) -> int: + return self._model.vocab_size() # type: ignore + + def vocab(self) -> List[str]: + return self._vocab + + @property + def bos_id(self) -> int: + return self._model.bos_id() # type: ignore + + @property + def eos_id(self) -> int: + return self._model.eos_id() # type: ignore + + @cached_property + def _control_tokens(self) -> Set[int]: + return {tok for tok in range(self.n_words) if self._model.IsControl(tok)} + + def encode(self, s: str, bos: bool, eos: bool) -> List[int]: + assert isinstance(s, str) + t: List[int] = self._model.encode(s) + if bos: + t = [self.bos_id, *t] + if eos: + t = [*t, self.eos_id] + return t + + def decode(self, t: List[int]) -> str: + return self._model.decode(t) # type: ignore + + def id_to_piece(self, token_id: int) -> str: + return self._model.id_to_piece(token_id) # type: ignore + + def to_string(self, tokens: List[int]) -> str: + """ + Converts tokens into a string for debugging purposes + """ + text = "" + curr_tokens: List[int] = [] + for tok in tokens: + if tok in self._control_tokens: + if curr_tokens: + text += "".join([self.id_to_piece(tok) for tok in curr_tokens]) + curr_tokens = [] + + text += self.id_to_piece(tok) + + else: + curr_tokens.append(tok) + + if curr_tokens: + text += "".join([self.id_to_piece(tok) for tok in curr_tokens]) + + return text + + @property + def pad_id(self) -> int: + return self._model.pad_id() # type: ignore + + @property + def unk_id(self) -> int: + return self._model.unk_id() # type: ignore + + +class InstructTokenizerBase( + InstructTokenizer, Generic[InstructRequestType, FIMRequestType, TokenizedType, AssistantMessageType] +): + def __init__(self, tokenizer: Tokenizer, mm_encoder: Optional[MultiModalEncoder] = None): + self.tokenizer = tokenizer + self.mm_encoder = mm_encoder + super().__init__(tokenizer, mm_encoder) + + def start(self) -> List[int]: + return [self.tokenizer.bos_id] + + @staticmethod + def find_first_last_user(request: InstructRequest) -> Tuple[int, int]: + # find last user message + last_user_idx = -1 + first_user_idx = -1 + for i, msg in list(enumerate(request.messages)): + if isinstance(msg, UserMessage): + if first_user_idx == -1: + first_user_idx = i + last_user_idx = i + return first_user_idx, last_user_idx + + @abstractmethod + def encode_tool_message(self, message: ToolMessage, is_before_last_user_message: bool) -> List[int]: + raise NotImplementedError("Tool message not implemented") + + @abstractmethod + def encode_assistant_message(self, message: AssistantMessageType, is_before_last_user_message: bool) -> List[int]: + raise NotImplementedError("Assistant message not implemented") + + def _truncate_for_max_tokens( + self, + tokenized: List[Optional[List[int]]], + messages: List[AssistantMessageType], + max_tokens: int, + last_user_message_index: int, + ) -> None: + # Tokenizer ⩽ V3 does not support truncation + return + + def encode_instruct( + self, + request: InstructRequest[AssistantMessageType, Tool], + ) -> Tokenized: + # init at bos + images: List[np.ndarray] = [] + prefix_ids: Optional[List[int]] = None + tokens_list: List[Optional[List[int]]] = [] + + # find last user message + first_user_idx, last_user_idx = self.find_first_last_user(request) + for msg_idx, msg in enumerate(request.messages): + if isinstance(msg, UserMessage): + new_tokens, new_images = self.encode_user_message( + msg, + request.available_tools, + msg_idx == last_user_idx, + msg_idx == first_user_idx, + system_prompt=request.system_prompt, + force_img_first=True, # img is always first when providing text/img chunk pair + ) + images.extend(new_images) + elif isinstance(msg, ToolMessage): + new_tokens = self.encode_tool_message(msg, msg_idx < last_user_idx) + elif isinstance(msg, AssistantMessage): + new_tokens = self.encode_assistant_message(msg, msg_idx < last_user_idx) + if msg_idx == len(request.messages) - 1: + prefix_ids = new_tokens + elif isinstance(msg, SystemMessage): + new_tokens = self.encode_system_message(msg) + + tokens_list.append(new_tokens) + + if request.truncate_at_max_tokens is not None: + self._truncate_for_max_tokens( + tokens_list, + request.messages, + request.truncate_at_max_tokens, + last_user_idx, + ) + tokens = self.start() + + for tok in tokens_list: + if tok is not None: + tokens.extend(tok) + + return Tokenized( + tokens=tokens, + text=self.tokenizer.to_string(tokens), + prefix_ids=prefix_ids, + images=images, + ) + + def decode(self, tokens: List[int]) -> str: + return self.tokenizer.decode(tokens) + + +class InstructTokenizerV1( + InstructTokenizerBase, Generic[InstructRequestType, FIMRequestType, TokenizedType, AssistantMessageType] +): + def encode_user_message( + self, + message: UserMessage, + available_tools: Optional[List[Tool]], + is_last: bool, + is_first: bool, + system_prompt: Optional[str] = None, + force_img_first: bool = False, + ) -> Tuple[List[int], List[np.ndarray]]: + assert message.content is not None + assert isinstance(message.content, str), "Message content must be normalized" + assert self.mm_encoder is None, "InstructTokenizerV1 cannot encode images" + + content = "" + if is_first and system_prompt: + content = system_prompt + "\n\n" + message.content + else: + content = message.content + + message_txt = f"[INST] {content} [/INST]" + curr_tokens, image_tokens = self.encode_user_content(content=message_txt, is_last=False, system_prompt=None) + return curr_tokens, image_tokens + + def encode_user_content( + self, + content: Union[str, List[ContentChunk]], + is_last: bool, + system_prompt: Optional[str] = None, + force_img_first: bool = False, + ) -> Tuple[List[int], List[np.ndarray]]: + assert isinstance(content, str) + + if is_last and system_prompt: + content = system_prompt + "\n\n" + content + + tokens = self.tokenizer.encode(content, bos=False, eos=False) + return tokens, [] + + def encode_tool_message(self, message: ToolMessage, is_before_last_user_message: bool) -> List[int]: + raise TokenizerException("Tools not implemented for tokenizer V1") + + def encode_assistant_message(self, message: AssistantMessageType, is_before_last_user_message: bool) -> List[int]: + assert isinstance(message, AssistantMessage), message + if message.tool_calls is not None and len(message.tool_calls) > 0: + raise TokenizerException("Tools not implemented for tokenizer V1") + elif message.content: + curr_tokens = self.tokenizer.encode(message.content, bos=False, eos=False) + else: + raise TokenizerException(f"{message.content} // {message.tool_calls}") + if not message.prefix: + curr_tokens.append(self.tokenizer.eos_id) + return curr_tokens + + def encode_fim(self, request: FIMRequest) -> Tokenized: + raise TokenizerException("FIM not available for tokenizer V1") + + +class InstructTokenizerV2( + InstructTokenizerV1, Generic[InstructRequestType, FIMRequestType, TokenizedType, AssistantMessageType] +): + def __init__(self, tokenizer: Tokenizer, mm_encoder: Optional[MultiModalEncoder] = None): + super().__init__(tokenizer, mm_encoder) + self.BEGIN_INST = self.tokenizer.get_control_token(SpecialTokens.begin_inst.value) + self.END_INST = self.tokenizer.get_control_token(SpecialTokens.end_inst.value) + self.BEGIN_AVAILABLE_TOOLS = self.tokenizer.get_control_token(SpecialTokens.begin_tools.value) + self.END_AVAILABLE_TOOLS = self.tokenizer.get_control_token(SpecialTokens.end_tools.value) + self.BEGIN_TOOL_RESULTS = self.tokenizer.get_control_token(SpecialTokens.begin_tool_results.value) + self.END_TOOL_RESULTS = self.tokenizer.get_control_token(SpecialTokens.end_tool_results.value) + self.TOOL_CALLS = self.tokenizer.get_control_token(SpecialTokens.tool_calls.value) + self.BOS = self.tokenizer.get_control_token(SpecialTokens.bos.value) + self.PREFIX = self.tokenizer.get_control_token(SpecialTokens.prefix.value) + self.SUFFIX = self.tokenizer.get_control_token(SpecialTokens.suffix.value) + + def encode_user_message( + self, + message: UserMessage, + available_tools: Optional[List[Tool]], + is_last: bool, + is_first: bool, + system_prompt: Optional[str] = None, + force_img_first: bool = False, + ) -> Tuple[List[int], List[np.ndarray]]: + assert message.content is not None + tools_tokens: List[int] = [] + if is_last and available_tools: + tools = [tool.model_dump() for tool in available_tools] + tools_json_tokens = self.tokenizer.encode(json.dumps(tools, ensure_ascii=False), bos=False, eos=False) + tools_tokens = [ + self.BEGIN_AVAILABLE_TOOLS, + *tools_json_tokens, + self.END_AVAILABLE_TOOLS, + ] + + tokens, image_tokens = self.encode_user_content( + content=message.content, + is_last=is_last, + system_prompt=system_prompt, + force_img_first=force_img_first, + ) + + prefix_tokens = [*tools_tokens, self.BEGIN_INST] + suffix_tokens = [self.END_INST] + + curr_tokens = prefix_tokens + tokens + suffix_tokens + + return curr_tokens, image_tokens + + def _parse_json_content(self, content: str) -> Any: + try: + return json.loads(content) + except json.JSONDecodeError: + return content + + def _prepare_tool_result(self, tool_message: ToolMessage) -> Dict[str, Any]: + """ + Bit of a hack due to the way tool results are tokenized + """ + assert tool_message.content is not None, "Tool message content cannot be None" + return { + "name": tool_message.name, + "content": self._parse_json_content(tool_message.content), + } + + def encode_tool_message(self, message: ToolMessage, is_before_last_user_message: bool) -> List[int]: + if is_before_last_user_message: + # don't tokenize last tool response before last user msg + return [] + + # Currently only supports single tool results + tool_result_str = json.dumps([self._prepare_tool_result(message)], ensure_ascii=False) + curr_tokens = [ + self.BEGIN_TOOL_RESULTS, + *self.tokenizer.encode(tool_result_str, bos=False, eos=False), + self.END_TOOL_RESULTS, + ] + return curr_tokens + + def _prepare_function_call(self, tool_call: ToolCall) -> Dict[str, Any]: + """ + Bit of a hack due to the way function calls are tokenized + """ + return { + "name": tool_call.function.name, + "arguments": self._parse_json_content(tool_call.function.arguments), + } + + def _encode_normal_content_assistant_message(self, message: AssistantMessageType) -> List[int]: + assert message.content, f"Assistant message must have content. Got {message}" + return self.tokenizer.encode(message.content.rstrip(" "), bos=False, eos=False) + + def _encode_tool_calls_in_assistant_message(self, message: AssistantMessageType) -> List[int]: + assert message.tool_calls, f"Assistant message must have tool calls. Got {message}" + prepared_tool_calls = [] + for tool_call in message.tool_calls: + prepared_tool_calls.append(self._prepare_function_call(tool_call)) + tool_call_str = json.dumps(prepared_tool_calls, ensure_ascii=False) + curr_tokens = [ + self.TOOL_CALLS, + *self.tokenizer.encode(tool_call_str, bos=False, eos=False), + ] + return curr_tokens + + def encode_assistant_message(self, message: AssistantMessageType, is_before_last_user_message: bool) -> List[int]: + if message.tool_calls: + if is_before_last_user_message: + # don't tokenize tool call before last user message + return [] + curr_tokens = self._encode_tool_calls_in_assistant_message(message) + elif message.content: + curr_tokens = self._encode_normal_content_assistant_message(message) + else: + raise TokenizerException(f"Invalid assistant message: {message.content}") + if not message.prefix: + curr_tokens.append(self.tokenizer.eos_id) + return curr_tokens + + def _encode_infilling(self, text: str) -> List[int]: + """ + Remove prefix space in the case of SentencePieceTokenizers + Thanks Fabian ! + """ + + return self.tokenizer.encode("☺" + text, bos=False, eos=False)[2:] + + def encode_fim(self, request: FIMRequest) -> Tokenized: + prefix_tokens = self.tokenizer.encode(request.prompt, bos=False, eos=False) + suffix_tokens = self._encode_infilling(request.suffix) if request.suffix else [] + tokens = [ + self.BOS, + self.SUFFIX, + *suffix_tokens, + self.PREFIX, + *prefix_tokens, + ] + return Tokenized(tokens=tokens, text=self.tokenizer.to_string(tokens)) + + +class InstructTokenizerV3( + InstructTokenizerV2, Generic[InstructRequestType, FIMRequestType, TokenizedType, AssistantMessageType] +): + """ + The only difference with V3 tokenizer is that it encodes the tool messages differently + """ + + def __init__(self, tokenizer: Tokenizer, mm_encoder: Optional[MultiModalEncoder] = None) -> None: + super().__init__(tokenizer, mm_encoder=mm_encoder) + + def _prepare_function_call(self, tool_call: ToolCall) -> Dict[str, Any]: + function_call = { + "name": tool_call.function.name, + "arguments": self._parse_json_content(tool_call.function.arguments), + } + + if tool_call.id and tool_call.id != "null": + function_call["id"] = tool_call.id + + return function_call + + def _prepare_tool_result(self, tool_message: ToolMessage) -> Dict[str, Any]: + assert tool_message.content is not None, "Tool message content cannot be None" + assert tool_message.tool_call_id is not None, "Tool message has to have the tool call id defined in v3" + + return { + "content": self._parse_json_content(tool_message.content), + "call_id": tool_message.tool_call_id, + } + + def encode_tool_message(self, message: ToolMessage, is_before_last_user_message: bool) -> List[int]: + """ + Same as V2 but tools not wrapped in a list and history is tokenized also + """ + tool_result_str = json.dumps(self._prepare_tool_result(message), ensure_ascii=False) + curr_tokens = [ + self.BEGIN_TOOL_RESULTS, + *self.tokenizer.encode(tool_result_str, bos=False, eos=False), + self.END_TOOL_RESULTS, + ] + return curr_tokens + + def encode_assistant_message(self, message: AssistantMessageType, is_before_last_user_message: bool) -> List[int]: + """ + Same as V2 but always encode tool history + """ + return super().encode_assistant_message(message, False) + + def encode_user_content( + self, + content: Union[str, List[ContentChunk]], + is_last: bool, + system_prompt: Optional[str] = None, + force_img_first: bool = False, + ) -> Tuple[List[int], List[np.ndarray]]: + if isinstance(content, str): + return super().encode_user_content(content, is_last, system_prompt) + + tokens: List[int] = [] + images: List[np.ndarray] = [] + + has_one_img_one_text_first = ( + len(content) == 2 and isinstance(content[0], TextChunk) and not isinstance(content[1], TextChunk) + ) + if force_img_first and has_one_img_one_text_first: + # make sure that if exactly one image and text chunk are passed we force the image chunk to be first + content = [content[1], content[0]] + + first_chunk = True + for chunk in content: + content = "" + if first_chunk and is_last and system_prompt: + first_chunk = False + content = system_prompt + "\n\n" + if isinstance(chunk, TextChunk): + content += chunk.text + tokens.extend(self.tokenizer.encode(content, bos=False, eos=False)) + else: + assert self.mm_encoder is not None, "Make sure to define a multi-modal encoder at init" + if content: + tokens.extend(self.tokenizer.encode(content, bos=False, eos=False)) + + img_encoding = self.mm_encoder(chunk) + + tokens.extend(img_encoding.tokens) + images.append(img_encoding.image) + + return tokens, images + + +class InstructTokenizerV7(InstructTokenizerV3): + """ + The difference with V3 tokenizer is that it encodes the system prompts differently: + - in V7 the system prompts are treated as separate SystemMessages + - they are no longer prepended to the last user message + - they are printed between special tokens + Tool call results are encoded as : + - [begin tool call] call_id_tokens [tool_content] content tokens [end tool call] + """ + + def __init__(self, tokenizer: Tokenizer, mm_encoder: Optional[MultiModalEncoder] = None) -> None: + super().__init__(tokenizer, mm_encoder) + self.BEGIN_SYSTEM = self.tokenizer.get_control_token(SpecialTokens.begin_system.value) + self.END_SYSTEM = self.tokenizer.get_control_token(SpecialTokens.end_system.value) + self.BEGIN_TOOL_CONTENT = self.tokenizer.get_control_token(SpecialTokens.begin_tool_content.value) + + def _truncate_for_max_tokens( + self, + tokenized_messages: List[Optional[List[int]]], + messages: List[AssistantMessageType], + max_tokens: int, + last_user_message_index: int, + ) -> None: + # drop some messages to fit in max_tokens. Rules: + # - don't drop any system messages + # - when a user message is dropped, all following assistant|tool message should be dropped until the next + # user message + # - we never drop the last message + to_drop = sum(len(t) for t in tokenized_messages if t is not None) - max_tokens + + def drop(idx: int) -> None: + nonlocal to_drop + if isinstance(messages[idx], SystemMessage): + # never drop system messages + return + if idx == last_user_message_index: + # never drop the last user message + return + tok = tokenized_messages[idx] + assert tok is not None + to_drop -= len(tok) + tokenized_messages[idx] = None + + current_idx = 0 + while to_drop > 0 and current_idx < len(messages): + drop(current_idx) + current_idx += 1 + if isinstance(messages[current_idx - 1], UserMessage): + # if we just dropped a UserMessage, + # also drop everything until the next user message + while current_idx < len(messages) and not isinstance(messages[current_idx], UserMessage): + drop(current_idx) + current_idx += 1 + + if to_drop > 0: + raise TokenizerException("Input couldn't fit in truncate_at_max_token") + + def encode_system_message(self, message: SystemMessage) -> List[int]: + assert message.content is not None + assert isinstance(message.content, str), "Message content must be normalized" + tokens = [ + self.BEGIN_SYSTEM, + *self.tokenizer.encode(message.content, bos=False, eos=False), + self.END_SYSTEM, + ] + return tokens + + def encode_user_message( + self, + message: UserMessage, + available_tools: Optional[List[Tool]], + is_last: bool, + is_first: bool, + system_prompt: Optional[str] = None, + force_img_first: bool = False, + ) -> Tuple[List[int], List[np.ndarray]]: + assert system_prompt is None, "in Tokenizer V7 we don't encode system prompts in user messages" + return super().encode_user_message( + message, + available_tools, + is_last=is_last, + is_first=is_first, + system_prompt=None, + force_img_first=force_img_first, + ) + + def encode_tool_message(self, message: ToolMessage, is_before_last_user_message: bool) -> List[int]: + """ + Same as V3 but tools not wrapped in a list and history is tokenized also + """ + assert message.tool_call_id is not None + tool_call_id_tokens = self.tokenizer.encode(message.tool_call_id, bos=False, eos=False) + tokens = self.tokenizer.encode(message.content, bos=False, eos=False) + + prefix_tokens = [ + self.BEGIN_TOOL_RESULTS, + *tool_call_id_tokens, + self.BEGIN_TOOL_CONTENT, + ] + curr_tokens = [ + *prefix_tokens, + *tokens, + self.END_TOOL_RESULTS, + ] + return curr_tokens + + def encode_assistant_message(self, message: AssistantMessageType, is_before_last_user_message: bool) -> List[int]: + if not message.content and not message.tool_calls: + raise TokenizerException(f"Invalid assistant message: {message}") + curr_tokens: list = [] + if message.content: + if isinstance(message.content, str): + curr_tokens += self._encode_normal_content_assistant_message(message) + elif isinstance(message.content, list): + curr_tokens += self.encode_content_chunks( + message.content, is_last=False, system_prompt=None, force_img_first=True + ).tokens + if message.tool_calls: + curr_tokens += self._encode_tool_calls_in_assistant_message(message) + if not message.prefix: + curr_tokens.append(self.tokenizer.eos_id) + + return curr_tokens diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/tekken.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/tekken.py new file mode 100644 index 0000000000000000000000000000000000000000..da6859c164ca3ca844fdb2e87d9a9171aeb0d807 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/tekken.py @@ -0,0 +1,312 @@ +import base64 +import json +import logging +from enum import Enum +from functools import cached_property +from itertools import groupby +from pathlib import Path +from typing import Dict, List, Optional, Type, TypedDict, Union + +import tiktoken + +from mistral_common.tokens.tokenizers.base import ( + SpecialTokens, + Tokenizer, + TokenizerVersion, +) +from mistral_common.tokens.tokenizers.multimodal import MultimodalConfig + +logger = logging.getLogger(__name__) + + +def is_tekken(path: Union[str, Path]) -> bool: + if isinstance(path, str): + path = Path(path) + return path.is_file() and "tekken" in path.name and path.suffix == ".json" + + +# Formatting specification of the JSON file +class TokenInfo(TypedDict): + rank: int + token_bytes: str # base64 encoded + token_str: Optional[str] + + +class TekkenConfig(TypedDict): + pattern: str + num_vocab_tokens: int + default_vocab_size: int + default_num_special_tokens: int + version: str + + +class ModelData(TypedDict): + vocab: List[TokenInfo] + config: TekkenConfig + version: int + type: str + multimodal: MultimodalConfig + + +class SpecialTokenPolicy(Enum): + """What to do with special tokens when encoding/decoding.""" + + IGNORE = 0 + KEEP = 1 + RAISE = 2 + + +class Tekkenizer(Tokenizer): + SPECIAL_TOKENS = ( + "", + SpecialTokens.bos, + SpecialTokens.eos, + SpecialTokens.begin_inst, + SpecialTokens.end_inst, + SpecialTokens.begin_tools, + SpecialTokens.end_tools, + SpecialTokens.begin_tool_results, + SpecialTokens.end_tool_results, + SpecialTokens.tool_calls, + SpecialTokens.img, + "", + SpecialTokens.img_break, + SpecialTokens.img_end, + SpecialTokens.prefix, + SpecialTokens.middle, + SpecialTokens.suffix, + SpecialTokens.begin_system, + SpecialTokens.end_system, + SpecialTokens.begin_tool_content, + ) + SPECIAL_TOKEN_TEMPLATE = "" + + # # note that params has a vocab_size field, but it's not used + + def __init__( + self, + vocab: List[TokenInfo], + pattern: str, + vocab_size: int, + num_special_tokens: int, + version: TokenizerVersion, + *, + name: str = "tekkenizer", + _path: Optional[str] = None, + mm_config: Optional[MultimodalConfig] = None, + ): + assert vocab_size <= len(vocab) + num_special_tokens, ( + vocab_size, + len(vocab), + num_special_tokens, + ) + self._vocab_size = vocab_size + self._path = _path + + special_tokens = list(self.SPECIAL_TOKENS) + assert len(special_tokens) == len(set(special_tokens)), f"Special tokens must be unique: {special_tokens}" + assert len(special_tokens) < num_special_tokens + + special_filler = [ + self.SPECIAL_TOKEN_TEMPLATE.format(id=i) for i in range(len(special_tokens), num_special_tokens) + ] + if special_filler: + logger.info(f"Adding special tokens {special_filler[0]}, ..., {special_filler[-1]}") + special_tokens = special_tokens + special_filler + assert len(set(special_tokens)) == len(special_tokens) == num_special_tokens, special_tokens + inner_vocab_size = vocab_size - num_special_tokens + + # reload vocab + self._tekken_token2id_nospecial = _reload_mergeable_ranks(vocab, max_vocab=inner_vocab_size) + assert set(range(inner_vocab_size)) == set(self._tekken_token2id_nospecial.values()), ( + inner_vocab_size, + self._tekken_token2id_nospecial, + ) + + self._model = tiktoken.Encoding( + name=name, + pat_str=pattern, + mergeable_ranks=self._tekken_token2id_nospecial, + special_tokens={}, # special tokens are handled manually + ) + self._all_special_tokens = special_tokens + self._vocab = [self.id_to_piece(i) for i in range(vocab_size)] + self._version = version + self._special_token_policy = SpecialTokenPolicy.RAISE + self._mm_config = mm_config + + @classmethod + def from_file(cls: Type["Tekkenizer"], path: Union[str, Path]) -> "Tekkenizer": + if isinstance(path, str): + path = Path(path) + assert path.exists() + with open(path, "r") as f: + untyped = json.load(f) + if mm := untyped.get("multimodal", None): + untyped["multimodal"] = MultimodalConfig(**mm) + model_data: ModelData = untyped + + _version_str = model_data["config"].get("version") + if _version_str not in TokenizerVersion.__members__: + raise ValueError( + f"Unknown version: {_version_str} in {path}. " + f"Make sure to use a valid version string: {list(TokenizerVersion.__members__)}" + ) + + return cls( + vocab=model_data["vocab"], + pattern=model_data["config"]["pattern"], + vocab_size=model_data["config"]["default_vocab_size"], + num_special_tokens=model_data["config"]["default_num_special_tokens"], + version=TokenizerVersion(_version_str), + name=path.name.replace(".json", ""), + _path=str(path), + mm_config=model_data.get("multimodal"), + ) + + @property + def multimodal(self) -> Optional[MultimodalConfig]: + return self._mm_config + + @multimodal.setter + def multimodal(self, value: MultimodalConfig) -> None: + raise ValueError("Can only set Multimodal config at init") + + @property + def num_special_tokens(self) -> int: + return len(self._all_special_tokens) + + @property + def n_words(self) -> int: + return self._vocab_size + + @property + def version(self) -> TokenizerVersion: + return self._version + + @property + def special_token_policy(self) -> SpecialTokenPolicy: + return self._special_token_policy + + @special_token_policy.setter + def special_token_policy(self, policy: SpecialTokenPolicy) -> None: + self._special_token_policy = policy + + @cached_property + def bos_id(self) -> int: + return self.SPECIAL_TOKENS.index("") + + @cached_property + def eos_id(self) -> int: + return self.SPECIAL_TOKENS.index("") + + @cached_property + def pad_id(self) -> int: + return self.SPECIAL_TOKENS.index("") + + @cached_property + def unk_id(self) -> int: + return self.SPECIAL_TOKENS.index("") + + def vocab(self) -> List[str]: + # when returning self._vocab this will collapse + # all tokens for which we have a decoding error into + # the string. This is bad and results in things + # like len(set(vocab)) != len(vocab)) + # be careful when using self._vocab + return self._vocab + + def encode(self, s: str, bos: bool, eos: bool) -> List[int]: + tokens: List[int] = self._model.encode(s) + tokens = [t + self.num_special_tokens for t in tokens] + if bos: + tokens = [self.bos_id, *tokens] + if eos: + tokens = [*tokens, self.eos_id] + return tokens + + def _decode_all(self, tokens: List[int], special_token_policy: SpecialTokenPolicy) -> List[str]: + # Lump special and non-special tokens together to minimize calls to decode + decoded: List[str] = [] + for is_special, group in groupby(tokens, lambda t: t < self.num_special_tokens): + if is_special: + if special_token_policy == SpecialTokenPolicy.RAISE: + raise ValueError( + f"Decoding `tokens` that contain special tokens ({list(group)}) is not allowed. \n" + "Either make sure `tokens` do not include any special tokens or, " + "if you want to decode `tokens` that includes special tokens, " + "change the tokenizer's special token policy to IGNORE or KEEP: \n" + "```\nfrom mistral_common.tokens.tokenizers.mistral import MistralTokenizer" + "\nfrom mistral_common.tokens.tokenizers.tekken import SpecialTokenPolicy" + "\n\ntokenizer = MistralTokenizer.v3(is_tekken=True)" + "\ntekken = tokenizer.instruct_tokenizer.tokenizer" + "\ntekken.special_token_policy = SpecialTokenPolicy.IGNORE # or SpecialTokenPolicy.KEEP" + "\n```" + ) + elif special_token_policy == SpecialTokenPolicy.KEEP: + decoded.extend(self._all_special_tokens[t] for t in group) + elif special_token_policy == SpecialTokenPolicy.IGNORE: + continue + # TODO: Could use "tokens_str" from vocab.json + # but need to handle null cases. + else: + decoded.append(self._model.decode([t - self.num_special_tokens for t in group])) + return decoded + + def is_byte(self, token_id: int) -> bool: + return 0 <= token_id - self.num_special_tokens < 256 + + def get_control_token(self, s: str) -> int: + try: + return self._all_special_tokens.index(s) + except ValueError: + raise ValueError(f"Unknown control token {s}") + + def decode(self, tokens: List[int]) -> str: + return "".join(self._decode_all(tokens, special_token_policy=self._special_token_policy)) + + def to_string(self, tokens: List[int]) -> str: + return "".join(self._decode_all(tokens, special_token_policy=SpecialTokenPolicy.KEEP)) + + def id_to_piece(self, token_id: int) -> str: + """convert a token id to its string representation.""" + return self._decode_all([token_id], special_token_policy=SpecialTokenPolicy.KEEP)[0] + + def id_to_byte_piece(self, token_id: int) -> bytes: + """convert a token id to its byte representation.""" + if token_id < self.num_special_tokens: + if self._special_token_policy == SpecialTokenPolicy.KEEP: + return self._all_special_tokens[token_id].encode("utf-8") + elif self._special_token_policy == SpecialTokenPolicy.RAISE: + raise ValueError(f"{token_id} is a special token") + + return self._model.decode_single_token_bytes(token_id - self.num_special_tokens) + + +def _reload_mergeable_ranks( + vocab: List[TokenInfo], + max_vocab: Union[int, None] = None, +) -> Dict[bytes, int]: + """ + Reload our tokenizer JSON file and convert it to Tiktoken format. + """ + logger.info(f"Vocab size: {len(vocab)}") + if max_vocab is not None: + assert len(vocab) >= max_vocab, (len(vocab), max_vocab) + vocab = vocab[:max_vocab] + logger.info(f"Cutting vocab to first {len(vocab)} tokens.") + + # build ranks + ranks: Dict[bytes, int] = {} + for i, x in enumerate(vocab): + assert x.keys() == {"rank", "token_bytes", "token_str"} + assert x["rank"] == i + merge = base64.b64decode(x["token_bytes"]) + assert i >= 256 or merge == bytes([i]), (i, merge) + ranks[merge] = x["rank"] + + # sanity check + assert len(ranks) == len(vocab) + assert set(ranks.values()) == set(range(len(ranks))) + + return ranks diff --git a/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/utils.py b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/utils.py new file mode 100644 index 0000000000000000000000000000000000000000..754fa8240b9c367e6bf4383e2af7e0f9a3bcb235 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/mistral_common/tokens/tokenizers/utils.py @@ -0,0 +1,6 @@ +from typing import Iterator, List + + +def chunks(lst: List[str], chunk_size: int) -> Iterator[List[str]]: + for i in range(0, len(lst), chunk_size): + yield lst[i : i + chunk_size] diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/__init__.py b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_deprecations.cpython-311.pyc b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_deprecations.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..75cc39b51c98bd7edb6befa70abeb34dcbb2c318 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_deprecations.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_mrecords.cpython-311.pyc b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_mrecords.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..c8fcbaa65d78164b6001d2a31bdeb78716dc965e Binary files /dev/null and b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_mrecords.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_old_ma.cpython-311.pyc b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_old_ma.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..05a3d3b312235aa298b352abaa7fce5207bb2892 Binary files /dev/null and b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_old_ma.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_regression.cpython-311.pyc b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_regression.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..9eff61b9598636ba6ff12a0a1002ce9aeaa6defb Binary files /dev/null and b/.venv/lib/python3.11/site-packages/numpy/ma/tests/__pycache__/test_regression.cpython-311.pyc differ diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_core.py b/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_core.py new file mode 100644 index 0000000000000000000000000000000000000000..08ddc46c2498e923105bcb47c34abf9b694bdd3f --- /dev/null +++ b/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_core.py @@ -0,0 +1,5687 @@ +# pylint: disable-msg=W0400,W0511,W0611,W0612,W0614,R0201,E1102 +"""Tests suite for MaskedArray & subclassing. + +:author: Pierre Gerard-Marchant +:contact: pierregm_at_uga_dot_edu +""" +__author__ = "Pierre GF Gerard-Marchant" + +import sys +import warnings +import copy +import operator +import itertools +import textwrap +import pytest + +from functools import reduce + + +import numpy as np +import numpy.ma.core +import numpy.core.fromnumeric as fromnumeric +import numpy.core.umath as umath +from numpy.testing import ( + assert_raises, assert_warns, suppress_warnings, IS_WASM + ) +from numpy.testing._private.utils import requires_memory +from numpy import ndarray +from numpy.compat import asbytes +from numpy.ma.testutils import ( + assert_, assert_array_equal, assert_equal, assert_almost_equal, + assert_equal_records, fail_if_equal, assert_not_equal, + assert_mask_equal + ) +from numpy.ma.core import ( + MAError, MaskError, MaskType, MaskedArray, abs, absolute, add, all, + allclose, allequal, alltrue, angle, anom, arange, arccos, arccosh, arctan2, + arcsin, arctan, argsort, array, asarray, choose, concatenate, + conjugate, cos, cosh, count, default_fill_value, diag, divide, doc_note, + empty, empty_like, equal, exp, flatten_mask, filled, fix_invalid, + flatten_structured_array, fromflex, getmask, getmaskarray, greater, + greater_equal, identity, inner, isMaskedArray, less, less_equal, log, + log10, make_mask, make_mask_descr, mask_or, masked, masked_array, + masked_equal, masked_greater, masked_greater_equal, masked_inside, + masked_less, masked_less_equal, masked_not_equal, masked_outside, + masked_print_option, masked_values, masked_where, max, maximum, + maximum_fill_value, min, minimum, minimum_fill_value, mod, multiply, + mvoid, nomask, not_equal, ones, ones_like, outer, power, product, put, + putmask, ravel, repeat, reshape, resize, shape, sin, sinh, sometrue, sort, + sqrt, subtract, sum, take, tan, tanh, transpose, where, zeros, zeros_like, + ) +from numpy.compat import pickle + +pi = np.pi + + +suppress_copy_mask_on_assignment = suppress_warnings() +suppress_copy_mask_on_assignment.filter( + numpy.ma.core.MaskedArrayFutureWarning, + "setting an item on a masked array which has a shared mask will not copy") + + +# For parametrized numeric testing +num_dts = [np.dtype(dt_) for dt_ in '?bhilqBHILQefdgFD'] +num_ids = [dt_.char for dt_ in num_dts] + + +class TestMaskedArray: + # Base test class for MaskedArrays. + + def setup_method(self): + # Base data definition. + x = np.array([1., 1., 1., -2., pi/2.0, 4., 5., -10., 10., 1., 2., 3.]) + y = np.array([5., 0., 3., 2., -1., -4., 0., -10., 10., 1., 0., 3.]) + a10 = 10. + m1 = [1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0] + m2 = [0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 1] + xm = masked_array(x, mask=m1) + ym = masked_array(y, mask=m2) + z = np.array([-.5, 0., .5, .8]) + zm = masked_array(z, mask=[0, 1, 0, 0]) + xf = np.where(m1, 1e+20, x) + xm.set_fill_value(1e+20) + self.d = (x, y, a10, m1, m2, xm, ym, z, zm, xf) + + def test_basicattributes(self): + # Tests some basic array attributes. + a = array([1, 3, 2]) + b = array([1, 3, 2], mask=[1, 0, 1]) + assert_equal(a.ndim, 1) + assert_equal(b.ndim, 1) + assert_equal(a.size, 3) + assert_equal(b.size, 3) + assert_equal(a.shape, (3,)) + assert_equal(b.shape, (3,)) + + def test_basic0d(self): + # Checks masking a scalar + x = masked_array(0) + assert_equal(str(x), '0') + x = masked_array(0, mask=True) + assert_equal(str(x), str(masked_print_option)) + x = masked_array(0, mask=False) + assert_equal(str(x), '0') + x = array(0, mask=1) + assert_(x.filled().dtype is x._data.dtype) + + def test_basic1d(self): + # Test of basic array creation and properties in 1 dimension. + (x, y, a10, m1, m2, xm, ym, z, zm, xf) = self.d + assert_(not isMaskedArray(x)) + assert_(isMaskedArray(xm)) + assert_((xm - ym).filled(0).any()) + fail_if_equal(xm.mask.astype(int), ym.mask.astype(int)) + s = x.shape + assert_equal(np.shape(xm), s) + assert_equal(xm.shape, s) + assert_equal(xm.dtype, x.dtype) + assert_equal(zm.dtype, z.dtype) + assert_equal(xm.size, reduce(lambda x, y:x * y, s)) + assert_equal(count(xm), len(m1) - reduce(lambda x, y:x + y, m1)) + assert_array_equal(xm, xf) + assert_array_equal(filled(xm, 1.e20), xf) + assert_array_equal(x, xm) + + def test_basic2d(self): + # Test of basic array creation and properties in 2 dimensions. + (x, y, a10, m1, m2, xm, ym, z, zm, xf) = self.d + for s in [(4, 3), (6, 2)]: + x.shape = s + y.shape = s + xm.shape = s + ym.shape = s + xf.shape = s + + assert_(not isMaskedArray(x)) + assert_(isMaskedArray(xm)) + assert_equal(shape(xm), s) + assert_equal(xm.shape, s) + assert_equal(xm.size, reduce(lambda x, y:x * y, s)) + assert_equal(count(xm), len(m1) - reduce(lambda x, y:x + y, m1)) + assert_equal(xm, xf) + assert_equal(filled(xm, 1.e20), xf) + assert_equal(x, xm) + + def test_concatenate_basic(self): + # Tests concatenations. + (x, y, a10, m1, m2, xm, ym, z, zm, xf) = self.d + # basic concatenation + assert_equal(np.concatenate((x, y)), concatenate((xm, ym))) + assert_equal(np.concatenate((x, y)), concatenate((x, y))) + assert_equal(np.concatenate((x, y)), concatenate((xm, y))) + assert_equal(np.concatenate((x, y, x)), concatenate((x, ym, x))) + + def test_concatenate_alongaxis(self): + # Tests concatenations. + (x, y, a10, m1, m2, xm, ym, z, zm, xf) = self.d + # Concatenation along an axis + s = (3, 4) + x.shape = y.shape = xm.shape = ym.shape = s + assert_equal(xm.mask, np.reshape(m1, s)) + assert_equal(ym.mask, np.reshape(m2, s)) + xmym = concatenate((xm, ym), 1) + assert_equal(np.concatenate((x, y), 1), xmym) + assert_equal(np.concatenate((xm.mask, ym.mask), 1), xmym._mask) + + x = zeros(2) + y = array(ones(2), mask=[False, True]) + z = concatenate((x, y)) + assert_array_equal(z, [0, 0, 1, 1]) + assert_array_equal(z.mask, [False, False, False, True]) + z = concatenate((y, x)) + assert_array_equal(z, [1, 1, 0, 0]) + assert_array_equal(z.mask, [False, True, False, False]) + + def test_concatenate_flexible(self): + # Tests the concatenation on flexible arrays. + data = masked_array(list(zip(np.random.rand(10), + np.arange(10))), + dtype=[('a', float), ('b', int)]) + + test = concatenate([data[:5], data[5:]]) + assert_equal_records(test, data) + + def test_creation_ndmin(self): + # Check the use of ndmin + x = array([1, 2, 3], mask=[1, 0, 0], ndmin=2) + assert_equal(x.shape, (1, 3)) + assert_equal(x._data, [[1, 2, 3]]) + assert_equal(x._mask, [[1, 0, 0]]) + + def test_creation_ndmin_from_maskedarray(self): + # Make sure we're not losing the original mask w/ ndmin + x = array([1, 2, 3]) + x[-1] = masked + xx = array(x, ndmin=2, dtype=float) + assert_equal(x.shape, x._mask.shape) + assert_equal(xx.shape, xx._mask.shape) + + def test_creation_maskcreation(self): + # Tests how masks are initialized at the creation of Maskedarrays. + data = arange(24, dtype=float) + data[[3, 6, 15]] = masked + dma_1 = MaskedArray(data) + assert_equal(dma_1.mask, data.mask) + dma_2 = MaskedArray(dma_1) + assert_equal(dma_2.mask, dma_1.mask) + dma_3 = MaskedArray(dma_1, mask=[1, 0, 0, 0] * 6) + fail_if_equal(dma_3.mask, dma_1.mask) + + x = array([1, 2, 3], mask=True) + assert_equal(x._mask, [True, True, True]) + x = array([1, 2, 3], mask=False) + assert_equal(x._mask, [False, False, False]) + y = array([1, 2, 3], mask=x._mask, copy=False) + assert_(np.may_share_memory(x.mask, y.mask)) + y = array([1, 2, 3], mask=x._mask, copy=True) + assert_(not np.may_share_memory(x.mask, y.mask)) + x = array([1, 2, 3], mask=None) + assert_equal(x._mask, [False, False, False]) + + def test_masked_singleton_array_creation_warns(self): + # The first works, but should not (ideally), there may be no way + # to solve this, however, as long as `np.ma.masked` is an ndarray. + np.array(np.ma.masked) + with pytest.warns(UserWarning): + # Tries to create a float array, using `float(np.ma.masked)`. + # We may want to define this is invalid behaviour in the future! + # (requiring np.ma.masked to be a known NumPy scalar probably + # with a DType.) + np.array([3., np.ma.masked]) + + def test_creation_with_list_of_maskedarrays(self): + # Tests creating a masked array from a list of masked arrays. + x = array(np.arange(5), mask=[1, 0, 0, 0, 0]) + data = array((x, x[::-1])) + assert_equal(data, [[0, 1, 2, 3, 4], [4, 3, 2, 1, 0]]) + assert_equal(data._mask, [[1, 0, 0, 0, 0], [0, 0, 0, 0, 1]]) + + x.mask = nomask + data = array((x, x[::-1])) + assert_equal(data, [[0, 1, 2, 3, 4], [4, 3, 2, 1, 0]]) + assert_(data.mask is nomask) + + def test_creation_with_list_of_maskedarrays_no_bool_cast(self): + # Tests the regression in gh-18551 + masked_str = np.ma.masked_array(['a', 'b'], mask=[True, False]) + normal_int = np.arange(2) + res = np.ma.asarray([masked_str, normal_int], dtype="U21") + assert_array_equal(res.mask, [[True, False], [False, False]]) + + # The above only failed due a long chain of oddity, try also with + # an object array that cannot be converted to bool always: + class NotBool(): + def __bool__(self): + raise ValueError("not a bool!") + masked_obj = np.ma.masked_array([NotBool(), 'b'], mask=[True, False]) + # Check that the NotBool actually fails like we would expect: + with pytest.raises(ValueError, match="not a bool!"): + np.asarray([masked_obj], dtype=bool) + + res = np.ma.asarray([masked_obj, normal_int]) + assert_array_equal(res.mask, [[True, False], [False, False]]) + + def test_creation_from_ndarray_with_padding(self): + x = np.array([('A', 0)], dtype={'names':['f0','f1'], + 'formats':['S4','i8'], + 'offsets':[0,8]}) + array(x) # used to fail due to 'V' padding field in x.dtype.descr + + def test_unknown_keyword_parameter(self): + with pytest.raises(TypeError, match="unexpected keyword argument"): + MaskedArray([1, 2, 3], maks=[0, 1, 0]) # `mask` is misspelled. + + def test_asarray(self): + (x, y, a10, m1, m2, xm, ym, z, zm, xf) = self.d + xm.fill_value = -9999 + xm._hardmask = True + xmm = asarray(xm) + assert_equal(xmm._data, xm._data) + assert_equal(xmm._mask, xm._mask) + assert_equal(xmm.fill_value, xm.fill_value) + assert_equal(xmm._hardmask, xm._hardmask) + + def test_asarray_default_order(self): + # See Issue #6646 + m = np.eye(3).T + assert_(not m.flags.c_contiguous) + + new_m = asarray(m) + assert_(new_m.flags.c_contiguous) + + def test_asarray_enforce_order(self): + # See Issue #6646 + m = np.eye(3).T + assert_(not m.flags.c_contiguous) + + new_m = asarray(m, order='C') + assert_(new_m.flags.c_contiguous) + + def test_fix_invalid(self): + # Checks fix_invalid. + with np.errstate(invalid='ignore'): + data = masked_array([np.nan, 0., 1.], mask=[0, 0, 1]) + data_fixed = fix_invalid(data) + assert_equal(data_fixed._data, [data.fill_value, 0., 1.]) + assert_equal(data_fixed._mask, [1., 0., 1.]) + + def test_maskedelement(self): + # Test of masked element + x = arange(6) + x[1] = masked + assert_(str(masked) == '--') + assert_(x[1] is masked) + assert_equal(filled(x[1], 0), 0) + + def test_set_element_as_object(self): + # Tests setting elements with object + a = empty(1, dtype=object) + x = (1, 2, 3, 4, 5) + a[0] = x + assert_equal(a[0], x) + assert_(a[0] is x) + + import datetime + dt = datetime.datetime.now() + a[0] = dt + assert_(a[0] is dt) + + def test_indexing(self): + # Tests conversions and indexing + x1 = np.array([1, 2, 4, 3]) + x2 = array(x1, mask=[1, 0, 0, 0]) + x3 = array(x1, mask=[0, 1, 0, 1]) + x4 = array(x1) + # test conversion to strings + str(x2) # raises? + repr(x2) # raises? + assert_equal(np.sort(x1), sort(x2, endwith=False)) + # tests of indexing + assert_(type(x2[1]) is type(x1[1])) + assert_(x1[1] == x2[1]) + assert_(x2[0] is masked) + assert_equal(x1[2], x2[2]) + assert_equal(x1[2:5], x2[2:5]) + assert_equal(x1[:], x2[:]) + assert_equal(x1[1:], x3[1:]) + x1[2] = 9 + x2[2] = 9 + assert_equal(x1, x2) + x1[1:3] = 99 + x2[1:3] = 99 + assert_equal(x1, x2) + x2[1] = masked + assert_equal(x1, x2) + x2[1:3] = masked + assert_equal(x1, x2) + x2[:] = x1 + x2[1] = masked + assert_(allequal(getmask(x2), array([0, 1, 0, 0]))) + x3[:] = masked_array([1, 2, 3, 4], [0, 1, 1, 0]) + assert_(allequal(getmask(x3), array([0, 1, 1, 0]))) + x4[:] = masked_array([1, 2, 3, 4], [0, 1, 1, 0]) + assert_(allequal(getmask(x4), array([0, 1, 1, 0]))) + assert_(allequal(x4, array([1, 2, 3, 4]))) + x1 = np.arange(5) * 1.0 + x2 = masked_values(x1, 3.0) + assert_equal(x1, x2) + assert_(allequal(array([0, 0, 0, 1, 0], MaskType), x2.mask)) + assert_equal(3.0, x2.fill_value) + x1 = array([1, 'hello', 2, 3], object) + x2 = np.array([1, 'hello', 2, 3], object) + s1 = x1[1] + s2 = x2[1] + assert_equal(type(s2), str) + assert_equal(type(s1), str) + assert_equal(s1, s2) + assert_(x1[1:1].shape == (0,)) + + def test_setitem_no_warning(self): + # Setitem shouldn't warn, because the assignment might be masked + # and warning for a masked assignment is weird (see gh-23000) + # (When the value is masked, otherwise a warning would be acceptable + # but is not given currently.) + x = np.ma.arange(60).reshape((6, 10)) + index = (slice(1, 5, 2), [7, 5]) + value = np.ma.masked_all((2, 2)) + value._data[...] = np.inf # not a valid integer... + x[index] = value + # The masked scalar is special cased, but test anyway (it's NaN): + x[...] = np.ma.masked + # Finally, a large value that cannot be cast to the float32 `x` + x = np.ma.arange(3., dtype=np.float32) + value = np.ma.array([2e234, 1, 1], mask=[True, False, False]) + x[...] = value + x[[0, 1, 2]] = value + + @suppress_copy_mask_on_assignment + def test_copy(self): + # Tests of some subtle points of copying and sizing. + n = [0, 0, 1, 0, 0] + m = make_mask(n) + m2 = make_mask(m) + assert_(m is m2) + m3 = make_mask(m, copy=True) + assert_(m is not m3) + + x1 = np.arange(5) + y1 = array(x1, mask=m) + assert_equal(y1._data.__array_interface__, x1.__array_interface__) + assert_(allequal(x1, y1.data)) + assert_equal(y1._mask.__array_interface__, m.__array_interface__) + + y1a = array(y1) + # Default for masked array is not to copy; see gh-10318. + assert_(y1a._data.__array_interface__ == + y1._data.__array_interface__) + assert_(y1a._mask.__array_interface__ == + y1._mask.__array_interface__) + + y2 = array(x1, mask=m3) + assert_(y2._data.__array_interface__ == x1.__array_interface__) + assert_(y2._mask.__array_interface__ == m3.__array_interface__) + assert_(y2[2] is masked) + y2[2] = 9 + assert_(y2[2] is not masked) + assert_(y2._mask.__array_interface__ == m3.__array_interface__) + assert_(allequal(y2.mask, 0)) + + y2a = array(x1, mask=m, copy=1) + assert_(y2a._data.__array_interface__ != x1.__array_interface__) + #assert_( y2a._mask is not m) + assert_(y2a._mask.__array_interface__ != m.__array_interface__) + assert_(y2a[2] is masked) + y2a[2] = 9 + assert_(y2a[2] is not masked) + #assert_( y2a._mask is not m) + assert_(y2a._mask.__array_interface__ != m.__array_interface__) + assert_(allequal(y2a.mask, 0)) + + y3 = array(x1 * 1.0, mask=m) + assert_(filled(y3).dtype is (x1 * 1.0).dtype) + + x4 = arange(4) + x4[2] = masked + y4 = resize(x4, (8,)) + assert_equal(concatenate([x4, x4]), y4) + assert_equal(getmask(y4), [0, 0, 1, 0, 0, 0, 1, 0]) + y5 = repeat(x4, (2, 2, 2, 2), axis=0) + assert_equal(y5, [0, 0, 1, 1, 2, 2, 3, 3]) + y6 = repeat(x4, 2, axis=0) + assert_equal(y5, y6) + y7 = x4.repeat((2, 2, 2, 2), axis=0) + assert_equal(y5, y7) + y8 = x4.repeat(2, 0) + assert_equal(y5, y8) + + y9 = x4.copy() + assert_equal(y9._data, x4._data) + assert_equal(y9._mask, x4._mask) + + x = masked_array([1, 2, 3], mask=[0, 1, 0]) + # Copy is False by default + y = masked_array(x) + assert_equal(y._data.ctypes.data, x._data.ctypes.data) + assert_equal(y._mask.ctypes.data, x._mask.ctypes.data) + y = masked_array(x, copy=True) + assert_not_equal(y._data.ctypes.data, x._data.ctypes.data) + assert_not_equal(y._mask.ctypes.data, x._mask.ctypes.data) + + def test_copy_0d(self): + # gh-9430 + x = np.ma.array(43, mask=True) + xc = x.copy() + assert_equal(xc.mask, True) + + def test_copy_on_python_builtins(self): + # Tests copy works on python builtins (issue#8019) + assert_(isMaskedArray(np.ma.copy([1,2,3]))) + assert_(isMaskedArray(np.ma.copy((1,2,3)))) + + def test_copy_immutable(self): + # Tests that the copy method is immutable, GitHub issue #5247 + a = np.ma.array([1, 2, 3]) + b = np.ma.array([4, 5, 6]) + a_copy_method = a.copy + b.copy + assert_equal(a_copy_method(), [1, 2, 3]) + + def test_deepcopy(self): + from copy import deepcopy + a = array([0, 1, 2], mask=[False, True, False]) + copied = deepcopy(a) + assert_equal(copied.mask, a.mask) + assert_not_equal(id(a._mask), id(copied._mask)) + + copied[1] = 1 + assert_equal(copied.mask, [0, 0, 0]) + assert_equal(a.mask, [0, 1, 0]) + + copied = deepcopy(a) + assert_equal(copied.mask, a.mask) + copied.mask[1] = False + assert_equal(copied.mask, [0, 0, 0]) + assert_equal(a.mask, [0, 1, 0]) + + def test_format(self): + a = array([0, 1, 2], mask=[False, True, False]) + assert_equal(format(a), "[0 -- 2]") + assert_equal(format(masked), "--") + assert_equal(format(masked, ""), "--") + + # Postponed from PR #15410, perhaps address in the future. + # assert_equal(format(masked, " >5"), " --") + # assert_equal(format(masked, " <5"), "-- ") + + # Expect a FutureWarning for using format_spec with MaskedElement + with assert_warns(FutureWarning): + with_format_string = format(masked, " >5") + assert_equal(with_format_string, "--") + + def test_str_repr(self): + a = array([0, 1, 2], mask=[False, True, False]) + assert_equal(str(a), '[0 -- 2]') + assert_equal( + repr(a), + textwrap.dedent('''\ + masked_array(data=[0, --, 2], + mask=[False, True, False], + fill_value=999999)''') + ) + + # arrays with a continuation + a = np.ma.arange(2000) + a[1:50] = np.ma.masked + assert_equal( + repr(a), + textwrap.dedent('''\ + masked_array(data=[0, --, --, ..., 1997, 1998, 1999], + mask=[False, True, True, ..., False, False, False], + fill_value=999999)''') + ) + + # line-wrapped 1d arrays are correctly aligned + a = np.ma.arange(20) + assert_equal( + repr(a), + textwrap.dedent('''\ + masked_array(data=[ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, + 14, 15, 16, 17, 18, 19], + mask=False, + fill_value=999999)''') + ) + + # 2d arrays cause wrapping + a = array([[1, 2, 3], [4, 5, 6]], dtype=np.int8) + a[1,1] = np.ma.masked + assert_equal( + repr(a), + textwrap.dedent('''\ + masked_array( + data=[[1, 2, 3], + [4, --, 6]], + mask=[[False, False, False], + [False, True, False]], + fill_value=999999, + dtype=int8)''') + ) + + # but not it they're a row vector + assert_equal( + repr(a[:1]), + textwrap.dedent('''\ + masked_array(data=[[1, 2, 3]], + mask=[[False, False, False]], + fill_value=999999, + dtype=int8)''') + ) + + # dtype=int is implied, so not shown + assert_equal( + repr(a.astype(int)), + textwrap.dedent('''\ + masked_array( + data=[[1, 2, 3], + [4, --, 6]], + mask=[[False, False, False], + [False, True, False]], + fill_value=999999)''') + ) + + def test_str_repr_legacy(self): + oldopts = np.get_printoptions() + np.set_printoptions(legacy='1.13') + try: + a = array([0, 1, 2], mask=[False, True, False]) + assert_equal(str(a), '[0 -- 2]') + assert_equal(repr(a), 'masked_array(data = [0 -- 2],\n' + ' mask = [False True False],\n' + ' fill_value = 999999)\n') + + a = np.ma.arange(2000) + a[1:50] = np.ma.masked + assert_equal( + repr(a), + 'masked_array(data = [0 -- -- ..., 1997 1998 1999],\n' + ' mask = [False True True ..., False False False],\n' + ' fill_value = 999999)\n' + ) + finally: + np.set_printoptions(**oldopts) + + def test_0d_unicode(self): + u = 'caf\xe9' + utype = type(u) + + arr_nomask = np.ma.array(u) + arr_masked = np.ma.array(u, mask=True) + + assert_equal(utype(arr_nomask), u) + assert_equal(utype(arr_masked), '--') + + def test_pickling(self): + # Tests pickling + for dtype in (int, float, str, object): + a = arange(10).astype(dtype) + a.fill_value = 999 + + masks = ([0, 0, 0, 1, 0, 1, 0, 1, 0, 1], # partially masked + True, # Fully masked + False) # Fully unmasked + + for proto in range(2, pickle.HIGHEST_PROTOCOL + 1): + for mask in masks: + a.mask = mask + a_pickled = pickle.loads(pickle.dumps(a, protocol=proto)) + assert_equal(a_pickled._mask, a._mask) + assert_equal(a_pickled._data, a._data) + if dtype in (object, int): + assert_equal(a_pickled.fill_value, 999) + else: + assert_equal(a_pickled.fill_value, dtype(999)) + assert_array_equal(a_pickled.mask, mask) + + def test_pickling_subbaseclass(self): + # Test pickling w/ a subclass of ndarray + x = np.array([(1.0, 2), (3.0, 4)], + dtype=[('x', float), ('y', int)]).view(np.recarray) + a = masked_array(x, mask=[(True, False), (False, True)]) + for proto in range(2, pickle.HIGHEST_PROTOCOL + 1): + a_pickled = pickle.loads(pickle.dumps(a, protocol=proto)) + assert_equal(a_pickled._mask, a._mask) + assert_equal(a_pickled, a) + assert_(isinstance(a_pickled._data, np.recarray)) + + def test_pickling_maskedconstant(self): + # Test pickling MaskedConstant + mc = np.ma.masked + for proto in range(2, pickle.HIGHEST_PROTOCOL + 1): + mc_pickled = pickle.loads(pickle.dumps(mc, protocol=proto)) + assert_equal(mc_pickled._baseclass, mc._baseclass) + assert_equal(mc_pickled._mask, mc._mask) + assert_equal(mc_pickled._data, mc._data) + + def test_pickling_wstructured(self): + # Tests pickling w/ structured array + a = array([(1, 1.), (2, 2.)], mask=[(0, 0), (0, 1)], + dtype=[('a', int), ('b', float)]) + for proto in range(2, pickle.HIGHEST_PROTOCOL + 1): + a_pickled = pickle.loads(pickle.dumps(a, protocol=proto)) + assert_equal(a_pickled._mask, a._mask) + assert_equal(a_pickled, a) + + def test_pickling_keepalignment(self): + # Tests pickling w/ F_CONTIGUOUS arrays + a = arange(10) + a.shape = (-1, 2) + b = a.T + for proto in range(2, pickle.HIGHEST_PROTOCOL + 1): + test = pickle.loads(pickle.dumps(b, protocol=proto)) + assert_equal(test, b) + + def test_single_element_subscript(self): + # Tests single element subscripts of Maskedarrays. + a = array([1, 3, 2]) + b = array([1, 3, 2], mask=[1, 0, 1]) + assert_equal(a[0].shape, ()) + assert_equal(b[0].shape, ()) + assert_equal(b[1].shape, ()) + + def test_topython(self): + # Tests some communication issues with Python. + assert_equal(1, int(array(1))) + assert_equal(1.0, float(array(1))) + assert_equal(1, int(array([[[1]]]))) + assert_equal(1.0, float(array([[1]]))) + assert_raises(TypeError, float, array([1, 1])) + + with suppress_warnings() as sup: + sup.filter(UserWarning, 'Warning: converting a masked element') + assert_(np.isnan(float(array([1], mask=[1])))) + + a = array([1, 2, 3], mask=[1, 0, 0]) + assert_raises(TypeError, lambda: float(a)) + assert_equal(float(a[-1]), 3.) + assert_(np.isnan(float(a[0]))) + assert_raises(TypeError, int, a) + assert_equal(int(a[-1]), 3) + assert_raises(MAError, lambda:int(a[0])) + + def test_oddfeatures_1(self): + # Test of other odd features + x = arange(20) + x = x.reshape(4, 5) + x.flat[5] = 12 + assert_(x[1, 0] == 12) + z = x + 10j * x + assert_equal(z.real, x) + assert_equal(z.imag, 10 * x) + assert_equal((z * conjugate(z)).real, 101 * x * x) + z.imag[...] = 0.0 + + x = arange(10) + x[3] = masked + assert_(str(x[3]) == str(masked)) + c = x >= 8 + assert_(count(where(c, masked, masked)) == 0) + assert_(shape(where(c, masked, masked)) == c.shape) + + z = masked_where(c, x) + assert_(z.dtype is x.dtype) + assert_(z[3] is masked) + assert_(z[4] is not masked) + assert_(z[7] is not masked) + assert_(z[8] is masked) + assert_(z[9] is masked) + assert_equal(x, z) + + def test_oddfeatures_2(self): + # Tests some more features. + x = array([1., 2., 3., 4., 5.]) + c = array([1, 1, 1, 0, 0]) + x[2] = masked + z = where(c, x, -x) + assert_equal(z, [1., 2., 0., -4., -5]) + c[0] = masked + z = where(c, x, -x) + assert_equal(z, [1., 2., 0., -4., -5]) + assert_(z[0] is masked) + assert_(z[1] is not masked) + assert_(z[2] is masked) + + @suppress_copy_mask_on_assignment + def test_oddfeatures_3(self): + # Tests some generic features + atest = array([10], mask=True) + btest = array([20]) + idx = atest.mask + atest[idx] = btest[idx] + assert_equal(atest, [20]) + + def test_filled_with_object_dtype(self): + a = np.ma.masked_all(1, dtype='O') + assert_equal(a.filled('x')[0], 'x') + + def test_filled_with_flexible_dtype(self): + # Test filled w/ flexible dtype + flexi = array([(1, 1, 1)], + dtype=[('i', int), ('s', '|S8'), ('f', float)]) + flexi[0] = masked + assert_equal(flexi.filled(), + np.array([(default_fill_value(0), + default_fill_value('0'), + default_fill_value(0.),)], dtype=flexi.dtype)) + flexi[0] = masked + assert_equal(flexi.filled(1), + np.array([(1, '1', 1.)], dtype=flexi.dtype)) + + def test_filled_with_mvoid(self): + # Test filled w/ mvoid + ndtype = [('a', int), ('b', float)] + a = mvoid((1, 2.), mask=[(0, 1)], dtype=ndtype) + # Filled using default + test = a.filled() + assert_equal(tuple(test), (1, default_fill_value(1.))) + # Explicit fill_value + test = a.filled((-1, -1)) + assert_equal(tuple(test), (1, -1)) + # Using predefined filling values + a.fill_value = (-999, -999) + assert_equal(tuple(a.filled()), (1, -999)) + + def test_filled_with_nested_dtype(self): + # Test filled w/ nested dtype + ndtype = [('A', int), ('B', [('BA', int), ('BB', int)])] + a = array([(1, (1, 1)), (2, (2, 2))], + mask=[(0, (1, 0)), (0, (0, 1))], dtype=ndtype) + test = a.filled(0) + control = np.array([(1, (0, 1)), (2, (2, 0))], dtype=ndtype) + assert_equal(test, control) + + test = a['B'].filled(0) + control = np.array([(0, 1), (2, 0)], dtype=a['B'].dtype) + assert_equal(test, control) + + # test if mask gets set correctly (see #6760) + Z = numpy.ma.zeros(2, numpy.dtype([("A", "(2,2)i1,(2,2)i1", (2,2))])) + assert_equal(Z.data.dtype, numpy.dtype([('A', [('f0', 'i1', (2, 2)), + ('f1', 'i1', (2, 2))], (2, 2))])) + assert_equal(Z.mask.dtype, numpy.dtype([('A', [('f0', '?', (2, 2)), + ('f1', '?', (2, 2))], (2, 2))])) + + def test_filled_with_f_order(self): + # Test filled w/ F-contiguous array + a = array(np.array([(0, 1, 2), (4, 5, 6)], order='F'), + mask=np.array([(0, 0, 1), (1, 0, 0)], order='F'), + order='F') # this is currently ignored + assert_(a.flags['F_CONTIGUOUS']) + assert_(a.filled(0).flags['F_CONTIGUOUS']) + + def test_optinfo_propagation(self): + # Checks that _optinfo dictionary isn't back-propagated + x = array([1, 2, 3, ], dtype=float) + x._optinfo['info'] = '???' + y = x.copy() + assert_equal(y._optinfo['info'], '???') + y._optinfo['info'] = '!!!' + assert_equal(x._optinfo['info'], '???') + + def test_optinfo_forward_propagation(self): + a = array([1,2,2,4]) + a._optinfo["key"] = "value" + assert_equal(a._optinfo["key"], (a == 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a != 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a > 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a >= 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a <= 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a + 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a - 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a * 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], (a / 2)._optinfo["key"]) + assert_equal(a._optinfo["key"], a[:2]._optinfo["key"]) + assert_equal(a._optinfo["key"], a[[0,0,2]]._optinfo["key"]) + assert_equal(a._optinfo["key"], np.exp(a)._optinfo["key"]) + assert_equal(a._optinfo["key"], np.abs(a)._optinfo["key"]) + assert_equal(a._optinfo["key"], array(a, copy=True)._optinfo["key"]) + assert_equal(a._optinfo["key"], np.zeros_like(a)._optinfo["key"]) + + def test_fancy_printoptions(self): + # Test printing a masked array w/ fancy dtype. + fancydtype = np.dtype([('x', int), ('y', [('t', int), ('s', float)])]) + test = array([(1, (2, 3.0)), (4, (5, 6.0))], + mask=[(1, (0, 1)), (0, (1, 0))], + dtype=fancydtype) + control = "[(--, (2, --)) (4, (--, 6.0))]" + assert_equal(str(test), control) + + # Test 0-d array with multi-dimensional dtype + t_2d0 = masked_array(data = (0, [[0.0, 0.0, 0.0], + [0.0, 0.0, 0.0]], + 0.0), + mask = (False, [[True, False, True], + [False, False, True]], + False), + dtype = "int, (2,3)float, float") + control = "(0, [[--, 0.0, --], [0.0, 0.0, --]], 0.0)" + assert_equal(str(t_2d0), control) + + def test_flatten_structured_array(self): + # Test flatten_structured_array on arrays + # On ndarray + ndtype = [('a', int), ('b', float)] + a = np.array([(1, 1), (2, 2)], dtype=ndtype) + test = flatten_structured_array(a) + control = np.array([[1., 1.], [2., 2.]], dtype=float) + assert_equal(test, control) + assert_equal(test.dtype, control.dtype) + # On masked_array + a = array([(1, 1), (2, 2)], mask=[(0, 1), (1, 0)], dtype=ndtype) + test = flatten_structured_array(a) + control = array([[1., 1.], [2., 2.]], + mask=[[0, 1], [1, 0]], dtype=float) + assert_equal(test, control) + assert_equal(test.dtype, control.dtype) + assert_equal(test.mask, control.mask) + # On masked array with nested structure + ndtype = [('a', int), ('b', [('ba', int), ('bb', float)])] + a = array([(1, (1, 1.1)), (2, (2, 2.2))], + mask=[(0, (1, 0)), (1, (0, 1))], dtype=ndtype) + test = flatten_structured_array(a) + control = array([[1., 1., 1.1], [2., 2., 2.2]], + mask=[[0, 1, 0], [1, 0, 1]], dtype=float) + assert_equal(test, control) + assert_equal(test.dtype, control.dtype) + assert_equal(test.mask, control.mask) + # Keeping the initial shape + ndtype = [('a', int), ('b', float)] + a = np.array([[(1, 1), ], [(2, 2), ]], dtype=ndtype) + test = flatten_structured_array(a) + control = np.array([[[1., 1.], ], [[2., 2.], ]], dtype=float) + assert_equal(test, control) + assert_equal(test.dtype, control.dtype) + + def test_void0d(self): + # Test creating a mvoid object + ndtype = [('a', int), ('b', int)] + a = np.array([(1, 2,)], dtype=ndtype)[0] + f = mvoid(a) + assert_(isinstance(f, mvoid)) + + a = masked_array([(1, 2)], mask=[(1, 0)], dtype=ndtype)[0] + assert_(isinstance(a, mvoid)) + + a = masked_array([(1, 2), (1, 2)], mask=[(1, 0), (0, 0)], dtype=ndtype) + f = mvoid(a._data[0], a._mask[0]) + assert_(isinstance(f, mvoid)) + + def test_mvoid_getitem(self): + # Test mvoid.__getitem__ + ndtype = [('a', int), ('b', int)] + a = masked_array([(1, 2,), (3, 4)], mask=[(0, 0), (1, 0)], + dtype=ndtype) + # w/o mask + f = a[0] + assert_(isinstance(f, mvoid)) + assert_equal((f[0], f['a']), (1, 1)) + assert_equal(f['b'], 2) + # w/ mask + f = a[1] + assert_(isinstance(f, mvoid)) + assert_(f[0] is masked) + assert_(f['a'] is masked) + assert_equal(f[1], 4) + + # exotic dtype + A = masked_array(data=[([0,1],)], + mask=[([True, False],)], + dtype=[("A", ">i2", (2,))]) + assert_equal(A[0]["A"], A["A"][0]) + assert_equal(A[0]["A"], masked_array(data=[0, 1], + mask=[True, False], dtype=">i2")) + + def test_mvoid_iter(self): + # Test iteration on __getitem__ + ndtype = [('a', int), ('b', int)] + a = masked_array([(1, 2,), (3, 4)], mask=[(0, 0), (1, 0)], + dtype=ndtype) + # w/o mask + assert_equal(list(a[0]), [1, 2]) + # w/ mask + assert_equal(list(a[1]), [masked, 4]) + + def test_mvoid_print(self): + # Test printing a mvoid + mx = array([(1, 1), (2, 2)], dtype=[('a', int), ('b', int)]) + assert_equal(str(mx[0]), "(1, 1)") + mx['b'][0] = masked + ini_display = masked_print_option._display + masked_print_option.set_display("-X-") + try: + assert_equal(str(mx[0]), "(1, -X-)") + assert_equal(repr(mx[0]), "(1, -X-)") + finally: + masked_print_option.set_display(ini_display) + + # also check if there are object datatypes (see gh-7493) + mx = array([(1,), (2,)], dtype=[('a', 'O')]) + assert_equal(str(mx[0]), "(1,)") + + def test_mvoid_multidim_print(self): + + # regression test for gh-6019 + t_ma = masked_array(data = [([1, 2, 3],)], + mask = [([False, True, False],)], + fill_value = ([999999, 999999, 999999],), + dtype = [('a', ' 1: + assert_equal(np.concatenate((x, y), 1), concatenate((xm, ym), 1)) + assert_equal(np.add.reduce(x, 1), add.reduce(x, 1)) + assert_equal(np.sum(x, 1), sum(x, 1)) + assert_equal(np.prod(x, 1), product(x, 1)) + + def test_binops_d2D(self): + # Test binary operations on 2D data + a = array([[1.], [2.], [3.]], mask=[[False], [True], [True]]) + b = array([[2., 3.], [4., 5.], [6., 7.]]) + + test = a * b + control = array([[2., 3.], [2., 2.], [3., 3.]], + mask=[[0, 0], [1, 1], [1, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + test = b * a + control = array([[2., 3.], [4., 5.], [6., 7.]], + mask=[[0, 0], [1, 1], [1, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + a = array([[1.], [2.], [3.]]) + b = array([[2., 3.], [4., 5.], [6., 7.]], + mask=[[0, 0], [0, 0], [0, 1]]) + test = a * b + control = array([[2, 3], [8, 10], [18, 3]], + mask=[[0, 0], [0, 0], [0, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + test = b * a + control = array([[2, 3], [8, 10], [18, 7]], + mask=[[0, 0], [0, 0], [0, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + def test_domained_binops_d2D(self): + # Test domained binary operations on 2D data + a = array([[1.], [2.], [3.]], mask=[[False], [True], [True]]) + b = array([[2., 3.], [4., 5.], [6., 7.]]) + + test = a / b + control = array([[1. / 2., 1. / 3.], [2., 2.], [3., 3.]], + mask=[[0, 0], [1, 1], [1, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + test = b / a + control = array([[2. / 1., 3. / 1.], [4., 5.], [6., 7.]], + mask=[[0, 0], [1, 1], [1, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + a = array([[1.], [2.], [3.]]) + b = array([[2., 3.], [4., 5.], [6., 7.]], + mask=[[0, 0], [0, 0], [0, 1]]) + test = a / b + control = array([[1. / 2, 1. / 3], [2. / 4, 2. / 5], [3. / 6, 3]], + mask=[[0, 0], [0, 0], [0, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + test = b / a + control = array([[2 / 1., 3 / 1.], [4 / 2., 5 / 2.], [6 / 3., 7]], + mask=[[0, 0], [0, 0], [0, 1]]) + assert_equal(test, control) + assert_equal(test.data, control.data) + assert_equal(test.mask, control.mask) + + def test_noshrinking(self): + # Check that we don't shrink a mask when not wanted + # Binary operations + a = masked_array([1., 2., 3.], mask=[False, False, False], + shrink=False) + b = a + 1 + assert_equal(b.mask, [0, 0, 0]) + # In place binary operation + a += 1 + assert_equal(a.mask, [0, 0, 0]) + # Domained binary operation + b = a / 1. + assert_equal(b.mask, [0, 0, 0]) + # In place binary operation + a /= 1. + assert_equal(a.mask, [0, 0, 0]) + + def test_ufunc_nomask(self): + # check the case ufuncs should set the mask to false + m = np.ma.array([1]) + # check we don't get array([False], dtype=bool) + assert_equal(np.true_divide(m, 5).mask.shape, ()) + + def test_noshink_on_creation(self): + # Check that the mask is not shrunk on array creation when not wanted + a = np.ma.masked_values([1., 2.5, 3.1], 1.5, shrink=False) + assert_equal(a.mask, [0, 0, 0]) + + def test_mod(self): + # Tests mod + (x, y, a10, m1, m2, xm, ym, z, zm, xf) = self.d + assert_equal(mod(x, y), mod(xm, ym)) + test = mod(ym, xm) + assert_equal(test, np.mod(ym, xm)) + assert_equal(test.mask, mask_or(xm.mask, ym.mask)) + test = mod(xm, ym) + assert_equal(test, np.mod(xm, ym)) + assert_equal(test.mask, mask_or(mask_or(xm.mask, ym.mask), (ym == 0))) + + def test_TakeTransposeInnerOuter(self): + # Test of take, transpose, inner, outer products + x = arange(24) + y = np.arange(24) + x[5:6] = masked + x = x.reshape(2, 3, 4) + y = y.reshape(2, 3, 4) + assert_equal(np.transpose(y, (2, 0, 1)), transpose(x, (2, 0, 1))) + assert_equal(np.take(y, (2, 0, 1), 1), take(x, (2, 0, 1), 1)) + assert_equal(np.inner(filled(x, 0), filled(y, 0)), + inner(x, y)) + assert_equal(np.outer(filled(x, 0), filled(y, 0)), + outer(x, y)) + y = array(['abc', 1, 'def', 2, 3], object) + y[2] = masked + t = take(y, [0, 3, 4]) + assert_(t[0] == 'abc') + assert_(t[1] == 2) + assert_(t[2] == 3) + + def test_imag_real(self): + # Check complex + xx = array([1 + 10j, 20 + 2j], mask=[1, 0]) + assert_equal(xx.imag, [10, 2]) + assert_equal(xx.imag.filled(), [1e+20, 2]) + assert_equal(xx.imag.dtype, xx._data.imag.dtype) + assert_equal(xx.real, [1, 20]) + assert_equal(xx.real.filled(), [1e+20, 20]) + assert_equal(xx.real.dtype, xx._data.real.dtype) + + def test_methods_with_output(self): + xm = array(np.random.uniform(0, 10, 12)).reshape(3, 4) + xm[:, 0] = xm[0] = xm[-1, -1] = masked + + funclist = ('sum', 'prod', 'var', 'std', 'max', 'min', 'ptp', 'mean',) + + for funcname in funclist: + npfunc = getattr(np, funcname) + xmmeth = getattr(xm, funcname) + # A ndarray as explicit input + output = np.empty(4, dtype=float) + output.fill(-9999) + result = npfunc(xm, axis=0, out=output) + # ... the result should be the given output + assert_(result is output) + assert_equal(result, xmmeth(axis=0, out=output)) + + output = empty(4, dtype=int) + result = xmmeth(axis=0, out=output) + assert_(result is output) + assert_(output[0] is masked) + + def test_eq_on_structured(self): + # Test the equality of structured arrays + ndtype = [('A', int), ('B', int)] + a = array([(1, 1), (2, 2)], mask=[(0, 1), (0, 0)], dtype=ndtype) + + test = (a == a) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [False, False]) + assert_(test.fill_value == True) + + test = (a == a[0]) + assert_equal(test.data, [True, False]) + assert_equal(test.mask, [False, False]) + assert_(test.fill_value == True) + + b = array([(1, 1), (2, 2)], mask=[(1, 0), (0, 0)], dtype=ndtype) + test = (a == b) + assert_equal(test.data, [False, True]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + test = (a[0] == b) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + b = array([(1, 1), (2, 2)], mask=[(0, 1), (1, 0)], dtype=ndtype) + test = (a == b) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [False, False]) + assert_(test.fill_value == True) + + # complicated dtype, 2-dimensional array. + ndtype = [('A', int), ('B', [('BA', int), ('BB', int)])] + a = array([[(1, (1, 1)), (2, (2, 2))], + [(3, (3, 3)), (4, (4, 4))]], + mask=[[(0, (1, 0)), (0, (0, 1))], + [(1, (0, 0)), (1, (1, 1))]], dtype=ndtype) + test = (a[0, 0] == a) + assert_equal(test.data, [[True, False], [False, False]]) + assert_equal(test.mask, [[False, False], [False, True]]) + assert_(test.fill_value == True) + + def test_ne_on_structured(self): + # Test the equality of structured arrays + ndtype = [('A', int), ('B', int)] + a = array([(1, 1), (2, 2)], mask=[(0, 1), (0, 0)], dtype=ndtype) + + test = (a != a) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [False, False]) + assert_(test.fill_value == True) + + test = (a != a[0]) + assert_equal(test.data, [False, True]) + assert_equal(test.mask, [False, False]) + assert_(test.fill_value == True) + + b = array([(1, 1), (2, 2)], mask=[(1, 0), (0, 0)], dtype=ndtype) + test = (a != b) + assert_equal(test.data, [True, False]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + test = (a[0] != b) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + b = array([(1, 1), (2, 2)], mask=[(0, 1), (1, 0)], dtype=ndtype) + test = (a != b) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [False, False]) + assert_(test.fill_value == True) + + # complicated dtype, 2-dimensional array. + ndtype = [('A', int), ('B', [('BA', int), ('BB', int)])] + a = array([[(1, (1, 1)), (2, (2, 2))], + [(3, (3, 3)), (4, (4, 4))]], + mask=[[(0, (1, 0)), (0, (0, 1))], + [(1, (0, 0)), (1, (1, 1))]], dtype=ndtype) + test = (a[0, 0] != a) + assert_equal(test.data, [[False, True], [True, True]]) + assert_equal(test.mask, [[False, False], [False, True]]) + assert_(test.fill_value == True) + + def test_eq_ne_structured_with_non_masked(self): + a = array([(1, 1), (2, 2), (3, 4)], + mask=[(0, 1), (0, 0), (1, 1)], dtype='i4,i4') + eq = a == a.data + ne = a.data != a + # Test the obvious. + assert_(np.all(eq)) + assert_(not np.any(ne)) + # Expect the mask set only for items with all fields masked. + expected_mask = a.mask == np.ones((), a.mask.dtype) + assert_array_equal(eq.mask, expected_mask) + assert_array_equal(ne.mask, expected_mask) + # The masked element will indicated not equal, because the + # masks did not match. + assert_equal(eq.data, [True, True, False]) + assert_array_equal(eq.data, ~ne.data) + + def test_eq_ne_structured_extra(self): + # ensure simple examples are symmetric and make sense. + # from https://github.com/numpy/numpy/pull/8590#discussion_r101126465 + dt = np.dtype('i4,i4') + for m1 in (mvoid((1, 2), mask=(0, 0), dtype=dt), + mvoid((1, 2), mask=(0, 1), dtype=dt), + mvoid((1, 2), mask=(1, 0), dtype=dt), + mvoid((1, 2), mask=(1, 1), dtype=dt)): + ma1 = m1.view(MaskedArray) + r1 = ma1.view('2i4') + for m2 in (np.array((1, 1), dtype=dt), + mvoid((1, 1), dtype=dt), + mvoid((1, 0), mask=(0, 1), dtype=dt), + mvoid((3, 2), mask=(0, 1), dtype=dt)): + ma2 = m2.view(MaskedArray) + r2 = ma2.view('2i4') + eq_expected = (r1 == r2).all() + assert_equal(m1 == m2, eq_expected) + assert_equal(m2 == m1, eq_expected) + assert_equal(ma1 == m2, eq_expected) + assert_equal(m1 == ma2, eq_expected) + assert_equal(ma1 == ma2, eq_expected) + # Also check it is the same if we do it element by element. + el_by_el = [m1[name] == m2[name] for name in dt.names] + assert_equal(array(el_by_el, dtype=bool).all(), eq_expected) + ne_expected = (r1 != r2).any() + assert_equal(m1 != m2, ne_expected) + assert_equal(m2 != m1, ne_expected) + assert_equal(ma1 != m2, ne_expected) + assert_equal(m1 != ma2, ne_expected) + assert_equal(ma1 != ma2, ne_expected) + el_by_el = [m1[name] != m2[name] for name in dt.names] + assert_equal(array(el_by_el, dtype=bool).any(), ne_expected) + + @pytest.mark.parametrize('dt', ['S', 'U']) + @pytest.mark.parametrize('fill', [None, 'A']) + def test_eq_for_strings(self, dt, fill): + # Test the equality of structured arrays + a = array(['a', 'b'], dtype=dt, mask=[0, 1], fill_value=fill) + + test = (a == a) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + test = (a == a[0]) + assert_equal(test.data, [True, False]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + b = array(['a', 'b'], dtype=dt, mask=[1, 0], fill_value=fill) + test = (a == b) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [True, True]) + assert_(test.fill_value == True) + + test = (a[0] == b) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + test = (b == a[0]) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + @pytest.mark.parametrize('dt', ['S', 'U']) + @pytest.mark.parametrize('fill', [None, 'A']) + def test_ne_for_strings(self, dt, fill): + # Test the equality of structured arrays + a = array(['a', 'b'], dtype=dt, mask=[0, 1], fill_value=fill) + + test = (a != a) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + test = (a != a[0]) + assert_equal(test.data, [False, True]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + b = array(['a', 'b'], dtype=dt, mask=[1, 0], fill_value=fill) + test = (a != b) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [True, True]) + assert_(test.fill_value == True) + + test = (a[0] != b) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + test = (b != a[0]) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + @pytest.mark.parametrize('dt1', num_dts, ids=num_ids) + @pytest.mark.parametrize('dt2', num_dts, ids=num_ids) + @pytest.mark.parametrize('fill', [None, 1]) + def test_eq_for_numeric(self, dt1, dt2, fill): + # Test the equality of structured arrays + a = array([0, 1], dtype=dt1, mask=[0, 1], fill_value=fill) + + test = (a == a) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + test = (a == a[0]) + assert_equal(test.data, [True, False]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + b = array([0, 1], dtype=dt2, mask=[1, 0], fill_value=fill) + test = (a == b) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [True, True]) + assert_(test.fill_value == True) + + test = (a[0] == b) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + test = (b == a[0]) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + @pytest.mark.parametrize("op", [operator.eq, operator.lt]) + def test_eq_broadcast_with_unmasked(self, op): + a = array([0, 1], mask=[0, 1]) + b = np.arange(10).reshape(5, 2) + result = op(a, b) + assert_(result.mask.shape == b.shape) + assert_equal(result.mask, np.zeros(b.shape, bool) | a.mask) + + @pytest.mark.parametrize("op", [operator.eq, operator.gt]) + def test_comp_no_mask_not_broadcast(self, op): + # Regression test for failing doctest in MaskedArray.nonzero + # after gh-24556. + a = array([[1, 2, 3], [4, 5, 6], [7, 8, 9]]) + result = op(a, 3) + assert_(not result.mask.shape) + assert_(result.mask is nomask) + + @pytest.mark.parametrize('dt1', num_dts, ids=num_ids) + @pytest.mark.parametrize('dt2', num_dts, ids=num_ids) + @pytest.mark.parametrize('fill', [None, 1]) + def test_ne_for_numeric(self, dt1, dt2, fill): + # Test the equality of structured arrays + a = array([0, 1], dtype=dt1, mask=[0, 1], fill_value=fill) + + test = (a != a) + assert_equal(test.data, [False, False]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + test = (a != a[0]) + assert_equal(test.data, [False, True]) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + b = array([0, 1], dtype=dt2, mask=[1, 0], fill_value=fill) + test = (a != b) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [True, True]) + assert_(test.fill_value == True) + + test = (a[0] != b) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + test = (b != a[0]) + assert_equal(test.data, [True, True]) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + @pytest.mark.parametrize('dt1', num_dts, ids=num_ids) + @pytest.mark.parametrize('dt2', num_dts, ids=num_ids) + @pytest.mark.parametrize('fill', [None, 1]) + @pytest.mark.parametrize('op', + [operator.le, operator.lt, operator.ge, operator.gt]) + def test_comparisons_for_numeric(self, op, dt1, dt2, fill): + # Test the equality of structured arrays + a = array([0, 1], dtype=dt1, mask=[0, 1], fill_value=fill) + + test = op(a, a) + assert_equal(test.data, op(a._data, a._data)) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + test = op(a, a[0]) + assert_equal(test.data, op(a._data, a._data[0])) + assert_equal(test.mask, [False, True]) + assert_(test.fill_value == True) + + b = array([0, 1], dtype=dt2, mask=[1, 0], fill_value=fill) + test = op(a, b) + assert_equal(test.data, op(a._data, b._data)) + assert_equal(test.mask, [True, True]) + assert_(test.fill_value == True) + + test = op(a[0], b) + assert_equal(test.data, op(a._data[0], b._data)) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + test = op(b, a[0]) + assert_equal(test.data, op(b._data, a._data[0])) + assert_equal(test.mask, [True, False]) + assert_(test.fill_value == True) + + @pytest.mark.parametrize('op', + [operator.le, operator.lt, operator.ge, operator.gt]) + @pytest.mark.parametrize('fill', [None, "N/A"]) + def test_comparisons_strings(self, op, fill): + # See gh-21770, mask propagation is broken for strings (and some other + # cases) so we explicitly test strings here. + # In principle only == and != may need special handling... + ma1 = masked_array(["a", "b", "cde"], mask=[0, 1, 0], fill_value=fill) + ma2 = masked_array(["cde", "b", "a"], mask=[0, 1, 0], fill_value=fill) + assert_equal(op(ma1, ma2)._data, op(ma1._data, ma2._data)) + + def test_eq_with_None(self): + # Really, comparisons with None should not be done, but check them + # anyway. Note that pep8 will flag these tests. + # Deprecation is in place for arrays, and when it happens this + # test will fail (and have to be changed accordingly). + + # With partial mask + with suppress_warnings() as sup: + sup.filter(FutureWarning, "Comparison to `None`") + a = array([None, 1], mask=[0, 1]) + assert_equal(a == None, array([True, False], mask=[0, 1])) + assert_equal(a.data == None, [True, False]) + assert_equal(a != None, array([False, True], mask=[0, 1])) + # With nomask + a = array([None, 1], mask=False) + assert_equal(a == None, [True, False]) + assert_equal(a != None, [False, True]) + # With complete mask + a = array([None, 2], mask=True) + assert_equal(a == None, array([False, True], mask=True)) + assert_equal(a != None, array([True, False], mask=True)) + # Fully masked, even comparison to None should return "masked" + a = masked + assert_equal(a == None, masked) + + def test_eq_with_scalar(self): + a = array(1) + assert_equal(a == 1, True) + assert_equal(a == 0, False) + assert_equal(a != 1, False) + assert_equal(a != 0, True) + b = array(1, mask=True) + assert_equal(b == 0, masked) + assert_equal(b == 1, masked) + assert_equal(b != 0, masked) + assert_equal(b != 1, masked) + + def test_eq_different_dimensions(self): + m1 = array([1, 1], mask=[0, 1]) + # test comparison with both masked and regular arrays. + for m2 in (array([[0, 1], [1, 2]]), + np.array([[0, 1], [1, 2]])): + test = (m1 == m2) + assert_equal(test.data, [[False, False], + [True, False]]) + assert_equal(test.mask, [[False, True], + [False, True]]) + + def test_numpyarithmetic(self): + # Check that the mask is not back-propagated when using numpy functions + a = masked_array([-1, 0, 1, 2, 3], mask=[0, 0, 0, 0, 1]) + control = masked_array([np.nan, np.nan, 0, np.log(2), -1], + mask=[1, 1, 0, 0, 1]) + + test = log(a) + assert_equal(test, control) + assert_equal(test.mask, control.mask) + assert_equal(a.mask, [0, 0, 0, 0, 1]) + + test = np.log(a) + assert_equal(test, control) + assert_equal(test.mask, control.mask) + assert_equal(a.mask, [0, 0, 0, 0, 1]) + + +class TestMaskedArrayAttributes: + + def test_keepmask(self): + # Tests the keep mask flag + x = masked_array([1, 2, 3], mask=[1, 0, 0]) + mx = masked_array(x) + assert_equal(mx.mask, x.mask) + mx = masked_array(x, mask=[0, 1, 0], keep_mask=False) + assert_equal(mx.mask, [0, 1, 0]) + mx = masked_array(x, mask=[0, 1, 0], keep_mask=True) + assert_equal(mx.mask, [1, 1, 0]) + # We default to true + mx = masked_array(x, mask=[0, 1, 0]) + assert_equal(mx.mask, [1, 1, 0]) + + def test_hardmask(self): + # Test hard_mask + d = arange(5) + n = [0, 0, 0, 1, 1] + m = make_mask(n) + xh = array(d, mask=m, hard_mask=True) + # We need to copy, to avoid updating d in xh ! + xs = array(d, mask=m, hard_mask=False, copy=True) + xh[[1, 4]] = [10, 40] + xs[[1, 4]] = [10, 40] + assert_equal(xh._data, [0, 10, 2, 3, 4]) + assert_equal(xs._data, [0, 10, 2, 3, 40]) + assert_equal(xs.mask, [0, 0, 0, 1, 0]) + assert_(xh._hardmask) + assert_(not xs._hardmask) + xh[1:4] = [10, 20, 30] + xs[1:4] = [10, 20, 30] + assert_equal(xh._data, [0, 10, 20, 3, 4]) + assert_equal(xs._data, [0, 10, 20, 30, 40]) + assert_equal(xs.mask, nomask) + xh[0] = masked + xs[0] = masked + assert_equal(xh.mask, [1, 0, 0, 1, 1]) + assert_equal(xs.mask, [1, 0, 0, 0, 0]) + xh[:] = 1 + xs[:] = 1 + assert_equal(xh._data, [0, 1, 1, 3, 4]) + assert_equal(xs._data, [1, 1, 1, 1, 1]) + assert_equal(xh.mask, [1, 0, 0, 1, 1]) + assert_equal(xs.mask, nomask) + # Switch to soft mask + xh.soften_mask() + xh[:] = arange(5) + assert_equal(xh._data, [0, 1, 2, 3, 4]) + assert_equal(xh.mask, nomask) + # Switch back to hard mask + xh.harden_mask() + xh[xh < 3] = masked + assert_equal(xh._data, [0, 1, 2, 3, 4]) + assert_equal(xh._mask, [1, 1, 1, 0, 0]) + xh[filled(xh > 1, False)] = 5 + assert_equal(xh._data, [0, 1, 2, 5, 5]) + assert_equal(xh._mask, [1, 1, 1, 0, 0]) + + xh = array([[1, 2], [3, 4]], mask=[[1, 0], [0, 0]], hard_mask=True) + xh[0] = 0 + assert_equal(xh._data, [[1, 0], [3, 4]]) + assert_equal(xh._mask, [[1, 0], [0, 0]]) + xh[-1, -1] = 5 + assert_equal(xh._data, [[1, 0], [3, 5]]) + assert_equal(xh._mask, [[1, 0], [0, 0]]) + xh[filled(xh < 5, False)] = 2 + assert_equal(xh._data, [[1, 2], [2, 5]]) + assert_equal(xh._mask, [[1, 0], [0, 0]]) + + def test_hardmask_again(self): + # Another test of hardmask + d = arange(5) + n = [0, 0, 0, 1, 1] + m = make_mask(n) + xh = array(d, mask=m, hard_mask=True) + xh[4:5] = 999 + xh[0:1] = 999 + assert_equal(xh._data, [999, 1, 2, 3, 4]) + + def test_hardmask_oncemore_yay(self): + # OK, yet another test of hardmask + # Make sure that harden_mask/soften_mask//unshare_mask returns self + a = array([1, 2, 3], mask=[1, 0, 0]) + b = a.harden_mask() + assert_equal(a, b) + b[0] = 0 + assert_equal(a, b) + assert_equal(b, array([1, 2, 3], mask=[1, 0, 0])) + a = b.soften_mask() + a[0] = 0 + assert_equal(a, b) + assert_equal(b, array([0, 2, 3], mask=[0, 0, 0])) + + def test_smallmask(self): + # Checks the behaviour of _smallmask + a = arange(10) + a[1] = masked + a[1] = 1 + assert_equal(a._mask, nomask) + a = arange(10) + a._smallmask = False + a[1] = masked + a[1] = 1 + assert_equal(a._mask, zeros(10)) + + def test_shrink_mask(self): + # Tests .shrink_mask() + a = array([1, 2, 3], mask=[0, 0, 0]) + b = a.shrink_mask() + assert_equal(a, b) + assert_equal(a.mask, nomask) + + # Mask cannot be shrunk on structured types, so is a no-op + a = np.ma.array([(1, 2.0)], [('a', int), ('b', float)]) + b = a.copy() + a.shrink_mask() + assert_equal(a.mask, b.mask) + + def test_flat(self): + # Test that flat can return all types of items [#4585, #4615] + # test 2-D record array + # ... on structured array w/ masked records + x = array([[(1, 1.1, 'one'), (2, 2.2, 'two'), (3, 3.3, 'thr')], + [(4, 4.4, 'fou'), (5, 5.5, 'fiv'), (6, 6.6, 'six')]], + dtype=[('a', int), ('b', float), ('c', '|S8')]) + x['a'][0, 1] = masked + x['b'][1, 0] = masked + x['c'][0, 2] = masked + x[-1, -1] = masked + xflat = x.flat + assert_equal(xflat[0], x[0, 0]) + assert_equal(xflat[1], x[0, 1]) + assert_equal(xflat[2], x[0, 2]) + assert_equal(xflat[:3], x[0]) + assert_equal(xflat[3], x[1, 0]) + assert_equal(xflat[4], x[1, 1]) + assert_equal(xflat[5], x[1, 2]) + assert_equal(xflat[3:], x[1]) + assert_equal(xflat[-1], x[-1, -1]) + i = 0 + j = 0 + for xf in xflat: + assert_equal(xf, x[j, i]) + i += 1 + if i >= x.shape[-1]: + i = 0 + j += 1 + + def test_assign_dtype(self): + # check that the mask's dtype is updated when dtype is changed + a = np.zeros(4, dtype='f4,i4') + + m = np.ma.array(a) + m.dtype = np.dtype('f4') + repr(m) # raises? + assert_equal(m.dtype, np.dtype('f4')) + + # check that dtype changes that change shape of mask too much + # are not allowed + def assign(): + m = np.ma.array(a) + m.dtype = np.dtype('f8') + assert_raises(ValueError, assign) + + b = a.view(dtype='f4', type=np.ma.MaskedArray) # raises? + assert_equal(b.dtype, np.dtype('f4')) + + # check that nomask is preserved + a = np.zeros(4, dtype='f4') + m = np.ma.array(a) + m.dtype = np.dtype('f4,i4') + assert_equal(m.dtype, np.dtype('f4,i4')) + assert_equal(m._mask, np.ma.nomask) + + +class TestFillingValues: + + def test_check_on_scalar(self): + # Test _check_fill_value set to valid and invalid values + _check_fill_value = np.ma.core._check_fill_value + + fval = _check_fill_value(0, int) + assert_equal(fval, 0) + fval = _check_fill_value(None, int) + assert_equal(fval, default_fill_value(0)) + + fval = _check_fill_value(0, "|S3") + assert_equal(fval, b"0") + fval = _check_fill_value(None, "|S3") + assert_equal(fval, default_fill_value(b"camelot!")) + assert_raises(TypeError, _check_fill_value, 1e+20, int) + assert_raises(TypeError, _check_fill_value, 'stuff', int) + + def test_check_on_fields(self): + # Tests _check_fill_value with records + _check_fill_value = np.ma.core._check_fill_value + ndtype = [('a', int), ('b', float), ('c', "|S3")] + # A check on a list should return a single record + fval = _check_fill_value([-999, -12345678.9, "???"], ndtype) + assert_(isinstance(fval, ndarray)) + assert_equal(fval.item(), [-999, -12345678.9, b"???"]) + # A check on None should output the defaults + fval = _check_fill_value(None, ndtype) + assert_(isinstance(fval, ndarray)) + assert_equal(fval.item(), [default_fill_value(0), + default_fill_value(0.), + asbytes(default_fill_value("0"))]) + #.....Using a structured type as fill_value should work + fill_val = np.array((-999, -12345678.9, "???"), dtype=ndtype) + fval = _check_fill_value(fill_val, ndtype) + assert_(isinstance(fval, ndarray)) + assert_equal(fval.item(), [-999, -12345678.9, b"???"]) + + #.....Using a flexible type w/ a different type shouldn't matter + # BEHAVIOR in 1.5 and earlier, and 1.13 and later: match structured + # types by position + fill_val = np.array((-999, -12345678.9, "???"), + dtype=[("A", int), ("B", float), ("C", "|S3")]) + fval = _check_fill_value(fill_val, ndtype) + assert_(isinstance(fval, ndarray)) + assert_equal(fval.item(), [-999, -12345678.9, b"???"]) + + #.....Using an object-array shouldn't matter either + fill_val = np.ndarray(shape=(1,), dtype=object) + fill_val[0] = (-999, -12345678.9, b"???") + fval = _check_fill_value(fill_val, object) + assert_(isinstance(fval, ndarray)) + assert_equal(fval.item(), [-999, -12345678.9, b"???"]) + # NOTE: This test was never run properly as "fill_value" rather than + # "fill_val" was assigned. Written properly, it fails. + #fill_val = np.array((-999, -12345678.9, "???")) + #fval = _check_fill_value(fill_val, ndtype) + #assert_(isinstance(fval, ndarray)) + #assert_equal(fval.item(), [-999, -12345678.9, b"???"]) + #.....One-field-only flexible type should work as well + ndtype = [("a", int)] + fval = _check_fill_value(-999999999, ndtype) + assert_(isinstance(fval, ndarray)) + assert_equal(fval.item(), (-999999999,)) + + def test_fillvalue_conversion(self): + # Tests the behavior of fill_value during conversion + # We had a tailored comment to make sure special attributes are + # properly dealt with + a = array([b'3', b'4', b'5']) + a._optinfo.update({'comment':"updated!"}) + + b = array(a, dtype=int) + assert_equal(b._data, [3, 4, 5]) + assert_equal(b.fill_value, default_fill_value(0)) + + b = array(a, dtype=float) + assert_equal(b._data, [3, 4, 5]) + assert_equal(b.fill_value, default_fill_value(0.)) + + b = a.astype(int) + assert_equal(b._data, [3, 4, 5]) + assert_equal(b.fill_value, default_fill_value(0)) + assert_equal(b._optinfo['comment'], "updated!") + + b = a.astype([('a', '|S3')]) + assert_equal(b['a']._data, a._data) + assert_equal(b['a'].fill_value, a.fill_value) + + def test_default_fill_value(self): + # check all calling conventions + f1 = default_fill_value(1.) + f2 = default_fill_value(np.array(1.)) + f3 = default_fill_value(np.array(1.).dtype) + assert_equal(f1, f2) + assert_equal(f1, f3) + + def test_default_fill_value_structured(self): + fields = array([(1, 1, 1)], + dtype=[('i', int), ('s', '|S8'), ('f', float)]) + + f1 = default_fill_value(fields) + f2 = default_fill_value(fields.dtype) + expected = np.array((default_fill_value(0), + default_fill_value('0'), + default_fill_value(0.)), dtype=fields.dtype) + assert_equal(f1, expected) + assert_equal(f2, expected) + + def test_default_fill_value_void(self): + dt = np.dtype([('v', 'V7')]) + f = default_fill_value(dt) + assert_equal(f['v'], np.array(default_fill_value(dt['v']), dt['v'])) + + def test_fillvalue(self): + # Yet more fun with the fill_value + data = masked_array([1, 2, 3], fill_value=-999) + series = data[[0, 2, 1]] + assert_equal(series._fill_value, data._fill_value) + + mtype = [('f', float), ('s', '|S3')] + x = array([(1, 'a'), (2, 'b'), (pi, 'pi')], dtype=mtype) + x.fill_value = 999 + assert_equal(x.fill_value.item(), [999., b'999']) + assert_equal(x['f'].fill_value, 999) + assert_equal(x['s'].fill_value, b'999') + + x.fill_value = (9, '???') + assert_equal(x.fill_value.item(), (9, b'???')) + assert_equal(x['f'].fill_value, 9) + assert_equal(x['s'].fill_value, b'???') + + x = array([1, 2, 3.1]) + x.fill_value = 999 + assert_equal(np.asarray(x.fill_value).dtype, float) + assert_equal(x.fill_value, 999.) + assert_equal(x._fill_value, np.array(999.)) + + def test_subarray_fillvalue(self): + # gh-10483 test multi-field index fill value + fields = array([(1, 1, 1)], + dtype=[('i', int), ('s', '|S8'), ('f', float)]) + with suppress_warnings() as sup: + sup.filter(FutureWarning, "Numpy has detected") + subfields = fields[['i', 'f']] + assert_equal(tuple(subfields.fill_value), (999999, 1.e+20)) + # test comparison does not raise: + subfields[1:] == subfields[:-1] + + def test_fillvalue_exotic_dtype(self): + # Tests yet more exotic flexible dtypes + _check_fill_value = np.ma.core._check_fill_value + ndtype = [('i', int), ('s', '|S8'), ('f', float)] + control = np.array((default_fill_value(0), + default_fill_value('0'), + default_fill_value(0.),), + dtype=ndtype) + assert_equal(_check_fill_value(None, ndtype), control) + # The shape shouldn't matter + ndtype = [('f0', float, (2, 2))] + control = np.array((default_fill_value(0.),), + dtype=[('f0', float)]).astype(ndtype) + assert_equal(_check_fill_value(None, ndtype), control) + control = np.array((0,), dtype=[('f0', float)]).astype(ndtype) + assert_equal(_check_fill_value(0, ndtype), control) + + ndtype = np.dtype("int, (2,3)float, float") + control = np.array((default_fill_value(0), + default_fill_value(0.), + default_fill_value(0.),), + dtype="int, float, float").astype(ndtype) + test = _check_fill_value(None, ndtype) + assert_equal(test, control) + control = np.array((0, 0, 0), dtype="int, float, float").astype(ndtype) + assert_equal(_check_fill_value(0, ndtype), control) + # but when indexing, fill value should become scalar not tuple + # See issue #6723 + M = masked_array(control) + assert_equal(M["f1"].fill_value.ndim, 0) + + def test_fillvalue_datetime_timedelta(self): + # Test default fillvalue for datetime64 and timedelta64 types. + # See issue #4476, this would return '?' which would cause errors + # elsewhere + + for timecode in ("as", "fs", "ps", "ns", "us", "ms", "s", "m", + "h", "D", "W", "M", "Y"): + control = numpy.datetime64("NaT", timecode) + test = default_fill_value(numpy.dtype(" 0 + + # test different unary domains + sqrt(m) + log(m) + tan(m) + arcsin(m) + arccos(m) + arccosh(m) + + # test binary domains + divide(m, 2) + + # also check that allclose uses ma ufuncs, to avoid warning + allclose(m, 0.5) + +class TestMaskedArrayInPlaceArithmetic: + # Test MaskedArray Arithmetic + + def setup_method(self): + x = arange(10) + y = arange(10) + xm = arange(10) + xm[2] = masked + self.intdata = (x, y, xm) + self.floatdata = (x.astype(float), y.astype(float), xm.astype(float)) + self.othertypes = np.typecodes['AllInteger'] + np.typecodes['AllFloat'] + self.othertypes = [np.dtype(_).type for _ in self.othertypes] + self.uint8data = ( + x.astype(np.uint8), + y.astype(np.uint8), + xm.astype(np.uint8) + ) + + def test_inplace_addition_scalar(self): + # Test of inplace additions + (x, y, xm) = self.intdata + xm[2] = masked + x += 1 + assert_equal(x, y + 1) + xm += 1 + assert_equal(xm, y + 1) + + (x, _, xm) = self.floatdata + id1 = x.data.ctypes.data + x += 1. + assert_(id1 == x.data.ctypes.data) + assert_equal(x, y + 1.) + + def test_inplace_addition_array(self): + # Test of inplace additions + (x, y, xm) = self.intdata + m = xm.mask + a = arange(10, dtype=np.int16) + a[-1] = masked + x += a + xm += a + assert_equal(x, y + a) + assert_equal(xm, y + a) + assert_equal(xm.mask, mask_or(m, a.mask)) + + def test_inplace_subtraction_scalar(self): + # Test of inplace subtractions + (x, y, xm) = self.intdata + x -= 1 + assert_equal(x, y - 1) + xm -= 1 + assert_equal(xm, y - 1) + + def test_inplace_subtraction_array(self): + # Test of inplace subtractions + (x, y, xm) = self.floatdata + m = xm.mask + a = arange(10, dtype=float) + a[-1] = masked + x -= a + xm -= a + assert_equal(x, y - a) + assert_equal(xm, y - a) + assert_equal(xm.mask, mask_or(m, a.mask)) + + def test_inplace_multiplication_scalar(self): + # Test of inplace multiplication + (x, y, xm) = self.floatdata + x *= 2.0 + assert_equal(x, y * 2) + xm *= 2.0 + assert_equal(xm, y * 2) + + def test_inplace_multiplication_array(self): + # Test of inplace multiplication + (x, y, xm) = self.floatdata + m = xm.mask + a = arange(10, dtype=float) + a[-1] = masked + x *= a + xm *= a + assert_equal(x, y * a) + assert_equal(xm, y * a) + assert_equal(xm.mask, mask_or(m, a.mask)) + + def test_inplace_division_scalar_int(self): + # Test of inplace division + (x, y, xm) = self.intdata + x = arange(10) * 2 + xm = arange(10) * 2 + xm[2] = masked + x //= 2 + assert_equal(x, y) + xm //= 2 + assert_equal(xm, y) + + def test_inplace_division_scalar_float(self): + # Test of inplace division + (x, y, xm) = self.floatdata + x /= 2.0 + assert_equal(x, y / 2.0) + xm /= arange(10) + assert_equal(xm, ones((10,))) + + def test_inplace_division_array_float(self): + # Test of inplace division + (x, y, xm) = self.floatdata + m = xm.mask + a = arange(10, dtype=float) + a[-1] = masked + x /= a + xm /= a + assert_equal(x, y / a) + assert_equal(xm, y / a) + assert_equal(xm.mask, mask_or(mask_or(m, a.mask), (a == 0))) + + def test_inplace_division_misc(self): + + x = [1., 1., 1., -2., pi / 2., 4., 5., -10., 10., 1., 2., 3.] + y = [5., 0., 3., 2., -1., -4., 0., -10., 10., 1., 0., 3.] + m1 = [1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0] + m2 = [0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 1] + xm = masked_array(x, mask=m1) + ym = masked_array(y, mask=m2) + + z = xm / ym + assert_equal(z._mask, [1, 1, 1, 0, 0, 1, 1, 0, 0, 0, 1, 1]) + assert_equal(z._data, + [1., 1., 1., -1., -pi / 2., 4., 5., 1., 1., 1., 2., 3.]) + + xm = xm.copy() + xm /= ym + assert_equal(xm._mask, [1, 1, 1, 0, 0, 1, 1, 0, 0, 0, 1, 1]) + assert_equal(z._data, + [1., 1., 1., -1., -pi / 2., 4., 5., 1., 1., 1., 2., 3.]) + + def test_datafriendly_add(self): + # Test keeping data w/ (inplace) addition + x = array([1, 2, 3], mask=[0, 0, 1]) + # Test add w/ scalar + xx = x + 1 + assert_equal(xx.data, [2, 3, 3]) + assert_equal(xx.mask, [0, 0, 1]) + # Test iadd w/ scalar + x += 1 + assert_equal(x.data, [2, 3, 3]) + assert_equal(x.mask, [0, 0, 1]) + # Test add w/ array + x = array([1, 2, 3], mask=[0, 0, 1]) + xx = x + array([1, 2, 3], mask=[1, 0, 0]) + assert_equal(xx.data, [1, 4, 3]) + assert_equal(xx.mask, [1, 0, 1]) + # Test iadd w/ array + x = array([1, 2, 3], mask=[0, 0, 1]) + x += array([1, 2, 3], mask=[1, 0, 0]) + assert_equal(x.data, [1, 4, 3]) + assert_equal(x.mask, [1, 0, 1]) + + def test_datafriendly_sub(self): + # Test keeping data w/ (inplace) subtraction + # Test sub w/ scalar + x = array([1, 2, 3], mask=[0, 0, 1]) + xx = x - 1 + assert_equal(xx.data, [0, 1, 3]) + assert_equal(xx.mask, [0, 0, 1]) + # Test isub w/ scalar + x = array([1, 2, 3], mask=[0, 0, 1]) + x -= 1 + assert_equal(x.data, [0, 1, 3]) + assert_equal(x.mask, [0, 0, 1]) + # Test sub w/ array + x = array([1, 2, 3], mask=[0, 0, 1]) + xx = x - array([1, 2, 3], mask=[1, 0, 0]) + assert_equal(xx.data, [1, 0, 3]) + assert_equal(xx.mask, [1, 0, 1]) + # Test isub w/ array + x = array([1, 2, 3], mask=[0, 0, 1]) + x -= array([1, 2, 3], mask=[1, 0, 0]) + assert_equal(x.data, [1, 0, 3]) + assert_equal(x.mask, [1, 0, 1]) + + def test_datafriendly_mul(self): + # Test keeping data w/ (inplace) multiplication + # Test mul w/ scalar + x = array([1, 2, 3], mask=[0, 0, 1]) + xx = x * 2 + assert_equal(xx.data, [2, 4, 3]) + assert_equal(xx.mask, [0, 0, 1]) + # Test imul w/ scalar + x = array([1, 2, 3], mask=[0, 0, 1]) + x *= 2 + assert_equal(x.data, [2, 4, 3]) + assert_equal(x.mask, [0, 0, 1]) + # Test mul w/ array + x = array([1, 2, 3], mask=[0, 0, 1]) + xx = x * array([10, 20, 30], mask=[1, 0, 0]) + assert_equal(xx.data, [1, 40, 3]) + assert_equal(xx.mask, [1, 0, 1]) + # Test imul w/ array + x = array([1, 2, 3], mask=[0, 0, 1]) + x *= array([10, 20, 30], mask=[1, 0, 0]) + assert_equal(x.data, [1, 40, 3]) + assert_equal(x.mask, [1, 0, 1]) + + def test_datafriendly_div(self): + # Test keeping data w/ (inplace) division + # Test div on scalar + x = array([1, 2, 3], mask=[0, 0, 1]) + xx = x / 2. + assert_equal(xx.data, [1 / 2., 2 / 2., 3]) + assert_equal(xx.mask, [0, 0, 1]) + # Test idiv on scalar + x = array([1., 2., 3.], mask=[0, 0, 1]) + x /= 2. + assert_equal(x.data, [1 / 2., 2 / 2., 3]) + assert_equal(x.mask, [0, 0, 1]) + # Test div on array + x = array([1., 2., 3.], mask=[0, 0, 1]) + xx = x / array([10., 20., 30.], mask=[1, 0, 0]) + assert_equal(xx.data, [1., 2. / 20., 3.]) + assert_equal(xx.mask, [1, 0, 1]) + # Test idiv on array + x = array([1., 2., 3.], mask=[0, 0, 1]) + x /= array([10., 20., 30.], mask=[1, 0, 0]) + assert_equal(x.data, [1., 2 / 20., 3.]) + assert_equal(x.mask, [1, 0, 1]) + + def test_datafriendly_pow(self): + # Test keeping data w/ (inplace) power + # Test pow on scalar + x = array([1., 2., 3.], mask=[0, 0, 1]) + xx = x ** 2.5 + assert_equal(xx.data, [1., 2. ** 2.5, 3.]) + assert_equal(xx.mask, [0, 0, 1]) + # Test ipow on scalar + x **= 2.5 + assert_equal(x.data, [1., 2. ** 2.5, 3]) + assert_equal(x.mask, [0, 0, 1]) + + def test_datafriendly_add_arrays(self): + a = array([[1, 1], [3, 3]]) + b = array([1, 1], mask=[0, 0]) + a += b + assert_equal(a, [[2, 2], [4, 4]]) + if a.mask is not nomask: + assert_equal(a.mask, [[0, 0], [0, 0]]) + + a = array([[1, 1], [3, 3]]) + b = array([1, 1], mask=[0, 1]) + a += b + assert_equal(a, [[2, 2], [4, 4]]) + assert_equal(a.mask, [[0, 1], [0, 1]]) + + def test_datafriendly_sub_arrays(self): + a = array([[1, 1], [3, 3]]) + b = array([1, 1], mask=[0, 0]) + a -= b + assert_equal(a, [[0, 0], [2, 2]]) + if a.mask is not nomask: + assert_equal(a.mask, [[0, 0], [0, 0]]) + + a = array([[1, 1], [3, 3]]) + b = array([1, 1], mask=[0, 1]) + a -= b + assert_equal(a, [[0, 0], [2, 2]]) + assert_equal(a.mask, [[0, 1], [0, 1]]) + + def test_datafriendly_mul_arrays(self): + a = array([[1, 1], [3, 3]]) + b = array([1, 1], mask=[0, 0]) + a *= b + assert_equal(a, [[1, 1], [3, 3]]) + if a.mask is not nomask: + assert_equal(a.mask, [[0, 0], [0, 0]]) + + a = array([[1, 1], [3, 3]]) + b = array([1, 1], mask=[0, 1]) + a *= b + assert_equal(a, [[1, 1], [3, 3]]) + assert_equal(a.mask, [[0, 1], [0, 1]]) + + def test_inplace_addition_scalar_type(self): + # Test of inplace additions + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + xm[2] = masked + x += t(1) + assert_equal(x, y + t(1)) + xm += t(1) + assert_equal(xm, y + t(1)) + + def test_inplace_addition_array_type(self): + # Test of inplace additions + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + m = xm.mask + a = arange(10, dtype=t) + a[-1] = masked + x += a + xm += a + assert_equal(x, y + a) + assert_equal(xm, y + a) + assert_equal(xm.mask, mask_or(m, a.mask)) + + def test_inplace_subtraction_scalar_type(self): + # Test of inplace subtractions + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + x -= t(1) + assert_equal(x, y - t(1)) + xm -= t(1) + assert_equal(xm, y - t(1)) + + def test_inplace_subtraction_array_type(self): + # Test of inplace subtractions + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + m = xm.mask + a = arange(10, dtype=t) + a[-1] = masked + x -= a + xm -= a + assert_equal(x, y - a) + assert_equal(xm, y - a) + assert_equal(xm.mask, mask_or(m, a.mask)) + + def test_inplace_multiplication_scalar_type(self): + # Test of inplace multiplication + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + x *= t(2) + assert_equal(x, y * t(2)) + xm *= t(2) + assert_equal(xm, y * t(2)) + + def test_inplace_multiplication_array_type(self): + # Test of inplace multiplication + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + m = xm.mask + a = arange(10, dtype=t) + a[-1] = masked + x *= a + xm *= a + assert_equal(x, y * a) + assert_equal(xm, y * a) + assert_equal(xm.mask, mask_or(m, a.mask)) + + def test_inplace_floor_division_scalar_type(self): + # Test of inplace division + # Check for TypeError in case of unsupported types + unsupported = {np.dtype(t).type for t in np.typecodes["Complex"]} + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + x = arange(10, dtype=t) * t(2) + xm = arange(10, dtype=t) * t(2) + xm[2] = masked + try: + x //= t(2) + xm //= t(2) + assert_equal(x, y) + assert_equal(xm, y) + except TypeError: + msg = f"Supported type {t} throwing TypeError" + assert t in unsupported, msg + + def test_inplace_floor_division_array_type(self): + # Test of inplace division + # Check for TypeError in case of unsupported types + unsupported = {np.dtype(t).type for t in np.typecodes["Complex"]} + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + m = xm.mask + a = arange(10, dtype=t) + a[-1] = masked + try: + x //= a + xm //= a + assert_equal(x, y // a) + assert_equal(xm, y // a) + assert_equal( + xm.mask, + mask_or(mask_or(m, a.mask), (a == t(0))) + ) + except TypeError: + msg = f"Supported type {t} throwing TypeError" + assert t in unsupported, msg + + def test_inplace_division_scalar_type(self): + # Test of inplace division + for t in self.othertypes: + with suppress_warnings() as sup: + sup.record(UserWarning) + + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + x = arange(10, dtype=t) * t(2) + xm = arange(10, dtype=t) * t(2) + xm[2] = masked + + # May get a DeprecationWarning or a TypeError. + # + # This is a consequence of the fact that this is true divide + # and will require casting to float for calculation and + # casting back to the original type. This will only be raised + # with integers. Whether it is an error or warning is only + # dependent on how stringent the casting rules are. + # + # Will handle the same way. + try: + x /= t(2) + assert_equal(x, y) + except (DeprecationWarning, TypeError) as e: + warnings.warn(str(e), stacklevel=1) + try: + xm /= t(2) + assert_equal(xm, y) + except (DeprecationWarning, TypeError) as e: + warnings.warn(str(e), stacklevel=1) + + if issubclass(t, np.integer): + assert_equal(len(sup.log), 2, f'Failed on type={t}.') + else: + assert_equal(len(sup.log), 0, f'Failed on type={t}.') + + def test_inplace_division_array_type(self): + # Test of inplace division + for t in self.othertypes: + with suppress_warnings() as sup: + sup.record(UserWarning) + (x, y, xm) = (_.astype(t) for _ in self.uint8data) + m = xm.mask + a = arange(10, dtype=t) + a[-1] = masked + + # May get a DeprecationWarning or a TypeError. + # + # This is a consequence of the fact that this is true divide + # and will require casting to float for calculation and + # casting back to the original type. This will only be raised + # with integers. Whether it is an error or warning is only + # dependent on how stringent the casting rules are. + # + # Will handle the same way. + try: + x /= a + assert_equal(x, y / a) + except (DeprecationWarning, TypeError) as e: + warnings.warn(str(e), stacklevel=1) + try: + xm /= a + assert_equal(xm, y / a) + assert_equal( + xm.mask, + mask_or(mask_or(m, a.mask), (a == t(0))) + ) + except (DeprecationWarning, TypeError) as e: + warnings.warn(str(e), stacklevel=1) + + if issubclass(t, np.integer): + assert_equal(len(sup.log), 2, f'Failed on type={t}.') + else: + assert_equal(len(sup.log), 0, f'Failed on type={t}.') + + def test_inplace_pow_type(self): + # Test keeping data w/ (inplace) power + for t in self.othertypes: + with warnings.catch_warnings(): + warnings.filterwarnings("error") + # Test pow on scalar + x = array([1, 2, 3], mask=[0, 0, 1], dtype=t) + xx = x ** t(2) + xx_r = array([1, 2 ** 2, 3], mask=[0, 0, 1], dtype=t) + assert_equal(xx.data, xx_r.data) + assert_equal(xx.mask, xx_r.mask) + # Test ipow on scalar + x **= t(2) + assert_equal(x.data, xx_r.data) + assert_equal(x.mask, xx_r.mask) + + +class TestMaskedArrayMethods: + # Test class for miscellaneous MaskedArrays methods. + def setup_method(self): + # Base data definition. + x = np.array([8.375, 7.545, 8.828, 8.5, 1.757, 5.928, + 8.43, 7.78, 9.865, 5.878, 8.979, 4.732, + 3.012, 6.022, 5.095, 3.116, 5.238, 3.957, + 6.04, 9.63, 7.712, 3.382, 4.489, 6.479, + 7.189, 9.645, 5.395, 4.961, 9.894, 2.893, + 7.357, 9.828, 6.272, 3.758, 6.693, 0.993]) + X = x.reshape(6, 6) + XX = x.reshape(3, 2, 2, 3) + + m = np.array([0, 1, 0, 1, 0, 0, + 1, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 0, 1, + 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 0, 0, + 0, 0, 1, 0, 1, 0]) + mx = array(data=x, mask=m) + mX = array(data=X, mask=m.reshape(X.shape)) + mXX = array(data=XX, mask=m.reshape(XX.shape)) + + m2 = np.array([1, 1, 0, 1, 0, 0, + 1, 1, 1, 1, 0, 1, + 0, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 1, 0, + 0, 0, 1, 0, 1, 1]) + m2x = array(data=x, mask=m2) + m2X = array(data=X, mask=m2.reshape(X.shape)) + m2XX = array(data=XX, mask=m2.reshape(XX.shape)) + self.d = (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) + + def test_generic_methods(self): + # Tests some MaskedArray methods. + a = array([1, 3, 2]) + assert_equal(a.any(), a._data.any()) + assert_equal(a.all(), a._data.all()) + assert_equal(a.argmax(), a._data.argmax()) + assert_equal(a.argmin(), a._data.argmin()) + assert_equal(a.choose(0, 1, 2, 3, 4), a._data.choose(0, 1, 2, 3, 4)) + assert_equal(a.compress([1, 0, 1]), a._data.compress([1, 0, 1])) + assert_equal(a.conj(), a._data.conj()) + assert_equal(a.conjugate(), a._data.conjugate()) + + m = array([[1, 2], [3, 4]]) + assert_equal(m.diagonal(), m._data.diagonal()) + assert_equal(a.sum(), a._data.sum()) + assert_equal(a.take([1, 2]), a._data.take([1, 2])) + assert_equal(m.transpose(), m._data.transpose()) + + def test_allclose(self): + # Tests allclose on arrays + a = np.random.rand(10) + b = a + np.random.rand(10) * 1e-8 + assert_(allclose(a, b)) + # Test allclose w/ infs + a[0] = np.inf + assert_(not allclose(a, b)) + b[0] = np.inf + assert_(allclose(a, b)) + # Test allclose w/ masked + a = masked_array(a) + a[-1] = masked + assert_(allclose(a, b, masked_equal=True)) + assert_(not allclose(a, b, masked_equal=False)) + # Test comparison w/ scalar + a *= 1e-8 + a[0] = 0 + assert_(allclose(a, 0, masked_equal=True)) + + # Test that the function works for MIN_INT integer typed arrays + a = masked_array([np.iinfo(np.int_).min], dtype=np.int_) + assert_(allclose(a, a)) + + def test_allclose_timedelta(self): + # Allclose currently works for timedelta64 as long as `atol` is + # an integer or also a timedelta64 + a = np.array([[1, 2, 3, 4]], dtype="m8[ns]") + assert allclose(a, a, atol=0) + assert allclose(a, a, atol=np.timedelta64(1, "ns")) + + def test_allany(self): + # Checks the any/all methods/functions. + x = np.array([[0.13, 0.26, 0.90], + [0.28, 0.33, 0.63], + [0.31, 0.87, 0.70]]) + m = np.array([[True, False, False], + [False, False, False], + [True, True, False]], dtype=np.bool_) + mx = masked_array(x, mask=m) + mxbig = (mx > 0.5) + mxsmall = (mx < 0.5) + + assert_(not mxbig.all()) + assert_(mxbig.any()) + assert_equal(mxbig.all(0), [False, False, True]) + assert_equal(mxbig.all(1), [False, False, True]) + assert_equal(mxbig.any(0), [False, False, True]) + assert_equal(mxbig.any(1), [True, True, True]) + + assert_(not mxsmall.all()) + assert_(mxsmall.any()) + assert_equal(mxsmall.all(0), [True, True, False]) + assert_equal(mxsmall.all(1), [False, False, False]) + assert_equal(mxsmall.any(0), [True, True, False]) + assert_equal(mxsmall.any(1), [True, True, False]) + + def test_allany_oddities(self): + # Some fun with all and any + store = empty((), dtype=bool) + full = array([1, 2, 3], mask=True) + + assert_(full.all() is masked) + full.all(out=store) + assert_(store) + assert_(store._mask, True) + assert_(store is not masked) + + store = empty((), dtype=bool) + assert_(full.any() is masked) + full.any(out=store) + assert_(not store) + assert_(store._mask, True) + assert_(store is not masked) + + def test_argmax_argmin(self): + # Tests argmin & argmax on MaskedArrays. + (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) = self.d + + assert_equal(mx.argmin(), 35) + assert_equal(mX.argmin(), 35) + assert_equal(m2x.argmin(), 4) + assert_equal(m2X.argmin(), 4) + assert_equal(mx.argmax(), 28) + assert_equal(mX.argmax(), 28) + assert_equal(m2x.argmax(), 31) + assert_equal(m2X.argmax(), 31) + + assert_equal(mX.argmin(0), [2, 2, 2, 5, 0, 5]) + assert_equal(m2X.argmin(0), [2, 2, 4, 5, 0, 4]) + assert_equal(mX.argmax(0), [0, 5, 0, 5, 4, 0]) + assert_equal(m2X.argmax(0), [5, 5, 0, 5, 1, 0]) + + assert_equal(mX.argmin(1), [4, 1, 0, 0, 5, 5, ]) + assert_equal(m2X.argmin(1), [4, 4, 0, 0, 5, 3]) + assert_equal(mX.argmax(1), [2, 4, 1, 1, 4, 1]) + assert_equal(m2X.argmax(1), [2, 4, 1, 1, 1, 1]) + + def test_clip(self): + # Tests clip on MaskedArrays. + x = np.array([8.375, 7.545, 8.828, 8.5, 1.757, 5.928, + 8.43, 7.78, 9.865, 5.878, 8.979, 4.732, + 3.012, 6.022, 5.095, 3.116, 5.238, 3.957, + 6.04, 9.63, 7.712, 3.382, 4.489, 6.479, + 7.189, 9.645, 5.395, 4.961, 9.894, 2.893, + 7.357, 9.828, 6.272, 3.758, 6.693, 0.993]) + m = np.array([0, 1, 0, 1, 0, 0, 1, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0]) + mx = array(x, mask=m) + clipped = mx.clip(2, 8) + assert_equal(clipped.mask, mx.mask) + assert_equal(clipped._data, x.clip(2, 8)) + assert_equal(clipped._data, mx._data.clip(2, 8)) + + def test_clip_out(self): + # gh-14140 + a = np.arange(10) + m = np.ma.MaskedArray(a, mask=[0, 1] * 5) + m.clip(0, 5, out=m) + assert_equal(m.mask, [0, 1] * 5) + + def test_compress(self): + # test compress + a = masked_array([1., 2., 3., 4., 5.], fill_value=9999) + condition = (a > 1.5) & (a < 3.5) + assert_equal(a.compress(condition), [2., 3.]) + + a[[2, 3]] = masked + b = a.compress(condition) + assert_equal(b._data, [2., 3.]) + assert_equal(b._mask, [0, 1]) + assert_equal(b.fill_value, 9999) + assert_equal(b, a[condition]) + + condition = (a < 4.) + b = a.compress(condition) + assert_equal(b._data, [1., 2., 3.]) + assert_equal(b._mask, [0, 0, 1]) + assert_equal(b.fill_value, 9999) + assert_equal(b, a[condition]) + + a = masked_array([[10, 20, 30], [40, 50, 60]], + mask=[[0, 0, 1], [1, 0, 0]]) + b = a.compress(a.ravel() >= 22) + assert_equal(b._data, [30, 40, 50, 60]) + assert_equal(b._mask, [1, 1, 0, 0]) + + x = np.array([3, 1, 2]) + b = a.compress(x >= 2, axis=1) + assert_equal(b._data, [[10, 30], [40, 60]]) + assert_equal(b._mask, [[0, 1], [1, 0]]) + + def test_compressed(self): + # Tests compressed + a = array([1, 2, 3, 4], mask=[0, 0, 0, 0]) + b = a.compressed() + assert_equal(b, a) + a[0] = masked + b = a.compressed() + assert_equal(b, [2, 3, 4]) + + def test_empty(self): + # Tests empty/like + datatype = [('a', int), ('b', float), ('c', '|S8')] + a = masked_array([(1, 1.1, '1.1'), (2, 2.2, '2.2'), (3, 3.3, '3.3')], + dtype=datatype) + assert_equal(len(a.fill_value.item()), len(datatype)) + + b = empty_like(a) + assert_equal(b.shape, a.shape) + assert_equal(b.fill_value, a.fill_value) + + b = empty(len(a), dtype=datatype) + assert_equal(b.shape, a.shape) + assert_equal(b.fill_value, a.fill_value) + + # check empty_like mask handling + a = masked_array([1, 2, 3], mask=[False, True, False]) + b = empty_like(a) + assert_(not np.may_share_memory(a.mask, b.mask)) + b = a.view(masked_array) + assert_(np.may_share_memory(a.mask, b.mask)) + + def test_zeros(self): + # Tests zeros/like + datatype = [('a', int), ('b', float), ('c', '|S8')] + a = masked_array([(1, 1.1, '1.1'), (2, 2.2, '2.2'), (3, 3.3, '3.3')], + dtype=datatype) + assert_equal(len(a.fill_value.item()), len(datatype)) + + b = zeros(len(a), dtype=datatype) + assert_equal(b.shape, a.shape) + assert_equal(b.fill_value, a.fill_value) + + b = zeros_like(a) + assert_equal(b.shape, a.shape) + assert_equal(b.fill_value, a.fill_value) + + # check zeros_like mask handling + a = masked_array([1, 2, 3], mask=[False, True, False]) + b = zeros_like(a) + assert_(not np.may_share_memory(a.mask, b.mask)) + b = a.view() + assert_(np.may_share_memory(a.mask, b.mask)) + + def test_ones(self): + # Tests ones/like + datatype = [('a', int), ('b', float), ('c', '|S8')] + a = masked_array([(1, 1.1, '1.1'), (2, 2.2, '2.2'), (3, 3.3, '3.3')], + dtype=datatype) + assert_equal(len(a.fill_value.item()), len(datatype)) + + b = ones(len(a), dtype=datatype) + assert_equal(b.shape, a.shape) + assert_equal(b.fill_value, a.fill_value) + + b = ones_like(a) + assert_equal(b.shape, a.shape) + assert_equal(b.fill_value, a.fill_value) + + # check ones_like mask handling + a = masked_array([1, 2, 3], mask=[False, True, False]) + b = ones_like(a) + assert_(not np.may_share_memory(a.mask, b.mask)) + b = a.view() + assert_(np.may_share_memory(a.mask, b.mask)) + + @suppress_copy_mask_on_assignment + def test_put(self): + # Tests put. + d = arange(5) + n = [0, 0, 0, 1, 1] + m = make_mask(n) + x = array(d, mask=m) + assert_(x[3] is masked) + assert_(x[4] is masked) + x[[1, 4]] = [10, 40] + assert_(x[3] is masked) + assert_(x[4] is not masked) + assert_equal(x, [0, 10, 2, -1, 40]) + + x = masked_array(arange(10), mask=[1, 0, 0, 0, 0] * 2) + i = [0, 2, 4, 6] + x.put(i, [6, 4, 2, 0]) + assert_equal(x, asarray([6, 1, 4, 3, 2, 5, 0, 7, 8, 9, ])) + assert_equal(x.mask, [0, 0, 0, 0, 0, 1, 0, 0, 0, 0]) + x.put(i, masked_array([0, 2, 4, 6], [1, 0, 1, 0])) + assert_array_equal(x, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, ]) + assert_equal(x.mask, [1, 0, 0, 0, 1, 1, 0, 0, 0, 0]) + + x = masked_array(arange(10), mask=[1, 0, 0, 0, 0] * 2) + put(x, i, [6, 4, 2, 0]) + assert_equal(x, asarray([6, 1, 4, 3, 2, 5, 0, 7, 8, 9, ])) + assert_equal(x.mask, [0, 0, 0, 0, 0, 1, 0, 0, 0, 0]) + put(x, i, masked_array([0, 2, 4, 6], [1, 0, 1, 0])) + assert_array_equal(x, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, ]) + assert_equal(x.mask, [1, 0, 0, 0, 1, 1, 0, 0, 0, 0]) + + def test_put_nomask(self): + # GitHub issue 6425 + x = zeros(10) + z = array([3., -1.], mask=[False, True]) + + x.put([1, 2], z) + assert_(x[0] is not masked) + assert_equal(x[0], 0) + assert_(x[1] is not masked) + assert_equal(x[1], 3) + assert_(x[2] is masked) + assert_(x[3] is not masked) + assert_equal(x[3], 0) + + def test_put_hardmask(self): + # Tests put on hardmask + d = arange(5) + n = [0, 0, 0, 1, 1] + m = make_mask(n) + xh = array(d + 1, mask=m, hard_mask=True, copy=True) + xh.put([4, 2, 0, 1, 3], [1, 2, 3, 4, 5]) + assert_equal(xh._data, [3, 4, 2, 4, 5]) + + def test_putmask(self): + x = arange(6) + 1 + mx = array(x, mask=[0, 0, 0, 1, 1, 1]) + mask = [0, 0, 1, 0, 0, 1] + # w/o mask, w/o masked values + xx = x.copy() + putmask(xx, mask, 99) + assert_equal(xx, [1, 2, 99, 4, 5, 99]) + # w/ mask, w/o masked values + mxx = mx.copy() + putmask(mxx, mask, 99) + assert_equal(mxx._data, [1, 2, 99, 4, 5, 99]) + assert_equal(mxx._mask, [0, 0, 0, 1, 1, 0]) + # w/o mask, w/ masked values + values = array([10, 20, 30, 40, 50, 60], mask=[1, 1, 1, 0, 0, 0]) + xx = x.copy() + putmask(xx, mask, values) + assert_equal(xx._data, [1, 2, 30, 4, 5, 60]) + assert_equal(xx._mask, [0, 0, 1, 0, 0, 0]) + # w/ mask, w/ masked values + mxx = mx.copy() + putmask(mxx, mask, values) + assert_equal(mxx._data, [1, 2, 30, 4, 5, 60]) + assert_equal(mxx._mask, [0, 0, 1, 1, 1, 0]) + # w/ mask, w/ masked values + hardmask + mxx = mx.copy() + mxx.harden_mask() + putmask(mxx, mask, values) + assert_equal(mxx, [1, 2, 30, 4, 5, 60]) + + def test_ravel(self): + # Tests ravel + a = array([[1, 2, 3, 4, 5]], mask=[[0, 1, 0, 0, 0]]) + aravel = a.ravel() + assert_equal(aravel._mask.shape, aravel.shape) + a = array([0, 0], mask=[1, 1]) + aravel = a.ravel() + assert_equal(aravel._mask.shape, a.shape) + # Checks that small_mask is preserved + a = array([1, 2, 3, 4], mask=[0, 0, 0, 0], shrink=False) + assert_equal(a.ravel()._mask, [0, 0, 0, 0]) + # Test that the fill_value is preserved + a.fill_value = -99 + a.shape = (2, 2) + ar = a.ravel() + assert_equal(ar._mask, [0, 0, 0, 0]) + assert_equal(ar._data, [1, 2, 3, 4]) + assert_equal(ar.fill_value, -99) + # Test index ordering + assert_equal(a.ravel(order='C'), [1, 2, 3, 4]) + assert_equal(a.ravel(order='F'), [1, 3, 2, 4]) + + @pytest.mark.parametrize("order", "AKCF") + @pytest.mark.parametrize("data_order", "CF") + def test_ravel_order(self, order, data_order): + # Ravelling must ravel mask and data in the same order always to avoid + # misaligning the two in the ravel result. + arr = np.ones((5, 10), order=data_order) + arr[0, :] = 0 + mask = np.ones((10, 5), dtype=bool, order=data_order).T + mask[0, :] = False + x = array(arr, mask=mask) + assert x._data.flags.fnc != x._mask.flags.fnc + assert (x.filled(0) == 0).all() + raveled = x.ravel(order) + assert (raveled.filled(0) == 0).all() + + # NOTE: Can be wrong if arr order is neither C nor F and `order="K"` + assert_array_equal(arr.ravel(order), x.ravel(order)._data) + + def test_reshape(self): + # Tests reshape + x = arange(4) + x[0] = masked + y = x.reshape(2, 2) + assert_equal(y.shape, (2, 2,)) + assert_equal(y._mask.shape, (2, 2,)) + assert_equal(x.shape, (4,)) + assert_equal(x._mask.shape, (4,)) + + def test_sort(self): + # Test sort + x = array([1, 4, 2, 3], mask=[0, 1, 0, 0], dtype=np.uint8) + + sortedx = sort(x) + assert_equal(sortedx._data, [1, 2, 3, 4]) + assert_equal(sortedx._mask, [0, 0, 0, 1]) + + sortedx = sort(x, endwith=False) + assert_equal(sortedx._data, [4, 1, 2, 3]) + assert_equal(sortedx._mask, [1, 0, 0, 0]) + + x.sort() + assert_equal(x._data, [1, 2, 3, 4]) + assert_equal(x._mask, [0, 0, 0, 1]) + + x = array([1, 4, 2, 3], mask=[0, 1, 0, 0], dtype=np.uint8) + x.sort(endwith=False) + assert_equal(x._data, [4, 1, 2, 3]) + assert_equal(x._mask, [1, 0, 0, 0]) + + x = [1, 4, 2, 3] + sortedx = sort(x) + assert_(not isinstance(sorted, MaskedArray)) + + x = array([0, 1, -1, -2, 2], mask=nomask, dtype=np.int8) + sortedx = sort(x, endwith=False) + assert_equal(sortedx._data, [-2, -1, 0, 1, 2]) + x = array([0, 1, -1, -2, 2], mask=[0, 1, 0, 0, 1], dtype=np.int8) + sortedx = sort(x, endwith=False) + assert_equal(sortedx._data, [1, 2, -2, -1, 0]) + assert_equal(sortedx._mask, [1, 1, 0, 0, 0]) + + x = array([0, -1], dtype=np.int8) + sortedx = sort(x, kind="stable") + assert_equal(sortedx, array([-1, 0], dtype=np.int8)) + + def test_stable_sort(self): + x = array([1, 2, 3, 1, 2, 3], dtype=np.uint8) + expected = array([0, 3, 1, 4, 2, 5]) + computed = argsort(x, kind='stable') + assert_equal(computed, expected) + + def test_argsort_matches_sort(self): + x = array([1, 4, 2, 3], mask=[0, 1, 0, 0], dtype=np.uint8) + + for kwargs in [dict(), + dict(endwith=True), + dict(endwith=False), + dict(fill_value=2), + dict(fill_value=2, endwith=True), + dict(fill_value=2, endwith=False)]: + sortedx = sort(x, **kwargs) + argsortedx = x[argsort(x, **kwargs)] + assert_equal(sortedx._data, argsortedx._data) + assert_equal(sortedx._mask, argsortedx._mask) + + def test_sort_2d(self): + # Check sort of 2D array. + # 2D array w/o mask + a = masked_array([[8, 4, 1], [2, 0, 9]]) + a.sort(0) + assert_equal(a, [[2, 0, 1], [8, 4, 9]]) + a = masked_array([[8, 4, 1], [2, 0, 9]]) + a.sort(1) + assert_equal(a, [[1, 4, 8], [0, 2, 9]]) + # 2D array w/mask + a = masked_array([[8, 4, 1], [2, 0, 9]], mask=[[1, 0, 0], [0, 0, 1]]) + a.sort(0) + assert_equal(a, [[2, 0, 1], [8, 4, 9]]) + assert_equal(a._mask, [[0, 0, 0], [1, 0, 1]]) + a = masked_array([[8, 4, 1], [2, 0, 9]], mask=[[1, 0, 0], [0, 0, 1]]) + a.sort(1) + assert_equal(a, [[1, 4, 8], [0, 2, 9]]) + assert_equal(a._mask, [[0, 0, 1], [0, 0, 1]]) + # 3D + a = masked_array([[[7, 8, 9], [4, 5, 6], [1, 2, 3]], + [[1, 2, 3], [7, 8, 9], [4, 5, 6]], + [[7, 8, 9], [1, 2, 3], [4, 5, 6]], + [[4, 5, 6], [1, 2, 3], [7, 8, 9]]]) + a[a % 4 == 0] = masked + am = a.copy() + an = a.filled(99) + am.sort(0) + an.sort(0) + assert_equal(am, an) + am = a.copy() + an = a.filled(99) + am.sort(1) + an.sort(1) + assert_equal(am, an) + am = a.copy() + an = a.filled(99) + am.sort(2) + an.sort(2) + assert_equal(am, an) + + def test_sort_flexible(self): + # Test sort on structured dtype. + a = array( + data=[(3, 3), (3, 2), (2, 2), (2, 1), (1, 0), (1, 1), (1, 2)], + mask=[(0, 0), (0, 1), (0, 0), (0, 0), (1, 0), (0, 0), (0, 0)], + dtype=[('A', int), ('B', int)]) + mask_last = array( + data=[(1, 1), (1, 2), (2, 1), (2, 2), (3, 3), (3, 2), (1, 0)], + mask=[(0, 0), (0, 0), (0, 0), (0, 0), (0, 0), (0, 1), (1, 0)], + dtype=[('A', int), ('B', int)]) + mask_first = array( + data=[(1, 0), (1, 1), (1, 2), (2, 1), (2, 2), (3, 2), (3, 3)], + mask=[(1, 0), (0, 0), (0, 0), (0, 0), (0, 0), (0, 1), (0, 0)], + dtype=[('A', int), ('B', int)]) + + test = sort(a) + assert_equal(test, mask_last) + assert_equal(test.mask, mask_last.mask) + + test = sort(a, endwith=False) + assert_equal(test, mask_first) + assert_equal(test.mask, mask_first.mask) + + # Test sort on dtype with subarray (gh-8069) + # Just check that the sort does not error, structured array subarrays + # are treated as byte strings and that leads to differing behavior + # depending on endianness and `endwith`. + dt = np.dtype([('v', int, 2)]) + a = a.view(dt) + test = sort(a) + test = sort(a, endwith=False) + + def test_argsort(self): + # Test argsort + a = array([1, 5, 2, 4, 3], mask=[1, 0, 0, 1, 0]) + assert_equal(np.argsort(a), argsort(a)) + + def test_squeeze(self): + # Check squeeze + data = masked_array([[1, 2, 3]]) + assert_equal(data.squeeze(), [1, 2, 3]) + data = masked_array([[1, 2, 3]], mask=[[1, 1, 1]]) + assert_equal(data.squeeze(), [1, 2, 3]) + assert_equal(data.squeeze()._mask, [1, 1, 1]) + + # normal ndarrays return a view + arr = np.array([[1]]) + arr_sq = arr.squeeze() + assert_equal(arr_sq, 1) + arr_sq[...] = 2 + assert_equal(arr[0,0], 2) + + # so maskedarrays should too + m_arr = masked_array([[1]], mask=True) + m_arr_sq = m_arr.squeeze() + assert_(m_arr_sq is not np.ma.masked) + assert_equal(m_arr_sq.mask, True) + m_arr_sq[...] = 2 + assert_equal(m_arr[0,0], 2) + + def test_swapaxes(self): + # Tests swapaxes on MaskedArrays. + x = np.array([8.375, 7.545, 8.828, 8.5, 1.757, 5.928, + 8.43, 7.78, 9.865, 5.878, 8.979, 4.732, + 3.012, 6.022, 5.095, 3.116, 5.238, 3.957, + 6.04, 9.63, 7.712, 3.382, 4.489, 6.479, + 7.189, 9.645, 5.395, 4.961, 9.894, 2.893, + 7.357, 9.828, 6.272, 3.758, 6.693, 0.993]) + m = np.array([0, 1, 0, 1, 0, 0, + 1, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 0, 1, + 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 0, 0, + 0, 0, 1, 0, 1, 0]) + mX = array(x, mask=m).reshape(6, 6) + mXX = mX.reshape(3, 2, 2, 3) + + mXswapped = mX.swapaxes(0, 1) + assert_equal(mXswapped[-1], mX[:, -1]) + + mXXswapped = mXX.swapaxes(0, 2) + assert_equal(mXXswapped.shape, (2, 2, 3, 3)) + + def test_take(self): + # Tests take + x = masked_array([10, 20, 30, 40], [0, 1, 0, 1]) + assert_equal(x.take([0, 0, 3]), masked_array([10, 10, 40], [0, 0, 1])) + assert_equal(x.take([0, 0, 3]), x[[0, 0, 3]]) + assert_equal(x.take([[0, 1], [0, 1]]), + masked_array([[10, 20], [10, 20]], [[0, 1], [0, 1]])) + + # assert_equal crashes when passed np.ma.mask + assert_(x[1] is np.ma.masked) + assert_(x.take(1) is np.ma.masked) + + x = array([[10, 20, 30], [40, 50, 60]], mask=[[0, 0, 1], [1, 0, 0, ]]) + assert_equal(x.take([0, 2], axis=1), + array([[10, 30], [40, 60]], mask=[[0, 1], [1, 0]])) + assert_equal(take(x, [0, 2], axis=1), + array([[10, 30], [40, 60]], mask=[[0, 1], [1, 0]])) + + def test_take_masked_indices(self): + # Test take w/ masked indices + a = np.array((40, 18, 37, 9, 22)) + indices = np.arange(3)[None,:] + np.arange(5)[:, None] + mindices = array(indices, mask=(indices >= len(a))) + # No mask + test = take(a, mindices, mode='clip') + ctrl = array([[40, 18, 37], + [18, 37, 9], + [37, 9, 22], + [9, 22, 22], + [22, 22, 22]]) + assert_equal(test, ctrl) + # Masked indices + test = take(a, mindices) + ctrl = array([[40, 18, 37], + [18, 37, 9], + [37, 9, 22], + [9, 22, 40], + [22, 40, 40]]) + ctrl[3, 2] = ctrl[4, 1] = ctrl[4, 2] = masked + assert_equal(test, ctrl) + assert_equal(test.mask, ctrl.mask) + # Masked input + masked indices + a = array((40, 18, 37, 9, 22), mask=(0, 1, 0, 0, 0)) + test = take(a, mindices) + ctrl[0, 1] = ctrl[1, 0] = masked + assert_equal(test, ctrl) + assert_equal(test.mask, ctrl.mask) + + def test_tolist(self): + # Tests to list + # ... on 1D + x = array(np.arange(12)) + x[[1, -2]] = masked + xlist = x.tolist() + assert_(xlist[1] is None) + assert_(xlist[-2] is None) + # ... on 2D + x.shape = (3, 4) + xlist = x.tolist() + ctrl = [[0, None, 2, 3], [4, 5, 6, 7], [8, 9, None, 11]] + assert_equal(xlist[0], [0, None, 2, 3]) + assert_equal(xlist[1], [4, 5, 6, 7]) + assert_equal(xlist[2], [8, 9, None, 11]) + assert_equal(xlist, ctrl) + # ... on structured array w/ masked records + x = array(list(zip([1, 2, 3], + [1.1, 2.2, 3.3], + ['one', 'two', 'thr'])), + dtype=[('a', int), ('b', float), ('c', '|S8')]) + x[-1] = masked + assert_equal(x.tolist(), + [(1, 1.1, b'one'), + (2, 2.2, b'two'), + (None, None, None)]) + # ... on structured array w/ masked fields + a = array([(1, 2,), (3, 4)], mask=[(0, 1), (0, 0)], + dtype=[('a', int), ('b', int)]) + test = a.tolist() + assert_equal(test, [[1, None], [3, 4]]) + # ... on mvoid + a = a[0] + test = a.tolist() + assert_equal(test, [1, None]) + + def test_tolist_specialcase(self): + # Test mvoid.tolist: make sure we return a standard Python object + a = array([(0, 1), (2, 3)], dtype=[('a', int), ('b', int)]) + # w/o mask: each entry is a np.void whose elements are standard Python + for entry in a: + for item in entry.tolist(): + assert_(not isinstance(item, np.generic)) + # w/ mask: each entry is a ma.void whose elements should be + # standard Python + a.mask[0] = (0, 1) + for entry in a: + for item in entry.tolist(): + assert_(not isinstance(item, np.generic)) + + def test_toflex(self): + # Test the conversion to records + data = arange(10) + record = data.toflex() + assert_equal(record['_data'], data._data) + assert_equal(record['_mask'], data._mask) + + data[[0, 1, 2, -1]] = masked + record = data.toflex() + assert_equal(record['_data'], data._data) + assert_equal(record['_mask'], data._mask) + + ndtype = [('i', int), ('s', '|S3'), ('f', float)] + data = array([(i, s, f) for (i, s, f) in zip(np.arange(10), + 'ABCDEFGHIJKLM', + np.random.rand(10))], + dtype=ndtype) + data[[0, 1, 2, -1]] = masked + record = data.toflex() + assert_equal(record['_data'], data._data) + assert_equal(record['_mask'], data._mask) + + ndtype = np.dtype("int, (2,3)float, float") + data = array([(i, f, ff) for (i, f, ff) in zip(np.arange(10), + np.random.rand(10), + np.random.rand(10))], + dtype=ndtype) + data[[0, 1, 2, -1]] = masked + record = data.toflex() + assert_equal_records(record['_data'], data._data) + assert_equal_records(record['_mask'], data._mask) + + def test_fromflex(self): + # Test the reconstruction of a masked_array from a record + a = array([1, 2, 3]) + test = fromflex(a.toflex()) + assert_equal(test, a) + assert_equal(test.mask, a.mask) + + a = array([1, 2, 3], mask=[0, 0, 1]) + test = fromflex(a.toflex()) + assert_equal(test, a) + assert_equal(test.mask, a.mask) + + a = array([(1, 1.), (2, 2.), (3, 3.)], mask=[(1, 0), (0, 0), (0, 1)], + dtype=[('A', int), ('B', float)]) + test = fromflex(a.toflex()) + assert_equal(test, a) + assert_equal(test.data, a.data) + + def test_arraymethod(self): + # Test a _arraymethod w/ n argument + marray = masked_array([[1, 2, 3, 4, 5]], mask=[0, 0, 1, 0, 0]) + control = masked_array([[1], [2], [3], [4], [5]], + mask=[0, 0, 1, 0, 0]) + assert_equal(marray.T, control) + assert_equal(marray.transpose(), control) + + assert_equal(MaskedArray.cumsum(marray.T, 0), control.cumsum(0)) + + def test_arraymethod_0d(self): + # gh-9430 + x = np.ma.array(42, mask=True) + assert_equal(x.T.mask, x.mask) + assert_equal(x.T.data, x.data) + + def test_transpose_view(self): + x = np.ma.array([[1, 2, 3], [4, 5, 6]]) + x[0,1] = np.ma.masked + xt = x.T + + xt[1,0] = 10 + xt[0,1] = np.ma.masked + + assert_equal(x.data, xt.T.data) + assert_equal(x.mask, xt.T.mask) + + def test_diagonal_view(self): + x = np.ma.zeros((3,3)) + x[0,0] = 10 + x[1,1] = np.ma.masked + x[2,2] = 20 + xd = x.diagonal() + x[1,1] = 15 + assert_equal(xd.mask, x.diagonal().mask) + assert_equal(xd.data, x.diagonal().data) + + +class TestMaskedArrayMathMethods: + + def setup_method(self): + # Base data definition. + x = np.array([8.375, 7.545, 8.828, 8.5, 1.757, 5.928, + 8.43, 7.78, 9.865, 5.878, 8.979, 4.732, + 3.012, 6.022, 5.095, 3.116, 5.238, 3.957, + 6.04, 9.63, 7.712, 3.382, 4.489, 6.479, + 7.189, 9.645, 5.395, 4.961, 9.894, 2.893, + 7.357, 9.828, 6.272, 3.758, 6.693, 0.993]) + X = x.reshape(6, 6) + XX = x.reshape(3, 2, 2, 3) + + m = np.array([0, 1, 0, 1, 0, 0, + 1, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 0, 1, + 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 0, 0, + 0, 0, 1, 0, 1, 0]) + mx = array(data=x, mask=m) + mX = array(data=X, mask=m.reshape(X.shape)) + mXX = array(data=XX, mask=m.reshape(XX.shape)) + + m2 = np.array([1, 1, 0, 1, 0, 0, + 1, 1, 1, 1, 0, 1, + 0, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 1, 0, + 0, 0, 1, 0, 1, 1]) + m2x = array(data=x, mask=m2) + m2X = array(data=X, mask=m2.reshape(X.shape)) + m2XX = array(data=XX, mask=m2.reshape(XX.shape)) + self.d = (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) + + def test_cumsumprod(self): + # Tests cumsum & cumprod on MaskedArrays. + (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) = self.d + mXcp = mX.cumsum(0) + assert_equal(mXcp._data, mX.filled(0).cumsum(0)) + mXcp = mX.cumsum(1) + assert_equal(mXcp._data, mX.filled(0).cumsum(1)) + + mXcp = mX.cumprod(0) + assert_equal(mXcp._data, mX.filled(1).cumprod(0)) + mXcp = mX.cumprod(1) + assert_equal(mXcp._data, mX.filled(1).cumprod(1)) + + def test_cumsumprod_with_output(self): + # Tests cumsum/cumprod w/ output + xm = array(np.random.uniform(0, 10, 12)).reshape(3, 4) + xm[:, 0] = xm[0] = xm[-1, -1] = masked + + for funcname in ('cumsum', 'cumprod'): + npfunc = getattr(np, funcname) + xmmeth = getattr(xm, funcname) + + # A ndarray as explicit input + output = np.empty((3, 4), dtype=float) + output.fill(-9999) + result = npfunc(xm, axis=0, out=output) + # ... the result should be the given output + assert_(result is output) + assert_equal(result, xmmeth(axis=0, out=output)) + + output = empty((3, 4), dtype=int) + result = xmmeth(axis=0, out=output) + assert_(result is output) + + def test_ptp(self): + # Tests ptp on MaskedArrays. + (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) = self.d + (n, m) = X.shape + assert_equal(mx.ptp(), mx.compressed().ptp()) + rows = np.zeros(n, float) + cols = np.zeros(m, float) + for k in range(m): + cols[k] = mX[:, k].compressed().ptp() + for k in range(n): + rows[k] = mX[k].compressed().ptp() + assert_equal(mX.ptp(0), cols) + assert_equal(mX.ptp(1), rows) + + def test_add_object(self): + x = masked_array(['a', 'b'], mask=[1, 0], dtype=object) + y = x + 'x' + assert_equal(y[1], 'bx') + assert_(y.mask[0]) + + def test_sum_object(self): + # Test sum on object dtype + a = masked_array([1, 2, 3], mask=[1, 0, 0], dtype=object) + assert_equal(a.sum(), 5) + a = masked_array([[1, 2, 3], [4, 5, 6]], dtype=object) + assert_equal(a.sum(axis=0), [5, 7, 9]) + + def test_prod_object(self): + # Test prod on object dtype + a = masked_array([1, 2, 3], mask=[1, 0, 0], dtype=object) + assert_equal(a.prod(), 2 * 3) + a = masked_array([[1, 2, 3], [4, 5, 6]], dtype=object) + assert_equal(a.prod(axis=0), [4, 10, 18]) + + def test_meananom_object(self): + # Test mean/anom on object dtype + a = masked_array([1, 2, 3], dtype=object) + assert_equal(a.mean(), 2) + assert_equal(a.anom(), [-1, 0, 1]) + + def test_anom_shape(self): + a = masked_array([1, 2, 3]) + assert_equal(a.anom().shape, a.shape) + a.mask = True + assert_equal(a.anom().shape, a.shape) + assert_(np.ma.is_masked(a.anom())) + + def test_anom(self): + a = masked_array(np.arange(1, 7).reshape(2, 3)) + assert_almost_equal(a.anom(), + [[-2.5, -1.5, -0.5], [0.5, 1.5, 2.5]]) + assert_almost_equal(a.anom(axis=0), + [[-1.5, -1.5, -1.5], [1.5, 1.5, 1.5]]) + assert_almost_equal(a.anom(axis=1), + [[-1., 0., 1.], [-1., 0., 1.]]) + a.mask = [[0, 0, 1], [0, 1, 0]] + mval = -99 + assert_almost_equal(a.anom().filled(mval), + [[-2.25, -1.25, mval], [0.75, mval, 2.75]]) + assert_almost_equal(a.anom(axis=0).filled(mval), + [[-1.5, 0.0, mval], [1.5, mval, 0.0]]) + assert_almost_equal(a.anom(axis=1).filled(mval), + [[-0.5, 0.5, mval], [-1.0, mval, 1.0]]) + + def test_trace(self): + # Tests trace on MaskedArrays. + (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) = self.d + mXdiag = mX.diagonal() + assert_equal(mX.trace(), mX.diagonal().compressed().sum()) + assert_almost_equal(mX.trace(), + X.trace() - sum(mXdiag.mask * X.diagonal(), + axis=0)) + assert_equal(np.trace(mX), mX.trace()) + + # gh-5560 + arr = np.arange(2*4*4).reshape(2,4,4) + m_arr = np.ma.masked_array(arr, False) + assert_equal(arr.trace(axis1=1, axis2=2), m_arr.trace(axis1=1, axis2=2)) + + def test_dot(self): + # Tests dot on MaskedArrays. + (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) = self.d + fx = mx.filled(0) + r = mx.dot(mx) + assert_almost_equal(r.filled(0), fx.dot(fx)) + assert_(r.mask is nomask) + + fX = mX.filled(0) + r = mX.dot(mX) + assert_almost_equal(r.filled(0), fX.dot(fX)) + assert_(r.mask[1,3]) + r1 = empty_like(r) + mX.dot(mX, out=r1) + assert_almost_equal(r, r1) + + mYY = mXX.swapaxes(-1, -2) + fXX, fYY = mXX.filled(0), mYY.filled(0) + r = mXX.dot(mYY) + assert_almost_equal(r.filled(0), fXX.dot(fYY)) + r1 = empty_like(r) + mXX.dot(mYY, out=r1) + assert_almost_equal(r, r1) + + def test_dot_shape_mismatch(self): + # regression test + x = masked_array([[1,2],[3,4]], mask=[[0,1],[0,0]]) + y = masked_array([[1,2],[3,4]], mask=[[0,1],[0,0]]) + z = masked_array([[0,1],[3,3]]) + x.dot(y, out=z) + assert_almost_equal(z.filled(0), [[1, 0], [15, 16]]) + assert_almost_equal(z.mask, [[0, 1], [0, 0]]) + + def test_varmean_nomask(self): + # gh-5769 + foo = array([1,2,3,4], dtype='f8') + bar = array([1,2,3,4], dtype='f8') + assert_equal(type(foo.mean()), np.float64) + assert_equal(type(foo.var()), np.float64) + assert((foo.mean() == bar.mean()) is np.bool_(True)) + + # check array type is preserved and out works + foo = array(np.arange(16).reshape((4,4)), dtype='f8') + bar = empty(4, dtype='f4') + assert_equal(type(foo.mean(axis=1)), MaskedArray) + assert_equal(type(foo.var(axis=1)), MaskedArray) + assert_(foo.mean(axis=1, out=bar) is bar) + assert_(foo.var(axis=1, out=bar) is bar) + + def test_varstd(self): + # Tests var & std on MaskedArrays. + (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) = self.d + assert_almost_equal(mX.var(axis=None), mX.compressed().var()) + assert_almost_equal(mX.std(axis=None), mX.compressed().std()) + assert_almost_equal(mX.std(axis=None, ddof=1), + mX.compressed().std(ddof=1)) + assert_almost_equal(mX.var(axis=None, ddof=1), + mX.compressed().var(ddof=1)) + assert_equal(mXX.var(axis=3).shape, XX.var(axis=3).shape) + assert_equal(mX.var().shape, X.var().shape) + (mXvar0, mXvar1) = (mX.var(axis=0), mX.var(axis=1)) + assert_almost_equal(mX.var(axis=None, ddof=2), + mX.compressed().var(ddof=2)) + assert_almost_equal(mX.std(axis=None, ddof=2), + mX.compressed().std(ddof=2)) + for k in range(6): + assert_almost_equal(mXvar1[k], mX[k].compressed().var()) + assert_almost_equal(mXvar0[k], mX[:, k].compressed().var()) + assert_almost_equal(np.sqrt(mXvar0[k]), + mX[:, k].compressed().std()) + + @suppress_copy_mask_on_assignment + def test_varstd_specialcases(self): + # Test a special case for var + nout = np.array(-1, dtype=float) + mout = array(-1, dtype=float) + + x = array(arange(10), mask=True) + for methodname in ('var', 'std'): + method = getattr(x, methodname) + assert_(method() is masked) + assert_(method(0) is masked) + assert_(method(-1) is masked) + # Using a masked array as explicit output + method(out=mout) + assert_(mout is not masked) + assert_equal(mout.mask, True) + # Using a ndarray as explicit output + method(out=nout) + assert_(np.isnan(nout)) + + x = array(arange(10), mask=True) + x[-1] = 9 + for methodname in ('var', 'std'): + method = getattr(x, methodname) + assert_(method(ddof=1) is masked) + assert_(method(0, ddof=1) is masked) + assert_(method(-1, ddof=1) is masked) + # Using a masked array as explicit output + method(out=mout, ddof=1) + assert_(mout is not masked) + assert_equal(mout.mask, True) + # Using a ndarray as explicit output + method(out=nout, ddof=1) + assert_(np.isnan(nout)) + + def test_varstd_ddof(self): + a = array([[1, 1, 0], [1, 1, 0]], mask=[[0, 0, 1], [0, 0, 1]]) + test = a.std(axis=0, ddof=0) + assert_equal(test.filled(0), [0, 0, 0]) + assert_equal(test.mask, [0, 0, 1]) + test = a.std(axis=0, ddof=1) + assert_equal(test.filled(0), [0, 0, 0]) + assert_equal(test.mask, [0, 0, 1]) + test = a.std(axis=0, ddof=2) + assert_equal(test.filled(0), [0, 0, 0]) + assert_equal(test.mask, [1, 1, 1]) + + def test_diag(self): + # Test diag + x = arange(9).reshape((3, 3)) + x[1, 1] = masked + out = np.diag(x) + assert_equal(out, [0, 4, 8]) + out = diag(x) + assert_equal(out, [0, 4, 8]) + assert_equal(out.mask, [0, 1, 0]) + out = diag(out) + control = array([[0, 0, 0], [0, 4, 0], [0, 0, 8]], + mask=[[0, 0, 0], [0, 1, 0], [0, 0, 0]]) + assert_equal(out, control) + + def test_axis_methods_nomask(self): + # Test the combination nomask & methods w/ axis + a = array([[1, 2, 3], [4, 5, 6]]) + + assert_equal(a.sum(0), [5, 7, 9]) + assert_equal(a.sum(-1), [6, 15]) + assert_equal(a.sum(1), [6, 15]) + + assert_equal(a.prod(0), [4, 10, 18]) + assert_equal(a.prod(-1), [6, 120]) + assert_equal(a.prod(1), [6, 120]) + + assert_equal(a.min(0), [1, 2, 3]) + assert_equal(a.min(-1), [1, 4]) + assert_equal(a.min(1), [1, 4]) + + assert_equal(a.max(0), [4, 5, 6]) + assert_equal(a.max(-1), [3, 6]) + assert_equal(a.max(1), [3, 6]) + + @requires_memory(free_bytes=2 * 10000 * 1000 * 2) + def test_mean_overflow(self): + # Test overflow in masked arrays + # gh-20272 + a = masked_array(np.full((10000, 10000), 65535, dtype=np.uint16), + mask=np.zeros((10000, 10000))) + assert_equal(a.mean(), 65535.0) + + def test_diff_with_prepend(self): + # GH 22465 + x = np.array([1, 2, 2, 3, 4, 2, 1, 1]) + + a = np.ma.masked_equal(x[3:], value=2) + a_prep = np.ma.masked_equal(x[:3], value=2) + diff1 = np.ma.diff(a, prepend=a_prep, axis=0) + + b = np.ma.masked_equal(x, value=2) + diff2 = np.ma.diff(b, axis=0) + + assert_(np.ma.allequal(diff1, diff2)) + + def test_diff_with_append(self): + # GH 22465 + x = np.array([1, 2, 2, 3, 4, 2, 1, 1]) + + a = np.ma.masked_equal(x[:3], value=2) + a_app = np.ma.masked_equal(x[3:], value=2) + diff1 = np.ma.diff(a, append=a_app, axis=0) + + b = np.ma.masked_equal(x, value=2) + diff2 = np.ma.diff(b, axis=0) + + assert_(np.ma.allequal(diff1, diff2)) + + def test_diff_with_dim_0(self): + with pytest.raises( + ValueError, + match="diff requires input that is at least one dimensional" + ): + np.ma.diff(np.array(1)) + + def test_diff_with_n_0(self): + a = np.ma.masked_equal([1, 2, 2, 3, 4, 2, 1, 1], value=2) + diff = np.ma.diff(a, n=0, axis=0) + + assert_(np.ma.allequal(a, diff)) + + +class TestMaskedArrayMathMethodsComplex: + # Test class for miscellaneous MaskedArrays methods. + def setup_method(self): + # Base data definition. + x = np.array([8.375j, 7.545j, 8.828j, 8.5j, 1.757j, 5.928, + 8.43, 7.78, 9.865, 5.878, 8.979, 4.732, + 3.012, 6.022, 5.095, 3.116, 5.238, 3.957, + 6.04, 9.63, 7.712, 3.382, 4.489, 6.479j, + 7.189j, 9.645, 5.395, 4.961, 9.894, 2.893, + 7.357, 9.828, 6.272, 3.758, 6.693, 0.993j]) + X = x.reshape(6, 6) + XX = x.reshape(3, 2, 2, 3) + + m = np.array([0, 1, 0, 1, 0, 0, + 1, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 0, 1, + 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 0, 0, + 0, 0, 1, 0, 1, 0]) + mx = array(data=x, mask=m) + mX = array(data=X, mask=m.reshape(X.shape)) + mXX = array(data=XX, mask=m.reshape(XX.shape)) + + m2 = np.array([1, 1, 0, 1, 0, 0, + 1, 1, 1, 1, 0, 1, + 0, 0, 1, 1, 0, 1, + 0, 0, 0, 1, 1, 1, + 1, 0, 0, 1, 1, 0, + 0, 0, 1, 0, 1, 1]) + m2x = array(data=x, mask=m2) + m2X = array(data=X, mask=m2.reshape(X.shape)) + m2XX = array(data=XX, mask=m2.reshape(XX.shape)) + self.d = (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) + + def test_varstd(self): + # Tests var & std on MaskedArrays. + (x, X, XX, m, mx, mX, mXX, m2x, m2X, m2XX) = self.d + assert_almost_equal(mX.var(axis=None), mX.compressed().var()) + assert_almost_equal(mX.std(axis=None), mX.compressed().std()) + assert_equal(mXX.var(axis=3).shape, XX.var(axis=3).shape) + assert_equal(mX.var().shape, X.var().shape) + (mXvar0, mXvar1) = (mX.var(axis=0), mX.var(axis=1)) + assert_almost_equal(mX.var(axis=None, ddof=2), + mX.compressed().var(ddof=2)) + assert_almost_equal(mX.std(axis=None, ddof=2), + mX.compressed().std(ddof=2)) + for k in range(6): + assert_almost_equal(mXvar1[k], mX[k].compressed().var()) + assert_almost_equal(mXvar0[k], mX[:, k].compressed().var()) + assert_almost_equal(np.sqrt(mXvar0[k]), + mX[:, k].compressed().std()) + + +class TestMaskedArrayFunctions: + # Test class for miscellaneous functions. + + def setup_method(self): + x = np.array([1., 1., 1., -2., pi/2.0, 4., 5., -10., 10., 1., 2., 3.]) + y = np.array([5., 0., 3., 2., -1., -4., 0., -10., 10., 1., 0., 3.]) + m1 = [1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0] + m2 = [0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 1] + xm = masked_array(x, mask=m1) + ym = masked_array(y, mask=m2) + xm.set_fill_value(1e+20) + self.info = (xm, ym) + + def test_masked_where_bool(self): + x = [1, 2] + y = masked_where(False, x) + assert_equal(y, [1, 2]) + assert_equal(y[1], 2) + + def test_masked_equal_wlist(self): + x = [1, 2, 3] + mx = masked_equal(x, 3) + assert_equal(mx, x) + assert_equal(mx._mask, [0, 0, 1]) + mx = masked_not_equal(x, 3) + assert_equal(mx, x) + assert_equal(mx._mask, [1, 1, 0]) + + def test_masked_equal_fill_value(self): + x = [1, 2, 3] + mx = masked_equal(x, 3) + assert_equal(mx._mask, [0, 0, 1]) + assert_equal(mx.fill_value, 3) + + def test_masked_where_condition(self): + # Tests masking functions. + x = array([1., 2., 3., 4., 5.]) + x[2] = masked + assert_equal(masked_where(greater(x, 2), x), masked_greater(x, 2)) + assert_equal(masked_where(greater_equal(x, 2), x), + masked_greater_equal(x, 2)) + assert_equal(masked_where(less(x, 2), x), masked_less(x, 2)) + assert_equal(masked_where(less_equal(x, 2), x), + masked_less_equal(x, 2)) + assert_equal(masked_where(not_equal(x, 2), x), masked_not_equal(x, 2)) + assert_equal(masked_where(equal(x, 2), x), masked_equal(x, 2)) + assert_equal(masked_where(not_equal(x, 2), x), masked_not_equal(x, 2)) + assert_equal(masked_where([1, 1, 0, 0, 0], [1, 2, 3, 4, 5]), + [99, 99, 3, 4, 5]) + + def test_masked_where_oddities(self): + # Tests some generic features. + atest = ones((10, 10, 10), dtype=float) + btest = zeros(atest.shape, MaskType) + ctest = masked_where(btest, atest) + assert_equal(atest, ctest) + + def test_masked_where_shape_constraint(self): + a = arange(10) + with assert_raises(IndexError): + masked_equal(1, a) + test = masked_equal(a, 1) + assert_equal(test.mask, [0, 1, 0, 0, 0, 0, 0, 0, 0, 0]) + + def test_masked_where_structured(self): + # test that masked_where on a structured array sets a structured + # mask (see issue #2972) + a = np.zeros(10, dtype=[("A", " 6, x) + + def test_masked_otherfunctions(self): + assert_equal(masked_inside(list(range(5)), 1, 3), + [0, 199, 199, 199, 4]) + assert_equal(masked_outside(list(range(5)), 1, 3), [199, 1, 2, 3, 199]) + assert_equal(masked_inside(array(list(range(5)), + mask=[1, 0, 0, 0, 0]), 1, 3).mask, + [1, 1, 1, 1, 0]) + assert_equal(masked_outside(array(list(range(5)), + mask=[0, 1, 0, 0, 0]), 1, 3).mask, + [1, 1, 0, 0, 1]) + assert_equal(masked_equal(array(list(range(5)), + mask=[1, 0, 0, 0, 0]), 2).mask, + [1, 0, 1, 0, 0]) + assert_equal(masked_not_equal(array([2, 2, 1, 2, 1], + mask=[1, 0, 0, 0, 0]), 2).mask, + [1, 0, 1, 0, 1]) + + def test_round(self): + a = array([1.23456, 2.34567, 3.45678, 4.56789, 5.67890], + mask=[0, 1, 0, 0, 0]) + assert_equal(a.round(), [1., 2., 3., 5., 6.]) + assert_equal(a.round(1), [1.2, 2.3, 3.5, 4.6, 5.7]) + assert_equal(a.round(3), [1.235, 2.346, 3.457, 4.568, 5.679]) + b = empty_like(a) + a.round(out=b) + assert_equal(b, [1., 2., 3., 5., 6.]) + + x = array([1., 2., 3., 4., 5.]) + c = array([1, 1, 1, 0, 0]) + x[2] = masked + z = where(c, x, -x) + assert_equal(z, [1., 2., 0., -4., -5]) + c[0] = masked + z = where(c, x, -x) + assert_equal(z, [1., 2., 0., -4., -5]) + assert_(z[0] is masked) + assert_(z[1] is not masked) + assert_(z[2] is masked) + + def test_round_with_output(self): + # Testing round with an explicit output + + xm = array(np.random.uniform(0, 10, 12)).reshape(3, 4) + xm[:, 0] = xm[0] = xm[-1, -1] = masked + + # A ndarray as explicit input + output = np.empty((3, 4), dtype=float) + output.fill(-9999) + result = np.round(xm, decimals=2, out=output) + # ... the result should be the given output + assert_(result is output) + assert_equal(result, xm.round(decimals=2, out=output)) + + output = empty((3, 4), dtype=float) + result = xm.round(decimals=2, out=output) + assert_(result is output) + + def test_round_with_scalar(self): + # Testing round with scalar/zero dimension input + # GH issue 2244 + a = array(1.1, mask=[False]) + assert_equal(a.round(), 1) + + a = array(1.1, mask=[True]) + assert_(a.round() is masked) + + a = array(1.1, mask=[False]) + output = np.empty(1, dtype=float) + output.fill(-9999) + a.round(out=output) + assert_equal(output, 1) + + a = array(1.1, mask=[False]) + output = array(-9999., mask=[True]) + a.round(out=output) + assert_equal(output[()], 1) + + a = array(1.1, mask=[True]) + output = array(-9999., mask=[False]) + a.round(out=output) + assert_(output[()] is masked) + + def test_identity(self): + a = identity(5) + assert_(isinstance(a, MaskedArray)) + assert_equal(a, np.identity(5)) + + def test_power(self): + x = -1.1 + assert_almost_equal(power(x, 2.), 1.21) + assert_(power(x, masked) is masked) + x = array([-1.1, -1.1, 1.1, 1.1, 0.]) + b = array([0.5, 2., 0.5, 2., -1.], mask=[0, 0, 0, 0, 1]) + y = power(x, b) + assert_almost_equal(y, [0, 1.21, 1.04880884817, 1.21, 0.]) + assert_equal(y._mask, [1, 0, 0, 0, 1]) + b.mask = nomask + y = power(x, b) + assert_equal(y._mask, [1, 0, 0, 0, 1]) + z = x ** b + assert_equal(z._mask, y._mask) + assert_almost_equal(z, y) + assert_almost_equal(z._data, y._data) + x **= b + assert_equal(x._mask, y._mask) + assert_almost_equal(x, y) + assert_almost_equal(x._data, y._data) + + def test_power_with_broadcasting(self): + # Test power w/ broadcasting + a2 = np.array([[1., 2., 3.], [4., 5., 6.]]) + a2m = array(a2, mask=[[1, 0, 0], [0, 0, 1]]) + b1 = np.array([2, 4, 3]) + b2 = np.array([b1, b1]) + b2m = array(b2, mask=[[0, 1, 0], [0, 1, 0]]) + + ctrl = array([[1 ** 2, 2 ** 4, 3 ** 3], [4 ** 2, 5 ** 4, 6 ** 3]], + mask=[[1, 1, 0], [0, 1, 1]]) + # No broadcasting, base & exp w/ mask + test = a2m ** b2m + assert_equal(test, ctrl) + assert_equal(test.mask, ctrl.mask) + # No broadcasting, base w/ mask, exp w/o mask + test = a2m ** b2 + assert_equal(test, ctrl) + assert_equal(test.mask, a2m.mask) + # No broadcasting, base w/o mask, exp w/ mask + test = a2 ** b2m + assert_equal(test, ctrl) + assert_equal(test.mask, b2m.mask) + + ctrl = array([[2 ** 2, 4 ** 4, 3 ** 3], [2 ** 2, 4 ** 4, 3 ** 3]], + mask=[[0, 1, 0], [0, 1, 0]]) + test = b1 ** b2m + assert_equal(test, ctrl) + assert_equal(test.mask, ctrl.mask) + test = b2m ** b1 + assert_equal(test, ctrl) + assert_equal(test.mask, ctrl.mask) + + @pytest.mark.skipif(IS_WASM, reason="fp errors don't work in wasm") + def test_where(self): + # Test the where function + x = np.array([1., 1., 1., -2., pi/2.0, 4., 5., -10., 10., 1., 2., 3.]) + y = np.array([5., 0., 3., 2., -1., -4., 0., -10., 10., 1., 0., 3.]) + m1 = [1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0] + m2 = [0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 1] + xm = masked_array(x, mask=m1) + ym = masked_array(y, mask=m2) + xm.set_fill_value(1e+20) + + d = where(xm > 2, xm, -9) + assert_equal(d, [-9., -9., -9., -9., -9., 4., + -9., -9., 10., -9., -9., 3.]) + assert_equal(d._mask, xm._mask) + d = where(xm > 2, -9, ym) + assert_equal(d, [5., 0., 3., 2., -1., -9., + -9., -10., -9., 1., 0., -9.]) + assert_equal(d._mask, [1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0]) + d = where(xm > 2, xm, masked) + assert_equal(d, [-9., -9., -9., -9., -9., 4., + -9., -9., 10., -9., -9., 3.]) + tmp = xm._mask.copy() + tmp[(xm <= 2).filled(True)] = True + assert_equal(d._mask, tmp) + + with np.errstate(invalid="warn"): + # The fill value is 1e20, it cannot be converted to `int`: + with pytest.warns(RuntimeWarning, match="invalid value"): + ixm = xm.astype(int) + d = where(ixm > 2, ixm, masked) + assert_equal(d, [-9, -9, -9, -9, -9, 4, -9, -9, 10, -9, -9, 3]) + assert_equal(d.dtype, ixm.dtype) + + def test_where_object(self): + a = np.array(None) + b = masked_array(None) + r = b.copy() + assert_equal(np.ma.where(True, a, a), r) + assert_equal(np.ma.where(True, b, b), r) + + def test_where_with_masked_choice(self): + x = arange(10) + x[3] = masked + c = x >= 8 + # Set False to masked + z = where(c, x, masked) + assert_(z.dtype is x.dtype) + assert_(z[3] is masked) + assert_(z[4] is masked) + assert_(z[7] is masked) + assert_(z[8] is not masked) + assert_(z[9] is not masked) + assert_equal(x, z) + # Set True to masked + z = where(c, masked, x) + assert_(z.dtype is x.dtype) + assert_(z[3] is masked) + assert_(z[4] is not masked) + assert_(z[7] is not masked) + assert_(z[8] is masked) + assert_(z[9] is masked) + + def test_where_with_masked_condition(self): + x = array([1., 2., 3., 4., 5.]) + c = array([1, 1, 1, 0, 0]) + x[2] = masked + z = where(c, x, -x) + assert_equal(z, [1., 2., 0., -4., -5]) + c[0] = masked + z = where(c, x, -x) + assert_equal(z, [1., 2., 0., -4., -5]) + assert_(z[0] is masked) + assert_(z[1] is not masked) + assert_(z[2] is masked) + + x = arange(1, 6) + x[-1] = masked + y = arange(1, 6) * 10 + y[2] = masked + c = array([1, 1, 1, 0, 0], mask=[1, 0, 0, 0, 0]) + cm = c.filled(1) + z = where(c, x, y) + zm = where(cm, x, y) + assert_equal(z, zm) + assert_(getmask(zm) is nomask) + assert_equal(zm, [1, 2, 3, 40, 50]) + z = where(c, masked, 1) + assert_equal(z, [99, 99, 99, 1, 1]) + z = where(c, 1, masked) + assert_equal(z, [99, 1, 1, 99, 99]) + + def test_where_type(self): + # Test the type conservation with where + x = np.arange(4, dtype=np.int32) + y = np.arange(4, dtype=np.float32) * 2.2 + test = where(x > 1.5, y, x).dtype + control = np.result_type(np.int32, np.float32) + assert_equal(test, control) + + def test_where_broadcast(self): + # Issue 8599 + x = np.arange(9).reshape(3, 3) + y = np.zeros(3) + core = np.where([1, 0, 1], x, y) + ma = where([1, 0, 1], x, y) + + assert_equal(core, ma) + assert_equal(core.dtype, ma.dtype) + + def test_where_structured(self): + # Issue 8600 + dt = np.dtype([('a', int), ('b', int)]) + x = np.array([(1, 2), (3, 4), (5, 6)], dtype=dt) + y = np.array((10, 20), dtype=dt) + core = np.where([0, 1, 1], x, y) + ma = np.where([0, 1, 1], x, y) + + assert_equal(core, ma) + assert_equal(core.dtype, ma.dtype) + + def test_where_structured_masked(self): + dt = np.dtype([('a', int), ('b', int)]) + x = np.array([(1, 2), (3, 4), (5, 6)], dtype=dt) + + ma = where([0, 1, 1], x, masked) + expected = masked_where([1, 0, 0], x) + + assert_equal(ma.dtype, expected.dtype) + assert_equal(ma, expected) + assert_equal(ma.mask, expected.mask) + + def test_masked_invalid_error(self): + a = np.arange(5, dtype=object) + a[3] = np.PINF + a[2] = np.NaN + with pytest.raises(TypeError, + match="not supported for the input types"): + np.ma.masked_invalid(a) + + def test_masked_invalid_pandas(self): + # getdata() used to be bad for pandas series due to its _data + # attribute. This test is a regression test mainly and may be + # removed if getdata() is adjusted. + class Series(): + _data = "nonsense" + + def __array__(self): + return np.array([5, np.nan, np.inf]) + + arr = np.ma.masked_invalid(Series()) + assert_array_equal(arr._data, np.array(Series())) + assert_array_equal(arr._mask, [False, True, True]) + + @pytest.mark.parametrize("copy", [True, False]) + def test_masked_invalid_full_mask(self, copy): + # Matplotlib relied on masked_invalid always returning a full mask + # (Also astropy projects, but were ok with it gh-22720 and gh-22842) + a = np.ma.array([1, 2, 3, 4]) + assert a._mask is nomask + res = np.ma.masked_invalid(a, copy=copy) + assert res.mask is not nomask + # mask of a should not be mutated + assert a.mask is nomask + assert np.may_share_memory(a._data, res._data) != copy + + def test_choose(self): + # Test choose + choices = [[0, 1, 2, 3], [10, 11, 12, 13], + [20, 21, 22, 23], [30, 31, 32, 33]] + chosen = choose([2, 3, 1, 0], choices) + assert_equal(chosen, array([20, 31, 12, 3])) + chosen = choose([2, 4, 1, 0], choices, mode='clip') + assert_equal(chosen, array([20, 31, 12, 3])) + chosen = choose([2, 4, 1, 0], choices, mode='wrap') + assert_equal(chosen, array([20, 1, 12, 3])) + # Check with some masked indices + indices_ = array([2, 4, 1, 0], mask=[1, 0, 0, 1]) + chosen = choose(indices_, choices, mode='wrap') + assert_equal(chosen, array([99, 1, 12, 99])) + assert_equal(chosen.mask, [1, 0, 0, 1]) + # Check with some masked choices + choices = array(choices, mask=[[0, 0, 0, 1], [1, 1, 0, 1], + [1, 0, 0, 0], [0, 0, 0, 0]]) + indices_ = [2, 3, 1, 0] + chosen = choose(indices_, choices, mode='wrap') + assert_equal(chosen, array([20, 31, 12, 3])) + assert_equal(chosen.mask, [1, 0, 0, 1]) + + def test_choose_with_out(self): + # Test choose with an explicit out keyword + choices = [[0, 1, 2, 3], [10, 11, 12, 13], + [20, 21, 22, 23], [30, 31, 32, 33]] + store = empty(4, dtype=int) + chosen = choose([2, 3, 1, 0], choices, out=store) + assert_equal(store, array([20, 31, 12, 3])) + assert_(store is chosen) + # Check with some masked indices + out + store = empty(4, dtype=int) + indices_ = array([2, 3, 1, 0], mask=[1, 0, 0, 1]) + chosen = choose(indices_, choices, mode='wrap', out=store) + assert_equal(store, array([99, 31, 12, 99])) + assert_equal(store.mask, [1, 0, 0, 1]) + # Check with some masked choices + out ina ndarray ! + choices = array(choices, mask=[[0, 0, 0, 1], [1, 1, 0, 1], + [1, 0, 0, 0], [0, 0, 0, 0]]) + indices_ = [2, 3, 1, 0] + store = empty(4, dtype=int).view(ndarray) + chosen = choose(indices_, choices, mode='wrap', out=store) + assert_equal(store, array([999999, 31, 12, 999999])) + + def test_reshape(self): + a = arange(10) + a[0] = masked + # Try the default + b = a.reshape((5, 2)) + assert_equal(b.shape, (5, 2)) + assert_(b.flags['C']) + # Try w/ arguments as list instead of tuple + b = a.reshape(5, 2) + assert_equal(b.shape, (5, 2)) + assert_(b.flags['C']) + # Try w/ order + b = a.reshape((5, 2), order='F') + assert_equal(b.shape, (5, 2)) + assert_(b.flags['F']) + # Try w/ order + b = a.reshape(5, 2, order='F') + assert_equal(b.shape, (5, 2)) + assert_(b.flags['F']) + + c = np.reshape(a, (2, 5)) + assert_(isinstance(c, MaskedArray)) + assert_equal(c.shape, (2, 5)) + assert_(c[0, 0] is masked) + assert_(c.flags['C']) + + def test_make_mask_descr(self): + # Flexible + ntype = [('a', float), ('b', float)] + test = make_mask_descr(ntype) + assert_equal(test, [('a', bool), ('b', bool)]) + assert_(test is make_mask_descr(test)) + + # Standard w/ shape + ntype = (float, 2) + test = make_mask_descr(ntype) + assert_equal(test, (bool, 2)) + assert_(test is make_mask_descr(test)) + + # Standard standard + ntype = float + test = make_mask_descr(ntype) + assert_equal(test, np.dtype(bool)) + assert_(test is make_mask_descr(test)) + + # Nested + ntype = [('a', float), ('b', [('ba', float), ('bb', float)])] + test = make_mask_descr(ntype) + control = np.dtype([('a', 'b1'), ('b', [('ba', 'b1'), ('bb', 'b1')])]) + assert_equal(test, control) + assert_(test is make_mask_descr(test)) + + # Named+ shape + ntype = [('a', (float, 2))] + test = make_mask_descr(ntype) + assert_equal(test, np.dtype([('a', (bool, 2))])) + assert_(test is make_mask_descr(test)) + + # 2 names + ntype = [(('A', 'a'), float)] + test = make_mask_descr(ntype) + assert_equal(test, np.dtype([(('A', 'a'), bool)])) + assert_(test is make_mask_descr(test)) + + # nested boolean types should preserve identity + base_type = np.dtype([('a', int, 3)]) + base_mtype = make_mask_descr(base_type) + sub_type = np.dtype([('a', int), ('b', base_mtype)]) + test = make_mask_descr(sub_type) + assert_equal(test, np.dtype([('a', bool), ('b', [('a', bool, 3)])])) + assert_(test.fields['b'][0] is base_mtype) + + def test_make_mask(self): + # Test make_mask + # w/ a list as an input + mask = [0, 1] + test = make_mask(mask) + assert_equal(test.dtype, MaskType) + assert_equal(test, [0, 1]) + # w/ a ndarray as an input + mask = np.array([0, 1], dtype=bool) + test = make_mask(mask) + assert_equal(test.dtype, MaskType) + assert_equal(test, [0, 1]) + # w/ a flexible-type ndarray as an input - use default + mdtype = [('a', bool), ('b', bool)] + mask = np.array([(0, 0), (0, 1)], dtype=mdtype) + test = make_mask(mask) + assert_equal(test.dtype, MaskType) + assert_equal(test, [1, 1]) + # w/ a flexible-type ndarray as an input - use input dtype + mdtype = [('a', bool), ('b', bool)] + mask = np.array([(0, 0), (0, 1)], dtype=mdtype) + test = make_mask(mask, dtype=mask.dtype) + assert_equal(test.dtype, mdtype) + assert_equal(test, mask) + # w/ a flexible-type ndarray as an input - use input dtype + mdtype = [('a', float), ('b', float)] + bdtype = [('a', bool), ('b', bool)] + mask = np.array([(0, 0), (0, 1)], dtype=mdtype) + test = make_mask(mask, dtype=mask.dtype) + assert_equal(test.dtype, bdtype) + assert_equal(test, np.array([(0, 0), (0, 1)], dtype=bdtype)) + # Ensure this also works for void + mask = np.array((False, True), dtype='?,?')[()] + assert_(isinstance(mask, np.void)) + test = make_mask(mask, dtype=mask.dtype) + assert_equal(test, mask) + assert_(test is not mask) + mask = np.array((0, 1), dtype='i4,i4')[()] + test2 = make_mask(mask, dtype=mask.dtype) + assert_equal(test2, test) + # test that nomask is returned when m is nomask. + bools = [True, False] + dtypes = [MaskType, float] + msgformat = 'copy=%s, shrink=%s, dtype=%s' + for cpy, shr, dt in itertools.product(bools, bools, dtypes): + res = make_mask(nomask, copy=cpy, shrink=shr, dtype=dt) + assert_(res is nomask, msgformat % (cpy, shr, dt)) + + def test_mask_or(self): + # Initialize + mtype = [('a', bool), ('b', bool)] + mask = np.array([(0, 0), (0, 1), (1, 0), (0, 0)], dtype=mtype) + # Test using nomask as input + test = mask_or(mask, nomask) + assert_equal(test, mask) + test = mask_or(nomask, mask) + assert_equal(test, mask) + # Using False as input + test = mask_or(mask, False) + assert_equal(test, mask) + # Using another array w / the same dtype + other = np.array([(0, 1), (0, 1), (0, 1), (0, 1)], dtype=mtype) + test = mask_or(mask, other) + control = np.array([(0, 1), (0, 1), (1, 1), (0, 1)], dtype=mtype) + assert_equal(test, control) + # Using another array w / a different dtype + othertype = [('A', bool), ('B', bool)] + other = np.array([(0, 1), (0, 1), (0, 1), (0, 1)], dtype=othertype) + try: + test = mask_or(mask, other) + except ValueError: + pass + # Using nested arrays + dtype = [('a', bool), ('b', [('ba', bool), ('bb', bool)])] + amask = np.array([(0, (1, 0)), (0, (1, 0))], dtype=dtype) + bmask = np.array([(1, (0, 1)), (0, (0, 0))], dtype=dtype) + cntrl = np.array([(1, (1, 1)), (0, (1, 0))], dtype=dtype) + assert_equal(mask_or(amask, bmask), cntrl) + + def test_flatten_mask(self): + # Tests flatten mask + # Standard dtype + mask = np.array([0, 0, 1], dtype=bool) + assert_equal(flatten_mask(mask), mask) + # Flexible dtype + mask = np.array([(0, 0), (0, 1)], dtype=[('a', bool), ('b', bool)]) + test = flatten_mask(mask) + control = np.array([0, 0, 0, 1], dtype=bool) + assert_equal(test, control) + + mdtype = [('a', bool), ('b', [('ba', bool), ('bb', bool)])] + data = [(0, (0, 0)), (0, (0, 1))] + mask = np.array(data, dtype=mdtype) + test = flatten_mask(mask) + control = np.array([0, 0, 0, 0, 0, 1], dtype=bool) + assert_equal(test, control) + + def test_on_ndarray(self): + # Test functions on ndarrays + a = np.array([1, 2, 3, 4]) + m = array(a, mask=False) + test = anom(a) + assert_equal(test, m.anom()) + test = reshape(a, (2, 2)) + assert_equal(test, m.reshape(2, 2)) + + def test_compress(self): + # Test compress function on ndarray and masked array + # Address Github #2495. + arr = np.arange(8) + arr.shape = 4, 2 + cond = np.array([True, False, True, True]) + control = arr[[0, 2, 3]] + test = np.ma.compress(cond, arr, axis=0) + assert_equal(test, control) + marr = np.ma.array(arr) + test = np.ma.compress(cond, marr, axis=0) + assert_equal(test, control) + + def test_compressed(self): + # Test ma.compressed function. + # Address gh-4026 + a = np.ma.array([1, 2]) + test = np.ma.compressed(a) + assert_(type(test) is np.ndarray) + + # Test case when input data is ndarray subclass + class A(np.ndarray): + pass + + a = np.ma.array(A(shape=0)) + test = np.ma.compressed(a) + assert_(type(test) is A) + + # Test that compress flattens + test = np.ma.compressed([[1],[2]]) + assert_equal(test.ndim, 1) + test = np.ma.compressed([[[[[1]]]]]) + assert_equal(test.ndim, 1) + + # Test case when input is MaskedArray subclass + class M(MaskedArray): + pass + + test = np.ma.compressed(M([[[]], [[]]])) + assert_equal(test.ndim, 1) + + # with .compressed() overridden + class M(MaskedArray): + def compressed(self): + return 42 + + test = np.ma.compressed(M([[[]], [[]]])) + assert_equal(test, 42) + + def test_convolve(self): + a = masked_equal(np.arange(5), 2) + b = np.array([1, 1]) + test = np.ma.convolve(a, b) + assert_equal(test, masked_equal([0, 1, -1, -1, 7, 4], -1)) + + test = np.ma.convolve(a, b, propagate_mask=False) + assert_equal(test, masked_equal([0, 1, 1, 3, 7, 4], -1)) + + test = np.ma.convolve([1, 1], [1, 1, 1]) + assert_equal(test, masked_equal([1, 2, 2, 1], -1)) + + a = [1, 1] + b = masked_equal([1, -1, -1, 1], -1) + test = np.ma.convolve(a, b, propagate_mask=False) + assert_equal(test, masked_equal([1, 1, -1, 1, 1], -1)) + test = np.ma.convolve(a, b, propagate_mask=True) + assert_equal(test, masked_equal([-1, -1, -1, -1, -1], -1)) + + +class TestMaskedFields: + + def setup_method(self): + ilist = [1, 2, 3, 4, 5] + flist = [1.1, 2.2, 3.3, 4.4, 5.5] + slist = ['one', 'two', 'three', 'four', 'five'] + ddtype = [('a', int), ('b', float), ('c', '|S8')] + mdtype = [('a', bool), ('b', bool), ('c', bool)] + mask = [0, 1, 0, 0, 1] + base = array(list(zip(ilist, flist, slist)), mask=mask, dtype=ddtype) + self.data = dict(base=base, mask=mask, ddtype=ddtype, mdtype=mdtype) + + def test_set_records_masks(self): + base = self.data['base'] + mdtype = self.data['mdtype'] + # Set w/ nomask or masked + base.mask = nomask + assert_equal_records(base._mask, np.zeros(base.shape, dtype=mdtype)) + base.mask = masked + assert_equal_records(base._mask, np.ones(base.shape, dtype=mdtype)) + # Set w/ simple boolean + base.mask = False + assert_equal_records(base._mask, np.zeros(base.shape, dtype=mdtype)) + base.mask = True + assert_equal_records(base._mask, np.ones(base.shape, dtype=mdtype)) + # Set w/ list + base.mask = [0, 0, 0, 1, 1] + assert_equal_records(base._mask, + np.array([(x, x, x) for x in [0, 0, 0, 1, 1]], + dtype=mdtype)) + + def test_set_record_element(self): + # Check setting an element of a record) + base = self.data['base'] + (base_a, base_b, base_c) = (base['a'], base['b'], base['c']) + base[0] = (pi, pi, 'pi') + + assert_equal(base_a.dtype, int) + assert_equal(base_a._data, [3, 2, 3, 4, 5]) + + assert_equal(base_b.dtype, float) + assert_equal(base_b._data, [pi, 2.2, 3.3, 4.4, 5.5]) + + assert_equal(base_c.dtype, '|S8') + assert_equal(base_c._data, + [b'pi', b'two', b'three', b'four', b'five']) + + def test_set_record_slice(self): + base = self.data['base'] + (base_a, base_b, base_c) = (base['a'], base['b'], base['c']) + base[:3] = (pi, pi, 'pi') + + assert_equal(base_a.dtype, int) + assert_equal(base_a._data, [3, 3, 3, 4, 5]) + + assert_equal(base_b.dtype, float) + assert_equal(base_b._data, [pi, pi, pi, 4.4, 5.5]) + + assert_equal(base_c.dtype, '|S8') + assert_equal(base_c._data, + [b'pi', b'pi', b'pi', b'four', b'five']) + + def test_mask_element(self): + "Check record access" + base = self.data['base'] + base[0] = masked + + for n in ('a', 'b', 'c'): + assert_equal(base[n].mask, [1, 1, 0, 0, 1]) + assert_equal(base[n]._data, base._data[n]) + + def test_getmaskarray(self): + # Test getmaskarray on flexible dtype + ndtype = [('a', int), ('b', float)] + test = empty(3, dtype=ndtype) + assert_equal(getmaskarray(test), + np.array([(0, 0), (0, 0), (0, 0)], + dtype=[('a', '|b1'), ('b', '|b1')])) + test[:] = masked + assert_equal(getmaskarray(test), + np.array([(1, 1), (1, 1), (1, 1)], + dtype=[('a', '|b1'), ('b', '|b1')])) + + def test_view(self): + # Test view w/ flexible dtype + iterator = list(zip(np.arange(10), np.random.rand(10))) + data = np.array(iterator) + a = array(iterator, dtype=[('a', float), ('b', float)]) + a.mask[0] = (1, 0) + controlmask = np.array([1] + 19 * [0], dtype=bool) + # Transform globally to simple dtype + test = a.view(float) + assert_equal(test, data.ravel()) + assert_equal(test.mask, controlmask) + # Transform globally to dty + test = a.view((float, 2)) + assert_equal(test, data) + assert_equal(test.mask, controlmask.reshape(-1, 2)) + + def test_getitem(self): + ndtype = [('a', float), ('b', float)] + a = array(list(zip(np.random.rand(10), np.arange(10))), dtype=ndtype) + a.mask = np.array(list(zip([0, 0, 0, 0, 0, 0, 0, 0, 1, 1], + [1, 0, 0, 0, 0, 0, 0, 0, 1, 0])), + dtype=[('a', bool), ('b', bool)]) + + def _test_index(i): + assert_equal(type(a[i]), mvoid) + assert_equal_records(a[i]._data, a._data[i]) + assert_equal_records(a[i]._mask, a._mask[i]) + + assert_equal(type(a[i, ...]), MaskedArray) + assert_equal_records(a[i,...]._data, a._data[i,...]) + assert_equal_records(a[i,...]._mask, a._mask[i,...]) + + _test_index(1) # No mask + _test_index(0) # One element masked + _test_index(-2) # All element masked + + def test_setitem(self): + # Issue 4866: check that one can set individual items in [record][col] + # and [col][record] order + ndtype = np.dtype([('a', float), ('b', int)]) + ma = np.ma.MaskedArray([(1.0, 1), (2.0, 2)], dtype=ndtype) + ma['a'][1] = 3.0 + assert_equal(ma['a'], np.array([1.0, 3.0])) + ma[1]['a'] = 4.0 + assert_equal(ma['a'], np.array([1.0, 4.0])) + # Issue 2403 + mdtype = np.dtype([('a', bool), ('b', bool)]) + # soft mask + control = np.array([(False, True), (True, True)], dtype=mdtype) + a = np.ma.masked_all((2,), dtype=ndtype) + a['a'][0] = 2 + assert_equal(a.mask, control) + a = np.ma.masked_all((2,), dtype=ndtype) + a[0]['a'] = 2 + assert_equal(a.mask, control) + # hard mask + control = np.array([(True, True), (True, True)], dtype=mdtype) + a = np.ma.masked_all((2,), dtype=ndtype) + a.harden_mask() + a['a'][0] = 2 + assert_equal(a.mask, control) + a = np.ma.masked_all((2,), dtype=ndtype) + a.harden_mask() + a[0]['a'] = 2 + assert_equal(a.mask, control) + + def test_setitem_scalar(self): + # 8510 + mask_0d = np.ma.masked_array(1, mask=True) + arr = np.ma.arange(3) + arr[0] = mask_0d + assert_array_equal(arr.mask, [True, False, False]) + + def test_element_len(self): + # check that len() works for mvoid (Github issue #576) + for rec in self.data['base']: + assert_equal(len(rec), len(self.data['ddtype'])) + + +class TestMaskedObjectArray: + + def test_getitem(self): + arr = np.ma.array([None, None]) + for dt in [float, object]: + a0 = np.eye(2).astype(dt) + a1 = np.eye(3).astype(dt) + arr[0] = a0 + arr[1] = a1 + + assert_(arr[0] is a0) + assert_(arr[1] is a1) + assert_(isinstance(arr[0,...], MaskedArray)) + assert_(isinstance(arr[1,...], MaskedArray)) + assert_(arr[0,...][()] is a0) + assert_(arr[1,...][()] is a1) + + arr[0] = np.ma.masked + + assert_(arr[1] is a1) + assert_(isinstance(arr[0,...], MaskedArray)) + assert_(isinstance(arr[1,...], MaskedArray)) + assert_equal(arr[0,...].mask, True) + assert_(arr[1,...][()] is a1) + + # gh-5962 - object arrays of arrays do something special + assert_equal(arr[0].data, a0) + assert_equal(arr[0].mask, True) + assert_equal(arr[0,...][()].data, a0) + assert_equal(arr[0,...][()].mask, True) + + def test_nested_ma(self): + + arr = np.ma.array([None, None]) + # set the first object to be an unmasked masked constant. A little fiddly + arr[0,...] = np.array([np.ma.masked], object)[0,...] + + # check the above line did what we were aiming for + assert_(arr.data[0] is np.ma.masked) + + # test that getitem returned the value by identity + assert_(arr[0] is np.ma.masked) + + # now mask the masked value! + arr[0] = np.ma.masked + assert_(arr[0] is np.ma.masked) + + +class TestMaskedView: + + def setup_method(self): + iterator = list(zip(np.arange(10), np.random.rand(10))) + data = np.array(iterator) + a = array(iterator, dtype=[('a', float), ('b', float)]) + a.mask[0] = (1, 0) + controlmask = np.array([1] + 19 * [0], dtype=bool) + self.data = (data, a, controlmask) + + def test_view_to_nothing(self): + (data, a, controlmask) = self.data + test = a.view() + assert_(isinstance(test, MaskedArray)) + assert_equal(test._data, a._data) + assert_equal(test._mask, a._mask) + + def test_view_to_type(self): + (data, a, controlmask) = self.data + test = a.view(np.ndarray) + assert_(not isinstance(test, MaskedArray)) + assert_equal(test, a._data) + assert_equal_records(test, data.view(a.dtype).squeeze()) + + def test_view_to_simple_dtype(self): + (data, a, controlmask) = self.data + # View globally + test = a.view(float) + assert_(isinstance(test, MaskedArray)) + assert_equal(test, data.ravel()) + assert_equal(test.mask, controlmask) + + def test_view_to_flexible_dtype(self): + (data, a, controlmask) = self.data + + test = a.view([('A', float), ('B', float)]) + assert_equal(test.mask.dtype.names, ('A', 'B')) + assert_equal(test['A'], a['a']) + assert_equal(test['B'], a['b']) + + test = a[0].view([('A', float), ('B', float)]) + assert_(isinstance(test, MaskedArray)) + assert_equal(test.mask.dtype.names, ('A', 'B')) + assert_equal(test['A'], a['a'][0]) + assert_equal(test['B'], a['b'][0]) + + test = a[-1].view([('A', float), ('B', float)]) + assert_(isinstance(test, MaskedArray)) + assert_equal(test.dtype.names, ('A', 'B')) + assert_equal(test['A'], a['a'][-1]) + assert_equal(test['B'], a['b'][-1]) + + def test_view_to_subdtype(self): + (data, a, controlmask) = self.data + # View globally + test = a.view((float, 2)) + assert_(isinstance(test, MaskedArray)) + assert_equal(test, data) + assert_equal(test.mask, controlmask.reshape(-1, 2)) + # View on 1 masked element + test = a[0].view((float, 2)) + assert_(isinstance(test, MaskedArray)) + assert_equal(test, data[0]) + assert_equal(test.mask, (1, 0)) + # View on 1 unmasked element + test = a[-1].view((float, 2)) + assert_(isinstance(test, MaskedArray)) + assert_equal(test, data[-1]) + + def test_view_to_dtype_and_type(self): + (data, a, controlmask) = self.data + + test = a.view((float, 2), np.recarray) + assert_equal(test, data) + assert_(isinstance(test, np.recarray)) + assert_(not isinstance(test, MaskedArray)) + + +class TestOptionalArgs: + def test_ndarrayfuncs(self): + # test axis arg behaves the same as ndarray (including multiple axes) + + d = np.arange(24.0).reshape((2,3,4)) + m = np.zeros(24, dtype=bool).reshape((2,3,4)) + # mask out last element of last dimension + m[:,:,-1] = True + a = np.ma.array(d, mask=m) + + def testaxis(f, a, d): + numpy_f = numpy.__getattribute__(f) + ma_f = np.ma.__getattribute__(f) + + # test axis arg + assert_equal(ma_f(a, axis=1)[...,:-1], numpy_f(d[...,:-1], axis=1)) + assert_equal(ma_f(a, axis=(0,1))[...,:-1], + numpy_f(d[...,:-1], axis=(0,1))) + + def testkeepdims(f, a, d): + numpy_f = numpy.__getattribute__(f) + ma_f = np.ma.__getattribute__(f) + + # test keepdims arg + assert_equal(ma_f(a, keepdims=True).shape, + numpy_f(d, keepdims=True).shape) + assert_equal(ma_f(a, keepdims=False).shape, + numpy_f(d, keepdims=False).shape) + + # test both at once + assert_equal(ma_f(a, axis=1, keepdims=True)[...,:-1], + numpy_f(d[...,:-1], axis=1, keepdims=True)) + assert_equal(ma_f(a, axis=(0,1), keepdims=True)[...,:-1], + numpy_f(d[...,:-1], axis=(0,1), keepdims=True)) + + for f in ['sum', 'prod', 'mean', 'var', 'std']: + testaxis(f, a, d) + testkeepdims(f, a, d) + + for f in ['min', 'max']: + testaxis(f, a, d) + + d = (np.arange(24).reshape((2,3,4))%2 == 0) + a = np.ma.array(d, mask=m) + for f in ['all', 'any']: + testaxis(f, a, d) + testkeepdims(f, a, d) + + def test_count(self): + # test np.ma.count specially + + d = np.arange(24.0).reshape((2,3,4)) + m = np.zeros(24, dtype=bool).reshape((2,3,4)) + m[:,0,:] = True + a = np.ma.array(d, mask=m) + + assert_equal(count(a), 16) + assert_equal(count(a, axis=1), 2*ones((2,4))) + assert_equal(count(a, axis=(0,1)), 4*ones((4,))) + assert_equal(count(a, keepdims=True), 16*ones((1,1,1))) + assert_equal(count(a, axis=1, keepdims=True), 2*ones((2,1,4))) + assert_equal(count(a, axis=(0,1), keepdims=True), 4*ones((1,1,4))) + assert_equal(count(a, axis=-2), 2*ones((2,4))) + assert_raises(ValueError, count, a, axis=(1,1)) + assert_raises(np.AxisError, count, a, axis=3) + + # check the 'nomask' path + a = np.ma.array(d, mask=nomask) + + assert_equal(count(a), 24) + assert_equal(count(a, axis=1), 3*ones((2,4))) + assert_equal(count(a, axis=(0,1)), 6*ones((4,))) + assert_equal(count(a, keepdims=True), 24*ones((1,1,1))) + assert_equal(np.ndim(count(a, keepdims=True)), 3) + assert_equal(count(a, axis=1, keepdims=True), 3*ones((2,1,4))) + assert_equal(count(a, axis=(0,1), keepdims=True), 6*ones((1,1,4))) + assert_equal(count(a, axis=-2), 3*ones((2,4))) + assert_raises(ValueError, count, a, axis=(1,1)) + assert_raises(np.AxisError, count, a, axis=3) + + # check the 'masked' singleton + assert_equal(count(np.ma.masked), 0) + + # check 0-d arrays do not allow axis > 0 + assert_raises(np.AxisError, count, np.ma.array(1), axis=1) + + +class TestMaskedConstant: + def _do_add_test(self, add): + # sanity check + assert_(add(np.ma.masked, 1) is np.ma.masked) + + # now try with a vector + vector = np.array([1, 2, 3]) + result = add(np.ma.masked, vector) + + # lots of things could go wrong here + assert_(result is not np.ma.masked) + assert_(not isinstance(result, np.ma.core.MaskedConstant)) + assert_equal(result.shape, vector.shape) + assert_equal(np.ma.getmask(result), np.ones(vector.shape, dtype=bool)) + + def test_ufunc(self): + self._do_add_test(np.add) + + def test_operator(self): + self._do_add_test(lambda a, b: a + b) + + def test_ctor(self): + m = np.ma.array(np.ma.masked) + + # most importantly, we do not want to create a new MaskedConstant + # instance + assert_(not isinstance(m, np.ma.core.MaskedConstant)) + assert_(m is not np.ma.masked) + + def test_repr(self): + # copies should not exist, but if they do, it should be obvious that + # something is wrong + assert_equal(repr(np.ma.masked), 'masked') + + # create a new instance in a weird way + masked2 = np.ma.MaskedArray.__new__(np.ma.core.MaskedConstant) + assert_not_equal(repr(masked2), 'masked') + + def test_pickle(self): + from io import BytesIO + + for proto in range(2, pickle.HIGHEST_PROTOCOL + 1): + with BytesIO() as f: + pickle.dump(np.ma.masked, f, protocol=proto) + f.seek(0) + res = pickle.load(f) + assert_(res is np.ma.masked) + + def test_copy(self): + # gh-9328 + # copy is a no-op, like it is with np.True_ + assert_equal( + np.ma.masked.copy() is np.ma.masked, + np.True_.copy() is np.True_) + + def test__copy(self): + import copy + assert_( + copy.copy(np.ma.masked) is np.ma.masked) + + def test_deepcopy(self): + import copy + assert_( + copy.deepcopy(np.ma.masked) is np.ma.masked) + + def test_immutable(self): + orig = np.ma.masked + assert_raises(np.ma.core.MaskError, operator.setitem, orig, (), 1) + assert_raises(ValueError,operator.setitem, orig.data, (), 1) + assert_raises(ValueError, operator.setitem, orig.mask, (), False) + + view = np.ma.masked.view(np.ma.MaskedArray) + assert_raises(ValueError, operator.setitem, view, (), 1) + assert_raises(ValueError, operator.setitem, view.data, (), 1) + assert_raises(ValueError, operator.setitem, view.mask, (), False) + + def test_coercion_int(self): + a_i = np.zeros((), int) + assert_raises(MaskError, operator.setitem, a_i, (), np.ma.masked) + assert_raises(MaskError, int, np.ma.masked) + + def test_coercion_float(self): + a_f = np.zeros((), float) + assert_warns(UserWarning, operator.setitem, a_f, (), np.ma.masked) + assert_(np.isnan(a_f[()])) + + @pytest.mark.xfail(reason="See gh-9750") + def test_coercion_unicode(self): + a_u = np.zeros((), 'U10') + a_u[()] = np.ma.masked + assert_equal(a_u[()], '--') + + @pytest.mark.xfail(reason="See gh-9750") + def test_coercion_bytes(self): + a_b = np.zeros((), 'S10') + a_b[()] = np.ma.masked + assert_equal(a_b[()], b'--') + + def test_subclass(self): + # https://github.com/astropy/astropy/issues/6645 + class Sub(type(np.ma.masked)): pass + + a = Sub() + assert_(a is Sub()) + assert_(a is not np.ma.masked) + assert_not_equal(repr(a), 'masked') + + def test_attributes_readonly(self): + assert_raises(AttributeError, setattr, np.ma.masked, 'shape', (1,)) + assert_raises(AttributeError, setattr, np.ma.masked, 'dtype', np.int64) + + +class TestMaskedWhereAliases: + + # TODO: Test masked_object, masked_equal, ... + + def test_masked_values(self): + res = masked_values(np.array([-32768.0]), np.int16(-32768)) + assert_equal(res.mask, [True]) + + res = masked_values(np.inf, np.inf) + assert_equal(res.mask, True) + + res = np.ma.masked_values(np.inf, -np.inf) + assert_equal(res.mask, False) + + res = np.ma.masked_values([1, 2, 3, 4], 5, shrink=True) + assert_(res.mask is np.ma.nomask) + + res = np.ma.masked_values([1, 2, 3, 4], 5, shrink=False) + assert_equal(res.mask, [False] * 4) + + +def test_masked_array(): + a = np.ma.array([0, 1, 2, 3], mask=[0, 0, 1, 0]) + assert_equal(np.argwhere(a), [[1], [3]]) + +def test_masked_array_no_copy(): + # check nomask array is updated in place + a = np.ma.array([1, 2, 3, 4]) + _ = np.ma.masked_where(a == 3, a, copy=False) + assert_array_equal(a.mask, [False, False, True, False]) + # check masked array is updated in place + a = np.ma.array([1, 2, 3, 4], mask=[1, 0, 0, 0]) + _ = np.ma.masked_where(a == 3, a, copy=False) + assert_array_equal(a.mask, [True, False, True, False]) + # check masked array with masked_invalid is updated in place + a = np.ma.array([np.inf, 1, 2, 3, 4]) + _ = np.ma.masked_invalid(a, copy=False) + assert_array_equal(a.mask, [True, False, False, False, False]) + +def test_append_masked_array(): + a = np.ma.masked_equal([1,2,3], value=2) + b = np.ma.masked_equal([4,3,2], value=2) + + result = np.ma.append(a, b) + expected_data = [1, 2, 3, 4, 3, 2] + expected_mask = [False, True, False, False, False, True] + assert_array_equal(result.data, expected_data) + assert_array_equal(result.mask, expected_mask) + + a = np.ma.masked_all((2,2)) + b = np.ma.ones((3,1)) + + result = np.ma.append(a, b) + expected_data = [1] * 3 + expected_mask = [True] * 4 + [False] * 3 + assert_array_equal(result.data[-3], expected_data) + assert_array_equal(result.mask, expected_mask) + + result = np.ma.append(a, b, axis=None) + assert_array_equal(result.data[-3], expected_data) + assert_array_equal(result.mask, expected_mask) + + +def test_append_masked_array_along_axis(): + a = np.ma.masked_equal([1,2,3], value=2) + b = np.ma.masked_values([[4, 5, 6], [7, 8, 9]], 7) + + # When `axis` is specified, `values` must have the correct shape. + assert_raises(ValueError, np.ma.append, a, b, axis=0) + + result = np.ma.append(a[np.newaxis,:], b, axis=0) + expected = np.ma.arange(1, 10) + expected[[1, 6]] = np.ma.masked + expected = expected.reshape((3,3)) + assert_array_equal(result.data, expected.data) + assert_array_equal(result.mask, expected.mask) + +def test_default_fill_value_complex(): + # regression test for Python 3, where 'unicode' was not defined + assert_(default_fill_value(1 + 1j) == 1.e20 + 0.0j) + + +def test_ufunc_with_output(): + # check that giving an output argument always returns that output. + # Regression test for gh-8416. + x = array([1., 2., 3.], mask=[0, 0, 1]) + y = np.add(x, 1., out=x) + assert_(y is x) + + +def test_ufunc_with_out_varied(): + """ Test that masked arrays are immune to gh-10459 """ + # the mask of the output should not affect the result, however it is passed + a = array([ 1, 2, 3], mask=[1, 0, 0]) + b = array([10, 20, 30], mask=[1, 0, 0]) + out = array([ 0, 0, 0], mask=[0, 0, 1]) + expected = array([11, 22, 33], mask=[1, 0, 0]) + + out_pos = out.copy() + res_pos = np.add(a, b, out_pos) + + out_kw = out.copy() + res_kw = np.add(a, b, out=out_kw) + + out_tup = out.copy() + res_tup = np.add(a, b, out=(out_tup,)) + + assert_equal(res_kw.mask, expected.mask) + assert_equal(res_kw.data, expected.data) + assert_equal(res_tup.mask, expected.mask) + assert_equal(res_tup.data, expected.data) + assert_equal(res_pos.mask, expected.mask) + assert_equal(res_pos.data, expected.data) + + +def test_astype_mask_ordering(): + descr = np.dtype([('v', int, 3), ('x', [('y', float)])]) + x = array([ + [([1, 2, 3], (1.0,)), ([1, 2, 3], (2.0,))], + [([1, 2, 3], (3.0,)), ([1, 2, 3], (4.0,))]], dtype=descr) + x[0]['v'][0] = np.ma.masked + + x_a = x.astype(descr) + assert x_a.dtype.names == np.dtype(descr).names + assert x_a.mask.dtype.names == np.dtype(descr).names + assert_equal(x, x_a) + + assert_(x is x.astype(x.dtype, copy=False)) + assert_equal(type(x.astype(x.dtype, subok=False)), np.ndarray) + + x_f = x.astype(x.dtype, order='F') + assert_(x_f.flags.f_contiguous) + assert_(x_f.mask.flags.f_contiguous) + + # Also test the same indirectly, via np.array + x_a2 = np.array(x, dtype=descr, subok=True) + assert x_a2.dtype.names == np.dtype(descr).names + assert x_a2.mask.dtype.names == np.dtype(descr).names + assert_equal(x, x_a2) + + assert_(x is np.array(x, dtype=descr, copy=False, subok=True)) + + x_f2 = np.array(x, dtype=x.dtype, order='F', subok=True) + assert_(x_f2.flags.f_contiguous) + assert_(x_f2.mask.flags.f_contiguous) + + +@pytest.mark.parametrize('dt1', num_dts, ids=num_ids) +@pytest.mark.parametrize('dt2', num_dts, ids=num_ids) +@pytest.mark.filterwarnings('ignore::numpy.ComplexWarning') +def test_astype_basic(dt1, dt2): + # See gh-12070 + src = np.ma.array(ones(3, dt1), fill_value=1) + dst = src.astype(dt2) + + assert_(src.fill_value == 1) + assert_(src.dtype == dt1) + assert_(src.fill_value.dtype == dt1) + + assert_(dst.fill_value == 1) + assert_(dst.dtype == dt2) + assert_(dst.fill_value.dtype == dt2) + + assert_equal(src, dst) + + +def test_fieldless_void(): + dt = np.dtype([]) # a void dtype with no fields + x = np.empty(4, dt) + + # these arrays contain no values, so there's little to test - but this + # shouldn't crash + mx = np.ma.array(x) + assert_equal(mx.dtype, x.dtype) + assert_equal(mx.shape, x.shape) + + mx = np.ma.array(x, mask=x) + assert_equal(mx.dtype, x.dtype) + assert_equal(mx.shape, x.shape) + + +def test_mask_shape_assignment_does_not_break_masked(): + a = np.ma.masked + b = np.ma.array(1, mask=a.mask) + b.shape = (1,) + assert_equal(a.mask.shape, ()) + +@pytest.mark.skipif(sys.flags.optimize > 1, + reason="no docstrings present to inspect when PYTHONOPTIMIZE/Py_OptimizeFlag > 1") +def test_doc_note(): + def method(self): + """This docstring + + Has multiple lines + + And notes + + Notes + ----- + original note + """ + pass + + expected_doc = """This docstring + +Has multiple lines + +And notes + +Notes +----- +note + +original note""" + + assert_equal(np.ma.core.doc_note(method.__doc__, "note"), expected_doc) + + +def test_gh_22556(): + source = np.ma.array([0, [0, 1, 2]], dtype=object) + deepcopy = copy.deepcopy(source) + deepcopy[1].append('this should not appear in source') + assert len(source[1]) == 3 + + +def test_gh_21022(): + # testing for absence of reported error + source = np.ma.masked_array(data=[-1, -1], mask=True, dtype=np.float64) + axis = np.array(0) + result = np.prod(source, axis=axis, keepdims=False) + result = np.ma.masked_array(result, + mask=np.ones(result.shape, dtype=np.bool_)) + array = np.ma.masked_array(data=-1, mask=True, dtype=np.float64) + copy.deepcopy(array) + copy.deepcopy(result) + + +def test_deepcopy_2d_obj(): + source = np.ma.array([[0, "dog"], + [1, 1], + [[1, 2], "cat"]], + mask=[[0, 1], + [0, 0], + [0, 0]], + dtype=object) + deepcopy = copy.deepcopy(source) + deepcopy[2, 0].extend(['this should not appear in source', 3]) + assert len(source[2, 0]) == 2 + assert len(deepcopy[2, 0]) == 4 + assert_equal(deepcopy._mask, source._mask) + deepcopy._mask[0, 0] = 1 + assert source._mask[0, 0] == 0 + + +def test_deepcopy_0d_obj(): + source = np.ma.array(0, mask=[0], dtype=object) + deepcopy = copy.deepcopy(source) + deepcopy[...] = 17 + assert_equal(source, 0) + assert_equal(deepcopy, 17) diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_deprecations.py b/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_deprecations.py new file mode 100644 index 0000000000000000000000000000000000000000..40c8418f5c1809130672dca46e8c43469692da09 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_deprecations.py @@ -0,0 +1,84 @@ +"""Test deprecation and future warnings. + +""" +import pytest +import numpy as np +from numpy.testing import assert_warns +from numpy.ma.testutils import assert_equal +from numpy.ma.core import MaskedArrayFutureWarning +import io +import textwrap + +class TestArgsort: + """ gh-8701 """ + def _test_base(self, argsort, cls): + arr_0d = np.array(1).view(cls) + argsort(arr_0d) + + arr_1d = np.array([1, 2, 3]).view(cls) + argsort(arr_1d) + + # argsort has a bad default for >1d arrays + arr_2d = np.array([[1, 2], [3, 4]]).view(cls) + result = assert_warns( + np.ma.core.MaskedArrayFutureWarning, argsort, arr_2d) + assert_equal(result, argsort(arr_2d, axis=None)) + + # should be no warnings for explicitly specifying it + argsort(arr_2d, axis=None) + argsort(arr_2d, axis=-1) + + def test_function_ndarray(self): + return self._test_base(np.ma.argsort, np.ndarray) + + def test_function_maskedarray(self): + return self._test_base(np.ma.argsort, np.ma.MaskedArray) + + def test_method(self): + return self._test_base(np.ma.MaskedArray.argsort, np.ma.MaskedArray) + + +class TestMinimumMaximum: + + def test_axis_default(self): + # NumPy 1.13, 2017-05-06 + + data1d = np.ma.arange(6) + data2d = data1d.reshape(2, 3) + + ma_min = np.ma.minimum.reduce + ma_max = np.ma.maximum.reduce + + # check that the default axis is still None, but warns on 2d arrays + result = assert_warns(MaskedArrayFutureWarning, ma_max, data2d) + assert_equal(result, ma_max(data2d, axis=None)) + + result = assert_warns(MaskedArrayFutureWarning, ma_min, data2d) + assert_equal(result, ma_min(data2d, axis=None)) + + # no warnings on 1d, as both new and old defaults are equivalent + result = ma_min(data1d) + assert_equal(result, ma_min(data1d, axis=None)) + assert_equal(result, ma_min(data1d, axis=0)) + + result = ma_max(data1d) + assert_equal(result, ma_max(data1d, axis=None)) + assert_equal(result, ma_max(data1d, axis=0)) + + +class TestFromtextfile: + def test_fromtextfile_delimitor(self): + # NumPy 1.22.0, 2021-09-23 + + textfile = io.StringIO(textwrap.dedent( + """ + A,B,C,D + 'string 1';1;1.0;'mixed column' + 'string 2';2;2.0; + 'string 3';3;3.0;123 + 'string 4';4;4.0;3.14 + """ + )) + + with pytest.warns(DeprecationWarning): + result = np.ma.mrecords.fromtextfile(textfile, delimitor=';') diff --git a/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_mrecords.py b/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_mrecords.py new file mode 100644 index 0000000000000000000000000000000000000000..77123c3cda941636354a7b282777f3f0e55d3ab0 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/numpy/ma/tests/test_mrecords.py @@ -0,0 +1,493 @@ +# pylint: disable-msg=W0611, W0612, W0511,R0201 +"""Tests suite for mrecords. + +:author: Pierre Gerard-Marchant +:contact: pierregm_at_uga_dot_edu + +""" +import numpy as np +import numpy.ma as ma +from numpy import recarray +from numpy.ma import masked, nomask +from numpy.testing import temppath +from numpy.core.records import ( + fromrecords as recfromrecords, fromarrays as recfromarrays + ) +from numpy.ma.mrecords import ( + MaskedRecords, mrecarray, fromarrays, fromtextfile, fromrecords, + addfield + ) +from numpy.ma.testutils import ( + assert_, assert_equal, + assert_equal_records, + ) +from numpy.compat import pickle + + +class TestMRecords: + + ilist = [1, 2, 3, 4, 5] + flist = [1.1, 2.2, 3.3, 4.4, 5.5] + slist = [b'one', b'two', b'three', b'four', b'five'] + ddtype = [('a', int), ('b', float), ('c', '|S8')] + mask = [0, 1, 0, 0, 1] + base = ma.array(list(zip(ilist, flist, slist)), mask=mask, dtype=ddtype) + + def test_byview(self): + # Test creation by view + base = self.base + mbase = base.view(mrecarray) + assert_equal(mbase.recordmask, base.recordmask) + assert_equal_records(mbase._mask, base._mask) + assert_(isinstance(mbase._data, recarray)) + assert_equal_records(mbase._data, base._data.view(recarray)) + for field in ('a', 'b', 'c'): + assert_equal(base[field], mbase[field]) + assert_equal_records(mbase.view(mrecarray), mbase) + + def test_get(self): + # Tests fields retrieval + base = self.base.copy() + mbase = base.view(mrecarray) + # As fields.......... + for field in ('a', 'b', 'c'): + assert_equal(getattr(mbase, field), mbase[field]) + assert_equal(base[field], mbase[field]) + # as elements ....... + mbase_first = mbase[0] + assert_(isinstance(mbase_first, mrecarray)) + assert_equal(mbase_first.dtype, mbase.dtype) + assert_equal(mbase_first.tolist(), (1, 1.1, b'one')) + # Used to be mask, now it's recordmask + assert_equal(mbase_first.recordmask, nomask) + assert_equal(mbase_first._mask.item(), (False, False, False)) + assert_equal(mbase_first['a'], mbase['a'][0]) + mbase_last = mbase[-1] + assert_(isinstance(mbase_last, mrecarray)) + assert_equal(mbase_last.dtype, mbase.dtype) + assert_equal(mbase_last.tolist(), (None, None, None)) + # Used to be mask, now it's recordmask + assert_equal(mbase_last.recordmask, True) + assert_equal(mbase_last._mask.item(), (True, True, True)) + assert_equal(mbase_last['a'], mbase['a'][-1]) + assert_((mbase_last['a'] is masked)) + # as slice .......... + mbase_sl = mbase[:2] + assert_(isinstance(mbase_sl, mrecarray)) + assert_equal(mbase_sl.dtype, mbase.dtype) + # Used to be mask, now it's recordmask + assert_equal(mbase_sl.recordmask, [0, 1]) + assert_equal_records(mbase_sl.mask, + np.array([(False, False, False), + (True, True, True)], + dtype=mbase._mask.dtype)) + assert_equal_records(mbase_sl, base[:2].view(mrecarray)) + for field in ('a', 'b', 'c'): + assert_equal(getattr(mbase_sl, field), base[:2][field]) + + def test_set_fields(self): + # Tests setting fields. + base = self.base.copy() + mbase = base.view(mrecarray) + mbase = mbase.copy() + mbase.fill_value = (999999, 1e20, 'N/A') + # Change the data, the mask should be conserved + mbase.a._data[:] = 5 + assert_equal(mbase['a']._data, [5, 5, 5, 5, 5]) + assert_equal(mbase['a']._mask, [0, 1, 0, 0, 1]) + # Change the elements, and the mask will follow + mbase.a = 1 + assert_equal(mbase['a']._data, [1]*5) + assert_equal(ma.getmaskarray(mbase['a']), [0]*5) + # Use to be _mask, now it's recordmask + assert_equal(mbase.recordmask, [False]*5) + assert_equal(mbase._mask.tolist(), + np.array([(0, 0, 0), + (0, 1, 1), + (0, 0, 0), + (0, 0, 0), + (0, 1, 1)], + dtype=bool)) + # Set a field to mask ........................ + mbase.c = masked + # Use to be mask, and now it's still mask ! + assert_equal(mbase.c.mask, [1]*5) + assert_equal(mbase.c.recordmask, [1]*5) + assert_equal(ma.getmaskarray(mbase['c']), [1]*5) + assert_equal(ma.getdata(mbase['c']), [b'N/A']*5) + assert_equal(mbase._mask.tolist(), + np.array([(0, 0, 1), + (0, 1, 1), + (0, 0, 1), + (0, 0, 1), + (0, 1, 1)], + dtype=bool)) + # Set fields by slices ....................... + mbase = base.view(mrecarray).copy() + mbase.a[3:] = 5 + assert_equal(mbase.a, [1, 2, 3, 5, 5]) + assert_equal(mbase.a._mask, [0, 1, 0, 0, 0]) + mbase.b[3:] = masked + assert_equal(mbase.b, base['b']) + assert_equal(mbase.b._mask, [0, 1, 0, 1, 1]) + # Set fields globally.......................... + ndtype = [('alpha', '|S1'), ('num', int)] + data = ma.array([('a', 1), ('b', 2), ('c', 3)], dtype=ndtype) + rdata = data.view(MaskedRecords) + val = ma.array([10, 20, 30], mask=[1, 0, 0]) + + rdata['num'] = val + assert_equal(rdata.num, val) + assert_equal(rdata.num.mask, [1, 0, 0]) + + def test_set_fields_mask(self): + # Tests setting the mask of a field. + base = self.base.copy() + # This one has already a mask.... + mbase = base.view(mrecarray) + mbase['a'][-2] = masked + assert_equal(mbase.a, [1, 2, 3, 4, 5]) + assert_equal(mbase.a._mask, [0, 1, 0, 1, 1]) + # This one has not yet + mbase = fromarrays([np.arange(5), np.random.rand(5)], + dtype=[('a', int), ('b', float)]) + mbase['a'][-2] = masked + assert_equal(mbase.a, [0, 1, 2, 3, 4]) + assert_equal(mbase.a._mask, [0, 0, 0, 1, 0]) + + def test_set_mask(self): + base = self.base.copy() + mbase = base.view(mrecarray) + # Set the mask to True ....................... + mbase.mask = masked + assert_equal(ma.getmaskarray(mbase['b']), [1]*5) + assert_equal(mbase['a']._mask, mbase['b']._mask) + assert_equal(mbase['a']._mask, mbase['c']._mask) + assert_equal(mbase._mask.tolist(), + np.array([(1, 1, 1)]*5, dtype=bool)) + # Delete the mask ............................ + mbase.mask = nomask + assert_equal(ma.getmaskarray(mbase['c']), [0]*5) + assert_equal(mbase._mask.tolist(), + np.array([(0, 0, 0)]*5, dtype=bool)) + + def test_set_mask_fromarray(self): + base = self.base.copy() + mbase = base.view(mrecarray) + # Sets the mask w/ an array + mbase.mask = [1, 0, 0, 0, 1] + assert_equal(mbase.a.mask, [1, 0, 0, 0, 1]) + assert_equal(mbase.b.mask, [1, 0, 0, 0, 1]) + assert_equal(mbase.c.mask, [1, 0, 0, 0, 1]) + # Yay, once more ! + mbase.mask = [0, 0, 0, 0, 1] + assert_equal(mbase.a.mask, [0, 0, 0, 0, 1]) + assert_equal(mbase.b.mask, [0, 0, 0, 0, 1]) + assert_equal(mbase.c.mask, [0, 0, 0, 0, 1]) + + def test_set_mask_fromfields(self): + mbase = self.base.copy().view(mrecarray) + + nmask = np.array( + [(0, 1, 0), (0, 1, 0), (1, 0, 1), (1, 0, 1), (0, 0, 0)], + dtype=[('a', bool), ('b', bool), ('c', bool)]) + mbase.mask = nmask + assert_equal(mbase.a.mask, [0, 0, 1, 1, 0]) + assert_equal(mbase.b.mask, [1, 1, 0, 0, 0]) + assert_equal(mbase.c.mask, [0, 0, 1, 1, 0]) + # Reinitialize and redo + mbase.mask = False + mbase.fieldmask = nmask + assert_equal(mbase.a.mask, [0, 0, 1, 1, 0]) + assert_equal(mbase.b.mask, [1, 1, 0, 0, 0]) + assert_equal(mbase.c.mask, [0, 0, 1, 1, 0]) + + def test_set_elements(self): + base = self.base.copy() + # Set an element to mask ..................... + mbase = base.view(mrecarray).copy() + mbase[-2] = masked + assert_equal( + mbase._mask.tolist(), + np.array([(0, 0, 0), (1, 1, 1), (0, 0, 0), (1, 1, 1), (1, 1, 1)], + dtype=bool)) + # Used to be mask, now it's recordmask! + assert_equal(mbase.recordmask, [0, 1, 0, 1, 1]) + # Set slices ................................. + mbase = base.view(mrecarray).copy() + mbase[:2] = (5, 5, 5) + assert_equal(mbase.a._data, [5, 5, 3, 4, 5]) + assert_equal(mbase.a._mask, [0, 0, 0, 0, 1]) + assert_equal(mbase.b._data, [5., 5., 3.3, 4.4, 5.5]) + assert_equal(mbase.b._mask, [0, 0, 0, 0, 1]) + assert_equal(mbase.c._data, + [b'5', b'5', b'three', b'four', b'five']) + assert_equal(mbase.b._mask, [0, 0, 0, 0, 1]) + + mbase = base.view(mrecarray).copy() + mbase[:2] = masked + assert_equal(mbase.a._data, [1, 2, 3, 4, 5]) + assert_equal(mbase.a._mask, [1, 1, 0, 0, 1]) + assert_equal(mbase.b._data, [1.1, 2.2, 3.3, 4.4, 5.5]) + assert_equal(mbase.b._mask, [1, 1, 0, 0, 1]) + assert_equal(mbase.c._data, + [b'one', b'two', b'three', b'four', b'five']) + assert_equal(mbase.b._mask, [1, 1, 0, 0, 1]) + + def test_setslices_hardmask(self): + # Tests setting slices w/ hardmask. + base = self.base.copy() + mbase = base.view(mrecarray) + mbase.harden_mask() + try: + mbase[-2:] = (5, 5, 5) + assert_equal(mbase.a._data, [1, 2, 3, 5, 5]) + assert_equal(mbase.b._data, [1.1, 2.2, 3.3, 5, 5.5]) + assert_equal(mbase.c._data, + [b'one', b'two', b'three', b'5', b'five']) + assert_equal(mbase.a._mask, [0, 1, 0, 0, 1]) + assert_equal(mbase.b._mask, mbase.a._mask) + assert_equal(mbase.b._mask, mbase.c._mask) + except NotImplementedError: + # OK, not implemented yet... + pass + except AssertionError: + raise + else: + raise Exception("Flexible hard masks should be supported !") + # Not using a tuple should crash + try: + mbase[-2:] = 3 + except (NotImplementedError, TypeError): + pass + else: + raise TypeError("Should have expected a readable buffer object!") + + def test_hardmask(self): + # Test hardmask + base = self.base.copy() + mbase = base.view(mrecarray) + mbase.harden_mask() + assert_(mbase._hardmask) + mbase.mask = nomask + assert_equal_records(mbase._mask, base._mask) + mbase.soften_mask() + assert_(not mbase._hardmask) + mbase.mask = nomask + # So, the mask of a field is no longer set to nomask... + assert_equal_records(mbase._mask, + ma.make_mask_none(base.shape, base.dtype)) + assert_(ma.make_mask(mbase['b']._mask) is nomask) + assert_equal(mbase['a']._mask, mbase['b']._mask) + + def test_pickling(self): + # Test pickling + base = self.base.copy() + mrec = base.view(mrecarray) + for proto in range(2, pickle.HIGHEST_PROTOCOL + 1): + _ = pickle.dumps(mrec, protocol=proto) + mrec_ = pickle.loads(_) + assert_equal(mrec_.dtype, mrec.dtype) + assert_equal_records(mrec_._data, mrec._data) + assert_equal(mrec_._mask, mrec._mask) + assert_equal_records(mrec_._mask, mrec._mask) + + def test_filled(self): + # Test filling the array + _a = ma.array([1, 2, 3], mask=[0, 0, 1], dtype=int) + _b = ma.array([1.1, 2.2, 3.3], mask=[0, 0, 1], dtype=float) + _c = ma.array(['one', 'two', 'three'], mask=[0, 0, 1], dtype='|S8') + ddtype = [('a', int), ('b', float), ('c', '|S8')] + mrec = fromarrays([_a, _b, _c], dtype=ddtype, + fill_value=(99999, 99999., 'N/A')) + mrecfilled = mrec.filled() + assert_equal(mrecfilled['a'], np.array((1, 2, 99999), dtype=int)) + assert_equal(mrecfilled['b'], np.array((1.1, 2.2, 99999.), + dtype=float)) + assert_equal(mrecfilled['c'], np.array(('one', 'two', 'N/A'), + dtype='|S8')) + + def test_tolist(self): + # Test tolist. + _a = ma.array([1, 2, 3], mask=[0, 0, 1], dtype=int) + _b = ma.array([1.1, 2.2, 3.3], mask=[0, 0, 1], dtype=float) + _c = ma.array(['one', 'two', 'three'], mask=[1, 0, 0], dtype='|S8') + ddtype = [('a', int), ('b', float), ('c', '|S8')] + mrec = fromarrays([_a, _b, _c], dtype=ddtype, + fill_value=(99999, 99999., 'N/A')) + + assert_equal(mrec.tolist(), + [(1, 1.1, None), (2, 2.2, b'two'), + (None, None, b'three')]) + + def test_withnames(self): + # Test the creation w/ format and names + x = mrecarray(1, formats=float, names='base') + x[0]['base'] = 10 + assert_equal(x['base'][0], 10) + + def test_exotic_formats(self): + # Test that 'exotic' formats are processed properly + easy = mrecarray(1, dtype=[('i', int), ('s', '|S8'), ('f', float)]) + easy[0] = masked + assert_equal(easy.filled(1).item(), (1, b'1', 1.)) + + solo = mrecarray(1, dtype=[('f0', '' + + def _validate_input(self, value): + if not isinstance(value, ComplicatedSubArray): + raise ValueError("Can only set to MySubArray values") + return value + + def __setitem__(self, item, value): + # validation ensures direct assignment with ndarray or + # masked_print_option will fail + super().__setitem__(item, self._validate_input(value)) + + def __getitem__(self, item): + # ensure getter returns our own class also for scalars + value = super().__getitem__(item) + if not isinstance(value, np.ndarray): # scalar + value = value.__array__().view(ComplicatedSubArray) + return value + + @property + def flat(self): + return CSAIterator(self) + + @flat.setter + def flat(self, value): + y = self.ravel() + y[:] = value + + def __array_wrap__(self, obj, context=None): + obj = super().__array_wrap__(obj, context) + if context is not None and context[0] is np.multiply: + obj.info['multiplied'] = obj.info.get('multiplied', 0) + 1 + + return obj + + +class WrappedArray(NDArrayOperatorsMixin): + """ + Wrapping a MaskedArray rather than subclassing to test that + ufunc deferrals are commutative. + See: https://github.com/numpy/numpy/issues/15200) + """ + __slots__ = ('_array', 'attrs') + __array_priority__ = 20 + + def __init__(self, array, **attrs): + self._array = array + self.attrs = attrs + + def __repr__(self): + return f"{self.__class__.__name__}(\n{self._array}\n{self.attrs}\n)" + + def __array__(self): + return np.asarray(self._array) + + def __array_ufunc__(self, ufunc, method, *inputs, **kwargs): + if method == '__call__': + inputs = [arg._array if isinstance(arg, self.__class__) else arg + for arg in inputs] + return self.__class__(ufunc(*inputs, **kwargs), **self.attrs) + else: + return NotImplemented + + +class TestSubclassing: + # Test suite for masked subclasses of ndarray. + + def setup_method(self): + x = np.arange(5, dtype='float') + mx = msubarray(x, mask=[0, 1, 0, 0, 0]) + self.data = (x, mx) + + def test_data_subclassing(self): + # Tests whether the subclass is kept. + x = np.arange(5) + m = [0, 0, 1, 0, 0] + xsub = SubArray(x) + xmsub = masked_array(xsub, mask=m) + assert_(isinstance(xmsub, MaskedArray)) + assert_equal(xmsub._data, xsub) + assert_(isinstance(xmsub._data, SubArray)) + + def test_maskedarray_subclassing(self): + # Tests subclassing MaskedArray + (x, mx) = self.data + assert_(isinstance(mx._data, subarray)) + + def test_masked_unary_operations(self): + # Tests masked_unary_operation + (x, mx) = self.data + with np.errstate(divide='ignore'): + assert_(isinstance(log(mx), msubarray)) + assert_equal(log(x), np.log(x)) + + def test_masked_binary_operations(self): + # Tests masked_binary_operation + (x, mx) = self.data + # Result should be a msubarray + assert_(isinstance(add(mx, mx), msubarray)) + assert_(isinstance(add(mx, x), msubarray)) + # Result should work + assert_equal(add(mx, x), mx+x) + assert_(isinstance(add(mx, mx)._data, subarray)) + assert_(isinstance(add.outer(mx, mx), msubarray)) + assert_(isinstance(hypot(mx, mx), msubarray)) + assert_(isinstance(hypot(mx, x), msubarray)) + + def test_masked_binary_operations2(self): + # Tests domained_masked_binary_operation + (x, mx) = self.data + xmx = masked_array(mx.data.__array__(), mask=mx.mask) + assert_(isinstance(divide(mx, mx), msubarray)) + assert_(isinstance(divide(mx, x), msubarray)) + assert_equal(divide(mx, mx), divide(xmx, xmx)) + + def test_attributepropagation(self): + x = array(arange(5), mask=[0]+[1]*4) + my = masked_array(subarray(x)) + ym = msubarray(x) + # + z = (my+1) + assert_(isinstance(z, MaskedArray)) + assert_(not isinstance(z, MSubArray)) + assert_(isinstance(z._data, SubArray)) + assert_equal(z._data.info, {}) + # + z = (ym+1) + assert_(isinstance(z, MaskedArray)) + assert_(isinstance(z, MSubArray)) + assert_(isinstance(z._data, SubArray)) + assert_(z._data.info['added'] > 0) + # Test that inplace methods from data get used (gh-4617) + ym += 1 + assert_(isinstance(ym, MaskedArray)) + assert_(isinstance(ym, MSubArray)) + assert_(isinstance(ym._data, SubArray)) + assert_(ym._data.info['iadded'] > 0) + # + ym._set_mask([1, 0, 0, 0, 1]) + assert_equal(ym._mask, [1, 0, 0, 0, 1]) + ym._series._set_mask([0, 0, 0, 0, 1]) + assert_equal(ym._mask, [0, 0, 0, 0, 1]) + # + xsub = subarray(x, info={'name':'x'}) + mxsub = masked_array(xsub) + assert_(hasattr(mxsub, 'info')) + assert_equal(mxsub.info, xsub.info) + + def test_subclasspreservation(self): + # Checks that masked_array(...,subok=True) preserves the class. + x = np.arange(5) + m = [0, 0, 1, 0, 0] + xinfo = [(i, j) for (i, j) in zip(x, m)] + xsub = MSubArray(x, mask=m, info={'xsub':xinfo}) + # + mxsub = masked_array(xsub, subok=False) + assert_(not isinstance(mxsub, MSubArray)) + assert_(isinstance(mxsub, MaskedArray)) + assert_equal(mxsub._mask, m) + # + mxsub = asarray(xsub) + assert_(not isinstance(mxsub, MSubArray)) + assert_(isinstance(mxsub, MaskedArray)) + assert_equal(mxsub._mask, m) + # + mxsub = masked_array(xsub, subok=True) + assert_(isinstance(mxsub, MSubArray)) + assert_equal(mxsub.info, xsub.info) + assert_equal(mxsub._mask, xsub._mask) + # + mxsub = asanyarray(xsub) + assert_(isinstance(mxsub, MSubArray)) + assert_equal(mxsub.info, xsub.info) + assert_equal(mxsub._mask, m) + + def test_subclass_items(self): + """test that getter and setter go via baseclass""" + x = np.arange(5) + xcsub = ComplicatedSubArray(x) + mxcsub = masked_array(xcsub, mask=[True, False, True, False, False]) + # getter should return a ComplicatedSubArray, even for single item + # first check we wrote ComplicatedSubArray correctly + assert_(isinstance(xcsub[1], ComplicatedSubArray)) + assert_(isinstance(xcsub[1,...], ComplicatedSubArray)) + assert_(isinstance(xcsub[1:4], ComplicatedSubArray)) + + # now that it propagates inside the MaskedArray + assert_(isinstance(mxcsub[1], ComplicatedSubArray)) + assert_(isinstance(mxcsub[1,...].data, ComplicatedSubArray)) + assert_(mxcsub[0] is masked) + assert_(isinstance(mxcsub[0,...].data, ComplicatedSubArray)) + assert_(isinstance(mxcsub[1:4].data, ComplicatedSubArray)) + + # also for flattened version (which goes via MaskedIterator) + assert_(isinstance(mxcsub.flat[1].data, ComplicatedSubArray)) + assert_(mxcsub.flat[0] is masked) + assert_(isinstance(mxcsub.flat[1:4].base, ComplicatedSubArray)) + + # setter should only work with ComplicatedSubArray input + # first check we wrote ComplicatedSubArray correctly + assert_raises(ValueError, xcsub.__setitem__, 1, x[4]) + # now that it propagates inside the MaskedArray + assert_raises(ValueError, mxcsub.__setitem__, 1, x[4]) + assert_raises(ValueError, mxcsub.__setitem__, slice(1, 4), x[1:4]) + mxcsub[1] = xcsub[4] + mxcsub[1:4] = xcsub[1:4] + # also for flattened version (which goes via MaskedIterator) + assert_raises(ValueError, mxcsub.flat.__setitem__, 1, x[4]) + assert_raises(ValueError, mxcsub.flat.__setitem__, slice(1, 4), x[1:4]) + mxcsub.flat[1] = xcsub[4] + mxcsub.flat[1:4] = xcsub[1:4] + + def test_subclass_nomask_items(self): + x = np.arange(5) + xcsub = ComplicatedSubArray(x) + mxcsub_nomask = masked_array(xcsub) + + assert_(isinstance(mxcsub_nomask[1,...].data, ComplicatedSubArray)) + assert_(isinstance(mxcsub_nomask[0,...].data, ComplicatedSubArray)) + + assert_(isinstance(mxcsub_nomask[1], ComplicatedSubArray)) + assert_(isinstance(mxcsub_nomask[0], ComplicatedSubArray)) + + def test_subclass_repr(self): + """test that repr uses the name of the subclass + and 'array' for np.ndarray""" + x = np.arange(5) + mx = masked_array(x, mask=[True, False, True, False, False]) + assert_startswith(repr(mx), 'masked_array') + xsub = SubArray(x) + mxsub = masked_array(xsub, mask=[True, False, True, False, False]) + assert_startswith(repr(mxsub), + f'masked_{SubArray.__name__}(data=[--, 1, --, 3, 4]') + + def test_subclass_str(self): + """test str with subclass that has overridden str, setitem""" + # first without override + x = np.arange(5) + xsub = SubArray(x) + mxsub = masked_array(xsub, mask=[True, False, True, False, False]) + assert_equal(str(mxsub), '[-- 1 -- 3 4]') + + xcsub = ComplicatedSubArray(x) + assert_raises(ValueError, xcsub.__setitem__, 0, + np.ma.core.masked_print_option) + mxcsub = masked_array(xcsub, mask=[True, False, True, False, False]) + assert_equal(str(mxcsub), 'myprefix [-- 1 -- 3 4] mypostfix') + + def test_pure_subclass_info_preservation(self): + # Test that ufuncs and methods conserve extra information consistently; + # see gh-7122. + arr1 = SubMaskedArray('test', data=[1,2,3,4,5,6]) + arr2 = SubMaskedArray(data=[0,1,2,3,4,5]) + diff1 = np.subtract(arr1, arr2) + assert_('info' in diff1._optinfo) + assert_(diff1._optinfo['info'] == 'test') + diff2 = arr1 - arr2 + assert_('info' in diff2._optinfo) + assert_(diff2._optinfo['info'] == 'test') + + +class ArrayNoInheritance: + """Quantity-like class that does not inherit from ndarray""" + def __init__(self, data, units): + self.magnitude = data + self.units = units + + def __getattr__(self, attr): + return getattr(self.magnitude, attr) + + +def test_array_no_inheritance(): + data_masked = np.ma.array([1, 2, 3], mask=[True, False, True]) + data_masked_units = ArrayNoInheritance(data_masked, 'meters') + + # Get the masked representation of the Quantity-like class + new_array = np.ma.array(data_masked_units) + assert_equal(data_masked.data, new_array.data) + assert_equal(data_masked.mask, new_array.mask) + # Test sharing the mask + data_masked.mask = [True, False, False] + assert_equal(data_masked.mask, new_array.mask) + assert_(new_array.sharedmask) + + # Get the masked representation of the Quantity-like class + new_array = np.ma.array(data_masked_units, copy=True) + assert_equal(data_masked.data, new_array.data) + assert_equal(data_masked.mask, new_array.mask) + # Test that the mask is not shared when copy=True + data_masked.mask = [True, False, True] + assert_equal([True, False, False], new_array.mask) + assert_(not new_array.sharedmask) + + # Get the masked representation of the Quantity-like class + new_array = np.ma.array(data_masked_units, keep_mask=False) + assert_equal(data_masked.data, new_array.data) + # The change did not affect the original mask + assert_equal(data_masked.mask, [True, False, True]) + # Test that the mask is False and not shared when keep_mask=False + assert_(not new_array.mask) + assert_(not new_array.sharedmask) + + +class TestClassWrapping: + # Test suite for classes that wrap MaskedArrays + + def setup_method(self): + m = np.ma.masked_array([1, 3, 5], mask=[False, True, False]) + wm = WrappedArray(m) + self.data = (m, wm) + + def test_masked_unary_operations(self): + # Tests masked_unary_operation + (m, wm) = self.data + with np.errstate(divide='ignore'): + assert_(isinstance(np.log(wm), WrappedArray)) + + def test_masked_binary_operations(self): + # Tests masked_binary_operation + (m, wm) = self.data + # Result should be a WrappedArray + assert_(isinstance(np.add(wm, wm), WrappedArray)) + assert_(isinstance(np.add(m, wm), WrappedArray)) + assert_(isinstance(np.add(wm, m), WrappedArray)) + # add and '+' should call the same ufunc + assert_equal(np.add(m, wm), m + wm) + assert_(isinstance(np.hypot(m, wm), WrappedArray)) + assert_(isinstance(np.hypot(wm, m), WrappedArray)) + # Test domained binary operations + assert_(isinstance(np.divide(wm, m), WrappedArray)) + assert_(isinstance(np.divide(m, wm), WrappedArray)) + assert_equal(np.divide(wm, m) * m, np.divide(m, m) * wm) + # Test broadcasting + m2 = np.stack([m, m]) + assert_(isinstance(np.divide(wm, m2), WrappedArray)) + assert_(isinstance(np.divide(m2, wm), WrappedArray)) + assert_equal(np.divide(m2, wm), np.divide(wm, m2)) + + def test_mixins_have_slots(self): + mixin = NDArrayOperatorsMixin() + # Should raise an error + assert_raises(AttributeError, mixin.__setattr__, "not_a_real_attr", 1) + + m = np.ma.masked_array([1, 3, 5], mask=[False, True, False]) + wm = WrappedArray(m) + assert_raises(AttributeError, wm.__setattr__, "not_an_attr", 2) diff --git a/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/INSTALLER b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/INSTALLER new file mode 100644 index 0000000000000000000000000000000000000000..a1b589e38a32041e49332e5e81c2d363dc418d68 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/INSTALLER @@ -0,0 +1 @@ +pip diff --git a/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/License.txt b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/License.txt new file mode 100644 index 0000000000000000000000000000000000000000..b491c70e0aef319022ded661e111ddbd45b8a17f --- /dev/null +++ b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/License.txt @@ -0,0 +1,1568 @@ +End User License Agreement +-------------------------- + + +Preface +------- + +The Software License Agreement in Chapter 1 and the Supplement +in Chapter 2 contain license terms and conditions that govern +the use of NVIDIA software. By accepting this agreement, you +agree to comply with all the terms and conditions applicable +to the product(s) included herein. + + +NVIDIA Driver + + +Description + +This package contains the operating system driver and +fundamental system software components for NVIDIA GPUs. + + +NVIDIA CUDA Toolkit + + +Description + +The NVIDIA CUDA Toolkit provides command-line and graphical +tools for building, debugging and optimizing the performance +of applications accelerated by NVIDIA GPUs, runtime and math +libraries, and documentation including programming guides, +user manuals, and API references. + + +Default Install Location of CUDA Toolkit + +Windows platform: + +%ProgramFiles%\NVIDIA GPU Computing Toolkit\CUDA\v#.# + +Linux platform: + +/usr/local/cuda-#.# + +Mac platform: + +/Developer/NVIDIA/CUDA-#.# + + +NVIDIA CUDA Samples + + +Description + +This package includes over 100+ CUDA examples that demonstrate +various CUDA programming principles, and efficient CUDA +implementation of algorithms in specific application domains. + + +Default Install Location of CUDA Samples + +Windows platform: + +%ProgramData%\NVIDIA Corporation\CUDA Samples\v#.# + +Linux platform: + +/usr/local/cuda-#.#/samples + +and + +$HOME/NVIDIA_CUDA-#.#_Samples + +Mac platform: + +/Developer/NVIDIA/CUDA-#.#/samples + + +NVIDIA Nsight Visual Studio Edition (Windows only) + + +Description + +NVIDIA Nsight Development Platform, Visual Studio Edition is a +development environment integrated into Microsoft Visual +Studio that provides tools for debugging, profiling, analyzing +and optimizing your GPU computing and graphics applications. + + +Default Install Location of Nsight Visual Studio Edition + +Windows platform: + +%ProgramFiles(x86)%\NVIDIA Corporation\Nsight Visual Studio Edition #.# + + +1. License Agreement for NVIDIA Software Development Kits +--------------------------------------------------------- + + +Release Date: July 26, 2018 +--------------------------- + + +Important NoticeRead before downloading, installing, +copying or using the licensed software: +------------------------------------------------------- + +This license agreement, including exhibits attached +("Agreement”) is a legal agreement between you and NVIDIA +Corporation ("NVIDIA") and governs your use of a NVIDIA +software development kit (“SDK”). + +Each SDK has its own set of software and materials, but here +is a description of the types of items that may be included in +a SDK: source code, header files, APIs, data sets and assets +(examples include images, textures, models, scenes, videos, +native API input/output files), binary software, sample code, +libraries, utility programs, programming code and +documentation. + +This Agreement can be accepted only by an adult of legal age +of majority in the country in which the SDK is used. + +If you are entering into this Agreement on behalf of a company +or other legal entity, you represent that you have the legal +authority to bind the entity to this Agreement, in which case +“you” will mean the entity you represent. + +If you don’t have the required age or authority to accept +this Agreement, or if you don’t accept all the terms and +conditions of this Agreement, do not download, install or use +the SDK. + +You agree to use the SDK only for purposes that are permitted +by (a) this Agreement, and (b) any applicable law, regulation +or generally accepted practices or guidelines in the relevant +jurisdictions. + + +1.1. License + + +1.1.1. License Grant + +Subject to the terms of this Agreement, NVIDIA hereby grants +you a non-exclusive, non-transferable license, without the +right to sublicense (except as expressly provided in this +Agreement) to: + + 1. Install and use the SDK, + + 2. Modify and create derivative works of sample source code + delivered in the SDK, and + + 3. Distribute those portions of the SDK that are identified + in this Agreement as distributable, as incorporated in + object code format into a software application that meets + the distribution requirements indicated in this Agreement. + + +1.1.2. Distribution Requirements + +These are the distribution requirements for you to exercise +the distribution grant: + + 1. Your application must have material additional + functionality, beyond the included portions of the SDK. + + 2. The distributable portions of the SDK shall only be + accessed by your application. + + 3. The following notice shall be included in modifications + and derivative works of sample source code distributed: + “This software contains source code provided by NVIDIA + Corporation.” + + 4. Unless a developer tool is identified in this Agreement + as distributable, it is delivered for your internal use + only. + + 5. The terms under which you distribute your application + must be consistent with the terms of this Agreement, + including (without limitation) terms relating to the + license grant and license restrictions and protection of + NVIDIA’s intellectual property rights. Additionally, you + agree that you will protect the privacy, security and + legal rights of your application users. + + 6. You agree to notify NVIDIA in writing of any known or + suspected distribution or use of the SDK not in compliance + with the requirements of this Agreement, and to enforce + the terms of your agreements with respect to distributed + SDK. + + +1.1.3. Authorized Users + +You may allow employees and contractors of your entity or of +your subsidiary(ies) to access and use the SDK from your +secure network to perform work on your behalf. + +If you are an academic institution you may allow users +enrolled or employed by the academic institution to access and +use the SDK from your secure network. + +You are responsible for the compliance with the terms of this +Agreement by your authorized users. If you become aware that +your authorized users didn’t follow the terms of this +Agreement, you agree to take reasonable steps to resolve the +non-compliance and prevent new occurrences. + + +1.1.4. Pre-Release SDK + +The SDK versions identified as alpha, beta, preview or +otherwise as pre-release, may not be fully functional, may +contain errors or design flaws, and may have reduced or +different security, privacy, accessibility, availability, and +reliability standards relative to commercial versions of +NVIDIA software and materials. Use of a pre-release SDK may +result in unexpected results, loss of data, project delays or +other unpredictable damage or loss. + +You may use a pre-release SDK at your own risk, understanding +that pre-release SDKs are not intended for use in production +or business-critical systems. + +NVIDIA may choose not to make available a commercial version +of any pre-release SDK. NVIDIA may also choose to abandon +development and terminate the availability of a pre-release +SDK at any time without liability. + + +1.1.5. Updates + +NVIDIA may, at its option, make available patches, workarounds +or other updates to this SDK. Unless the updates are provided +with their separate governing terms, they are deemed part of +the SDK licensed to you as provided in this Agreement. You +agree that the form and content of the SDK that NVIDIA +provides may change without prior notice to you. While NVIDIA +generally maintains compatibility between versions, NVIDIA may +in some cases make changes that introduce incompatibilities in +future versions of the SDK. + + +1.1.6. Third Party Licenses + +The SDK may come bundled with, or otherwise include or be +distributed with, third party software licensed by a NVIDIA +supplier and/or open source software provided under an open +source license. Use of third party software is subject to the +third-party license terms, or in the absence of third party +terms, the terms of this Agreement. Copyright to third party +software is held by the copyright holders indicated in the +third-party software or license. + + +1.1.7. Reservation of Rights + +NVIDIA reserves all rights, title, and interest in and to the +SDK, not expressly granted to you under this Agreement. + + +1.2. Limitations + +The following license limitations apply to your use of the +SDK: + + 1. You may not reverse engineer, decompile or disassemble, + or remove copyright or other proprietary notices from any + portion of the SDK or copies of the SDK. + + 2. Except as expressly provided in this Agreement, you may + not copy, sell, rent, sublicense, transfer, distribute, + modify, or create derivative works of any portion of the + SDK. For clarity, you may not distribute or sublicense the + SDK as a stand-alone product. + + 3. Unless you have an agreement with NVIDIA for this + purpose, you may not indicate that an application created + with the SDK is sponsored or endorsed by NVIDIA. + + 4. You may not bypass, disable, or circumvent any + encryption, security, digital rights management or + authentication mechanism in the SDK. + + 5. You may not use the SDK in any manner that would cause it + to become subject to an open source software license. As + examples, licenses that require as a condition of use, + modification, and/or distribution that the SDK be: + + a. Disclosed or distributed in source code form; + + b. Licensed for the purpose of making derivative works; + or + + c. Redistributable at no charge. + + 6. Unless you have an agreement with NVIDIA for this + purpose, you may not use the SDK with any system or + application where the use or failure of the system or + application can reasonably be expected to threaten or + result in personal injury, death, or catastrophic loss. + Examples include use in avionics, navigation, military, + medical, life support or other life critical applications. + NVIDIA does not design, test or manufacture the SDK for + these critical uses and NVIDIA shall not be liable to you + or any third party, in whole or in part, for any claims or + damages arising from such uses. + + 7. You agree to defend, indemnify and hold harmless NVIDIA + and its affiliates, and their respective employees, + contractors, agents, officers and directors, from and + against any and all claims, damages, obligations, losses, + liabilities, costs or debt, fines, restitutions and + expenses (including but not limited to attorney’s fees + and costs incident to establishing the right of + indemnification) arising out of or related to your use of + the SDK outside of the scope of this Agreement, or not in + compliance with its terms. + + +1.3. Ownership + + 1. NVIDIA or its licensors hold all rights, title and + interest in and to the SDK and its modifications and + derivative works, including their respective intellectual + property rights, subject to your rights described in this + section. This SDK may include software and materials from + NVIDIA’s licensors, and these licensors are intended + third party beneficiaries that may enforce this Agreement + with respect to their intellectual property rights. + + 2. You hold all rights, title and interest in and to your + applications and your derivative works of the sample + source code delivered in the SDK, including their + respective intellectual property rights, subject to + NVIDIA’s rights described in this section. + + 3. You may, but don’t have to, provide to NVIDIA + suggestions, feature requests or other feedback regarding + the SDK, including possible enhancements or modifications + to the SDK. For any feedback that you voluntarily provide, + you hereby grant NVIDIA and its affiliates a perpetual, + non-exclusive, worldwide, irrevocable license to use, + reproduce, modify, license, sublicense (through multiple + tiers of sublicensees), and distribute (through multiple + tiers of distributors) it without the payment of any + royalties or fees to you. NVIDIA will use feedback at its + choice. NVIDIA is constantly looking for ways to improve + its products, so you may send feedback to NVIDIA through + the developer portal at https://developer.nvidia.com. + + +1.4. No Warranties + +THE SDK IS PROVIDED BY NVIDIA “AS IS” AND “WITH ALL +FAULTS.” TO THE MAXIMUM EXTENT PERMITTED BY LAW, NVIDIA AND +ITS AFFILIATES EXPRESSLY DISCLAIM ALL WARRANTIES OF ANY KIND +OR NATURE, WHETHER EXPRESS, IMPLIED OR STATUTORY, INCLUDING, +BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY, FITNESS +FOR A PARTICULAR PURPOSE, TITLE, NON-INFRINGEMENT, OR THE +ABSENCE OF ANY DEFECTS THEREIN, WHETHER LATENT OR PATENT. NO +WARRANTY IS MADE ON THE BASIS OF TRADE USAGE, COURSE OF +DEALING OR COURSE OF TRADE. + + +1.5. Limitation of Liability + +TO THE MAXIMUM EXTENT PERMITTED BY LAW, NVIDIA AND ITS +AFFILIATES SHALL NOT BE LIABLE FOR ANY SPECIAL, INCIDENTAL, +PUNITIVE OR CONSEQUENTIAL DAMAGES, OR ANY LOST PROFITS, LOSS +OF USE, LOSS OF DATA OR LOSS OF GOODWILL, OR THE COSTS OF +PROCURING SUBSTITUTE PRODUCTS, ARISING OUT OF OR IN CONNECTION +WITH THIS AGREEMENT OR THE USE OR PERFORMANCE OF THE SDK, +WHETHER SUCH LIABILITY ARISES FROM ANY CLAIM BASED UPON BREACH +OF CONTRACT, BREACH OF WARRANTY, TORT (INCLUDING NEGLIGENCE), +PRODUCT LIABILITY OR ANY OTHER CAUSE OF ACTION OR THEORY OF +LIABILITY. IN NO EVENT WILL NVIDIA’S AND ITS AFFILIATES +TOTAL CUMULATIVE LIABILITY UNDER OR ARISING OUT OF THIS +AGREEMENT EXCEED US$10.00. THE NATURE OF THE LIABILITY OR THE +NUMBER OF CLAIMS OR SUITS SHALL NOT ENLARGE OR EXTEND THIS +LIMIT. + +These exclusions and limitations of liability shall apply +regardless if NVIDIA or its affiliates have been advised of +the possibility of such damages, and regardless of whether a +remedy fails its essential purpose. These exclusions and +limitations of liability form an essential basis of the +bargain between the parties, and, absent any of these +exclusions or limitations of liability, the provisions of this +Agreement, including, without limitation, the economic terms, +would be substantially different. + + +1.6. Termination + + 1. This Agreement will continue to apply until terminated by + either you or NVIDIA as described below. + + 2. If you want to terminate this Agreement, you may do so by + stopping to use the SDK. + + 3. NVIDIA may, at any time, terminate this Agreement if: + + a. (i) you fail to comply with any term of this + Agreement and the non-compliance is not fixed within + thirty (30) days following notice from NVIDIA (or + immediately if you violate NVIDIA’s intellectual + property rights); + + b. (ii) you commence or participate in any legal + proceeding against NVIDIA with respect to the SDK; or + + c. (iii) NVIDIA decides to no longer provide the SDK in + a country or, in NVIDIA’s sole discretion, the + continued use of it is no longer commercially viable. + + 4. Upon any termination of this Agreement, you agree to + promptly discontinue use of the SDK and destroy all copies + in your possession or control. Your prior distributions in + accordance with this Agreement are not affected by the + termination of this Agreement. Upon written request, you + will certify in writing that you have complied with your + commitments under this section. Upon any termination of + this Agreement all provisions survive except for the + license grant provisions. + + +1.7. General + +If you wish to assign this Agreement or your rights and +obligations, including by merger, consolidation, dissolution +or operation of law, contact NVIDIA to ask for permission. Any +attempted assignment not approved by NVIDIA in writing shall +be void and of no effect. NVIDIA may assign, delegate or +transfer this Agreement and its rights and obligations, and if +to a non-affiliate you will be notified. + +You agree to cooperate with NVIDIA and provide reasonably +requested information to verify your compliance with this +Agreement. + +This Agreement will be governed in all respects by the laws of +the United States and of the State of Delaware as those laws +are applied to contracts entered into and performed entirely +within Delaware by Delaware residents, without regard to the +conflicts of laws principles. The United Nations Convention on +Contracts for the International Sale of Goods is specifically +disclaimed. You agree to all terms of this Agreement in the +English language. + +The state or federal courts residing in Santa Clara County, +California shall have exclusive jurisdiction over any dispute +or claim arising out of this Agreement. Notwithstanding this, +you agree that NVIDIA shall still be allowed to apply for +injunctive remedies or an equivalent type of urgent legal +relief in any jurisdiction. + +If any court of competent jurisdiction determines that any +provision of this Agreement is illegal, invalid or +unenforceable, such provision will be construed as limited to +the extent necessary to be consistent with and fully +enforceable under the law and the remaining provisions will +remain in full force and effect. Unless otherwise specified, +remedies are cumulative. + +Each party acknowledges and agrees that the other is an +independent contractor in the performance of this Agreement. + +The SDK has been developed entirely at private expense and is +“commercial items” consisting of “commercial computer +software” and “commercial computer software +documentation” provided with RESTRICTED RIGHTS. Use, +duplication or disclosure by the U.S. Government or a U.S. +Government subcontractor is subject to the restrictions in +this Agreement pursuant to DFARS 227.7202-3(a) or as set forth +in subparagraphs (c)(1) and (2) of the Commercial Computer +Software - Restricted Rights clause at FAR 52.227-19, as +applicable. Contractor/manufacturer is NVIDIA, 2788 San Tomas +Expressway, Santa Clara, CA 95051. + +The SDK is subject to United States export laws and +regulations. You agree that you will not ship, transfer or +export the SDK into any country, or use the SDK in any manner, +prohibited by the United States Bureau of Industry and +Security or economic sanctions regulations administered by the +U.S. Department of Treasury’s Office of Foreign Assets +Control (OFAC), or any applicable export laws, restrictions or +regulations. These laws include restrictions on destinations, +end users and end use. By accepting this Agreement, you +confirm that you are not a resident or citizen of any country +currently embargoed by the U.S. and that you are not otherwise +prohibited from receiving the SDK. + +Any notice delivered by NVIDIA to you under this Agreement +will be delivered via mail, email or fax. You agree that any +notices that NVIDIA sends you electronically will satisfy any +legal communication requirements. Please direct your legal +notices or other correspondence to NVIDIA Corporation, 2788 +San Tomas Expressway, Santa Clara, California 95051, United +States of America, Attention: Legal Department. + +This Agreement and any exhibits incorporated into this +Agreement constitute the entire agreement of the parties with +respect to the subject matter of this Agreement and supersede +all prior negotiations or documentation exchanged between the +parties relating to this SDK license. Any additional and/or +conflicting terms on documents issued by you are null, void, +and invalid. Any amendment or waiver under this Agreement +shall be in writing and signed by representatives of both +parties. + + +2. CUDA Toolkit Supplement to Software License Agreement for +NVIDIA Software Development Kits +------------------------------------------------------------ + + +Release date: August 16, 2018 +----------------------------- + +The terms in this supplement govern your use of the NVIDIA +CUDA Toolkit SDK under the terms of your license agreement +(“Agreement”) as modified by this supplement. Capitalized +terms used but not defined below have the meaning assigned to +them in the Agreement. + +This supplement is an exhibit to the Agreement and is +incorporated as an integral part of the Agreement. In the +event of conflict between the terms in this supplement and the +terms in the Agreement, the terms in this supplement govern. + + +2.1. License Scope + +The SDK is licensed for you to develop applications only for +use in systems with NVIDIA GPUs. + + +2.2. Distribution + +The portions of the SDK that are distributable under the +Agreement are listed in Attachment A. + + +2.3. Operating Systems + +Those portions of the SDK designed exclusively for use on the +Linux or FreeBSD operating systems, or other operating systems +derived from the source code to these operating systems, may +be copied and redistributed for use in accordance with this +Agreement, provided that the object code files are not +modified in any way (except for unzipping of compressed +files). + + +2.4. Audio and Video Encoders and Decoders + +You acknowledge and agree that it is your sole responsibility +to obtain any additional third-party licenses required to +make, have made, use, have used, sell, import, and offer for +sale your products or services that include or incorporate any +third-party software and content relating to audio and/or +video encoders and decoders from, including but not limited +to, Microsoft, Thomson, Fraunhofer IIS, Sisvel S.p.A., +MPEG-LA, and Coding Technologies. NVIDIA does not grant to you +under this Agreement any necessary patent or other rights with +respect to any audio and/or video encoders and decoders. + + +2.5. Licensing + +If the distribution terms in this Agreement are not suitable +for your organization, or for any questions regarding this +Agreement, please contact NVIDIA at +nvidia-compute-license-questions@nvidia.com. + + +2.6. Attachment A + +The following portions of the SDK are distributable under the +Agreement: + +Component + +CUDA Runtime + +Windows + +cudart.dll, cudart_static.lib, cudadevrt.lib + +Mac OSX + +libcudart.dylib, libcudart_static.a, libcudadevrt.a + +Linux + +libcudart.so, libcudart_static.a, libcudadevrt.a + +Android + +libcudart.so, libcudart_static.a, libcudadevrt.a + +Component + +CUDA FFT Library + +Windows + +cufft.dll, cufftw.dll, cufft.lib, cufftw.lib + +Mac OSX + +libcufft.dylib, libcufft_static.a, libcufftw.dylib, +libcufftw_static.a + +Linux + +libcufft.so, libcufft_static.a, libcufftw.so, +libcufftw_static.a + +Android + +libcufft.so, libcufft_static.a, libcufftw.so, +libcufftw_static.a + +Component + +CUDA BLAS Library + +Windows + +cublas.dll, cublasLt.dll + +Mac OSX + +libcublas.dylib, libcublasLt.dylib, libcublas_static.a, +libcublasLt_static.a + +Linux + +libcublas.so, libcublasLt.so, libcublas_static.a, +libcublasLt_static.a + +Android + +libcublas.so, libcublasLt.so, libcublas_static.a, +libcublasLt_static.a + +Component + +NVIDIA "Drop-in" BLAS Library + +Windows + +nvblas.dll + +Mac OSX + +libnvblas.dylib + +Linux + +libnvblas.so + +Component + +CUDA Sparse Matrix Library + +Windows + +cusparse.dll, cusparse.lib + +Mac OSX + +libcusparse.dylib, libcusparse_static.a + +Linux + +libcusparse.so, libcusparse_static.a + +Android + +libcusparse.so, libcusparse_static.a + +Component + +CUDA Linear Solver Library + +Windows + +cusolver.dll, cusolver.lib + +Mac OSX + +libcusolver.dylib, libcusolver_static.a + +Linux + +libcusolver.so, libcusolver_static.a + +Android + +libcusolver.so, libcusolver_static.a + +Component + +CUDA Random Number Generation Library + +Windows + +curand.dll, curand.lib + +Mac OSX + +libcurand.dylib, libcurand_static.a + +Linux + +libcurand.so, libcurand_static.a + +Android + +libcurand.so, libcurand_static.a + +Component + +CUDA Accelerated Graph Library + +Component + +NVIDIA Performance Primitives Library + +Windows + +nppc.dll, nppc.lib, nppial.dll, nppial.lib, nppicc.dll, +nppicc.lib, nppicom.dll, nppicom.lib, nppidei.dll, +nppidei.lib, nppif.dll, nppif.lib, nppig.dll, nppig.lib, +nppim.dll, nppim.lib, nppist.dll, nppist.lib, nppisu.dll, +nppisu.lib, nppitc.dll, nppitc.lib, npps.dll, npps.lib + +Mac OSX + +libnppc.dylib, libnppc_static.a, libnppial.dylib, +libnppial_static.a, libnppicc.dylib, libnppicc_static.a, +libnppicom.dylib, libnppicom_static.a, libnppidei.dylib, +libnppidei_static.a, libnppif.dylib, libnppif_static.a, +libnppig.dylib, libnppig_static.a, libnppim.dylib, +libnppisu_static.a, libnppitc.dylib, libnppitc_static.a, +libnpps.dylib, libnpps_static.a + +Linux + +libnppc.so, libnppc_static.a, libnppial.so, +libnppial_static.a, libnppicc.so, libnppicc_static.a, +libnppicom.so, libnppicom_static.a, libnppidei.so, +libnppidei_static.a, libnppif.so, libnppif_static.a +libnppig.so, libnppig_static.a, libnppim.so, +libnppim_static.a, libnppist.so, libnppist_static.a, +libnppisu.so, libnppisu_static.a, libnppitc.so +libnppitc_static.a, libnpps.so, libnpps_static.a + +Android + +libnppc.so, libnppc_static.a, libnppial.so, +libnppial_static.a, libnppicc.so, libnppicc_static.a, +libnppicom.so, libnppicom_static.a, libnppidei.so, +libnppidei_static.a, libnppif.so, libnppif_static.a +libnppig.so, libnppig_static.a, libnppim.so, +libnppim_static.a, libnppist.so, libnppist_static.a, +libnppisu.so, libnppisu_static.a, libnppitc.so +libnppitc_static.a, libnpps.so, libnpps_static.a + +Component + +NVIDIA JPEG Library + +Linux + +libnvjpeg.so, libnvjpeg_static.a + +Component + +Internal common library required for statically linking to +cuBLAS, cuSPARSE, cuFFT, cuRAND, nvJPEG and NPP + +Mac OSX + +libculibos.a + +Linux + +libculibos.a + +Component + +NVIDIA Runtime Compilation Library and Header + +All + +nvrtc.h + +Windows + +nvrtc.dll, nvrtc-builtins.dll + +Mac OSX + +libnvrtc.dylib, libnvrtc-builtins.dylib + +Linux + +libnvrtc.so, libnvrtc-builtins.so + +Component + +NVIDIA Optimizing Compiler Library + +Windows + +nvvm.dll + +Mac OSX + +libnvvm.dylib + +Linux + +libnvvm.so + +Component + +NVIDIA Common Device Math Functions Library + +Windows + +libdevice.10.bc + +Mac OSX + +libdevice.10.bc + +Linux + +libdevice.10.bc + +Component + +CUDA Occupancy Calculation Header Library + +All + +cuda_occupancy.h + +Component + +CUDA Half Precision Headers + +All + +cuda_fp16.h, cuda_fp16.hpp + +Component + +CUDA Profiling Tools Interface (CUPTI) Library + +Windows + +cupti.dll + +Mac OSX + +libcupti.dylib + +Linux + +libcupti.so + +Component + +NVIDIA Tools Extension Library + +Windows + +nvToolsExt.dll, nvToolsExt.lib + +Mac OSX + +libnvToolsExt.dylib + +Linux + +libnvToolsExt.so + +Component + +NVIDIA CUDA Driver Libraries + +Linux + +libcuda.so, libnvidia-fatbinaryloader.so, +libnvidia-ptxjitcompiler.so + +The NVIDIA CUDA Driver Libraries are only distributable in +applications that meet this criteria: + + 1. The application was developed starting from a NVIDIA CUDA + container obtained from Docker Hub or the NVIDIA GPU + Cloud, and + + 2. The resulting application is packaged as a Docker + container and distributed to users on Docker Hub or the + NVIDIA GPU Cloud only. + + +2.7. Attachment B + + +Additional Licensing Obligations + +The following third party components included in the SOFTWARE +are licensed to Licensee pursuant to the following terms and +conditions: + + 1. Licensee's use of the GDB third party component is + subject to the terms and conditions of GNU GPL v3: + + This product includes copyrighted third-party software licensed + under the terms of the GNU General Public License v3 ("GPL v3"). + All third-party software packages are copyright by their respective + authors. GPL v3 terms and conditions are hereby incorporated into + the Agreement by this reference: http://www.gnu.org/licenses/gpl.txt + + Consistent with these licensing requirements, the software + listed below is provided under the terms of the specified + open source software licenses. To obtain source code for + software provided under licenses that require + redistribution of source code, including the GNU General + Public License (GPL) and GNU Lesser General Public License + (LGPL), contact oss-requests@nvidia.com. This offer is + valid for a period of three (3) years from the date of the + distribution of this product by NVIDIA CORPORATION. + + Component License + CUDA-GDB GPL v3 + + 2. Licensee represents and warrants that any and all third + party licensing and/or royalty payment obligations in + connection with Licensee's use of the H.264 video codecs + are solely the responsibility of Licensee. + + 3. Licensee's use of the Thrust library is subject to the + terms and conditions of the Apache License Version 2.0. + All third-party software packages are copyright by their + respective authors. Apache License Version 2.0 terms and + conditions are hereby incorporated into the Agreement by + this reference. + http://www.apache.org/licenses/LICENSE-2.0.html + + In addition, Licensee acknowledges the following notice: + Thrust includes source code from the Boost Iterator, + Tuple, System, and Random Number libraries. + + Boost Software License - Version 1.0 - August 17th, 2003 + . . . . + + Permission is hereby granted, free of charge, to any person or + organization obtaining a copy of the software and accompanying + documentation covered by this license (the "Software") to use, + reproduce, display, distribute, execute, and transmit the Software, + and to prepare derivative works of the Software, and to permit + third-parties to whom the Software is furnished to do so, all + subject to the following: + + The copyright notices in the Software and this entire statement, + including the above license grant, this restriction and the following + disclaimer, must be included in all copies of the Software, in whole + or in part, and all derivative works of the Software, unless such + copies or derivative works are solely in the form of machine-executable + object code generated by a source language processor. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE AND + NON-INFRINGEMENT. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR + ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE FOR ANY DAMAGES OR + OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, ARISING + FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + OTHER DEALINGS IN THE SOFTWARE. + + 4. Licensee's use of the LLVM third party component is + subject to the following terms and conditions: + + ====================================================== + LLVM Release License + ====================================================== + University of Illinois/NCSA + Open Source License + + Copyright (c) 2003-2010 University of Illinois at Urbana-Champaign. + All rights reserved. + + Developed by: + + LLVM Team + + University of Illinois at Urbana-Champaign + + http://llvm.org + + Permission is hereby granted, free of charge, to any person obtaining a copy + of this software and associated documentation files (the "Software"), to + deal with the Software without restriction, including without limitation the + rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + sell copies of the Software, and to permit persons to whom the Software is + furnished to do so, subject to the following conditions: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimers. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimers in the + documentation and/or other materials provided with the distribution. + + * Neither the names of the LLVM Team, University of Illinois at Urbana- + Champaign, nor the names of its contributors may be used to endorse or + promote products derived from this Software without specific prior + written permission. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + THE CONTRIBUTORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + DEALINGS WITH THE SOFTWARE. + + 5. Licensee's use (e.g. nvprof) of the PCRE third party + component is subject to the following terms and + conditions: + + ------------ + PCRE LICENCE + ------------ + PCRE is a library of functions to support regular expressions whose syntax + and semantics are as close as possible to those of the Perl 5 language. + Release 8 of PCRE is distributed under the terms of the "BSD" licence, as + specified below. The documentation for PCRE, supplied in the "doc" + directory, is distributed under the same terms as the software itself. The + basic library functions are written in C and are freestanding. Also + included in the distribution is a set of C++ wrapper functions, and a just- + in-time compiler that can be used to optimize pattern matching. These are + both optional features that can be omitted when the library is built. + + THE BASIC LIBRARY FUNCTIONS + --------------------------- + Written by: Philip Hazel + Email local part: ph10 + Email domain: cam.ac.uk + University of Cambridge Computing Service, + Cambridge, England. + Copyright (c) 1997-2012 University of Cambridge + All rights reserved. + + PCRE JUST-IN-TIME COMPILATION SUPPORT + ------------------------------------- + Written by: Zoltan Herczeg + Email local part: hzmester + Emain domain: freemail.hu + Copyright(c) 2010-2012 Zoltan Herczeg + All rights reserved. + + STACK-LESS JUST-IN-TIME COMPILER + -------------------------------- + Written by: Zoltan Herczeg + Email local part: hzmester + Emain domain: freemail.hu + Copyright(c) 2009-2012 Zoltan Herczeg + All rights reserved. + + THE C++ WRAPPER FUNCTIONS + ------------------------- + Contributed by: Google Inc. + Copyright (c) 2007-2012, Google Inc. + All rights reserved. + + THE "BSD" LICENCE + ----------------- + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the name of Google + Inc. nor the names of their contributors may be used to endorse or + promote products derived from this software without specific prior + written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + POSSIBILITY OF SUCH DAMAGE. + + 6. Some of the cuBLAS library routines were written by or + derived from code written by Vasily Volkov and are subject + to the Modified Berkeley Software Distribution License as + follows: + + Copyright (c) 2007-2009, Regents of the University of California + + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + * Neither the name of the University of California, Berkeley nor + the names of its contributors may be used to endorse or promote + products derived from this software without specific prior + written permission. + + THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR + IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, + INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING + IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + POSSIBILITY OF SUCH DAMAGE. + + 7. Some of the cuBLAS library routines were written by or + derived from code written by Davide Barbieri and are + subject to the Modified Berkeley Software Distribution + License as follows: + + Copyright (c) 2008-2009 Davide Barbieri @ University of Rome Tor Vergata. + + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + * The name of the author may not be used to endorse or promote + products derived from this software without specific prior + written permission. + + THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR + IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, + INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING + IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + POSSIBILITY OF SUCH DAMAGE. + + 8. Some of the cuBLAS library routines were derived from + code developed by the University of Tennessee and are + subject to the Modified Berkeley Software Distribution + License as follows: + + Copyright (c) 2010 The University of Tennessee. + + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer listed in this license in the documentation and/or + other materials provided with the distribution. + * Neither the name of the copyright holders nor the names of its + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + 9. Some of the cuBLAS library routines were written by or + derived from code written by Jonathan Hogg and are subject + to the Modified Berkeley Software Distribution License as + follows: + + Copyright (c) 2012, The Science and Technology Facilities Council (STFC). + + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + * Neither the name of the STFC nor the names of its contributors + may be used to endorse or promote products derived from this + software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE STFC BE + LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + 10. Some of the cuBLAS library routines were written by or + derived from code written by Ahmad M. Abdelfattah, David + Keyes, and Hatem Ltaief, and are subject to the Apache + License, Version 2.0, as follows: + + -- (C) Copyright 2013 King Abdullah University of Science and Technology + Authors: + Ahmad Abdelfattah (ahmad.ahmad@kaust.edu.sa) + David Keyes (david.keyes@kaust.edu.sa) + Hatem Ltaief (hatem.ltaief@kaust.edu.sa) + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + * Neither the name of the King Abdullah University of Science and + Technology nor the names of its contributors may be used to endorse + or promote products derived from this software without specific prior + written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE + + 11. Some of the cuSPARSE library routines were written by or + derived from code written by Li-Wen Chang and are subject + to the NCSA Open Source License as follows: + + Copyright (c) 2012, University of Illinois. + + All rights reserved. + + Developed by: IMPACT Group, University of Illinois, http://impact.crhc.illinois.edu + + Permission is hereby granted, free of charge, to any person obtaining + a copy of this software and associated documentation files (the + "Software"), to deal with the Software without restriction, including + without limitation the rights to use, copy, modify, merge, publish, + distribute, sublicense, and/or sell copies of the Software, and to + permit persons to whom the Software is furnished to do so, subject to + the following conditions: + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimers in the documentation and/or other materials provided + with the distribution. + * Neither the names of IMPACT Group, University of Illinois, nor + the names of its contributors may be used to endorse or promote + products derived from this Software without specific prior + written permission. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + NONINFRINGEMENT. IN NO EVENT SHALL THE CONTRIBUTORS OR COPYRIGHT + HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER + IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR + IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS WITH THE + SOFTWARE. + + 12. Some of the cuRAND library routines were written by or + derived from code written by Mutsuo Saito and Makoto + Matsumoto and are subject to the following license: + + Copyright (c) 2009, 2010 Mutsuo Saito, Makoto Matsumoto and Hiroshima + University. All rights reserved. + + Copyright (c) 2011 Mutsuo Saito, Makoto Matsumoto, Hiroshima + University and University of Tokyo. All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + * Neither the name of the Hiroshima University nor the names of + its contributors may be used to endorse or promote products + derived from this software without specific prior written + permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + 13. Some of the cuRAND library routines were derived from + code developed by D. E. Shaw Research and are subject to + the following license: + + Copyright 2010-2011, D. E. Shaw Research. + + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + * Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions, and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + * Neither the name of D. E. Shaw Research nor the names of its + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + 14. Some of the Math library routines were written by or + derived from code developed by Norbert Juffa and are + subject to the following license: + + Copyright (c) 2015-2017, Norbert Juffa + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + 15. Licensee's use of the lz4 third party component is + subject to the following terms and conditions: + + Copyright (C) 2011-2013, Yann Collet. + BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following disclaimer + in the documentation and/or other materials provided with the + distribution. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + 16. The NPP library uses code from the Boost Math Toolkit, + and is subject to the following license: + + Boost Software License - Version 1.0 - August 17th, 2003 + . . . . + + Permission is hereby granted, free of charge, to any person or + organization obtaining a copy of the software and accompanying + documentation covered by this license (the "Software") to use, + reproduce, display, distribute, execute, and transmit the Software, + and to prepare derivative works of the Software, and to permit + third-parties to whom the Software is furnished to do so, all + subject to the following: + + The copyright notices in the Software and this entire statement, + including the above license grant, this restriction and the following + disclaimer, must be included in all copies of the Software, in whole + or in part, and all derivative works of the Software, unless such + copies or derivative works are solely in the form of machine-executable + object code generated by a source language processor. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE AND + NON-INFRINGEMENT. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR + ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE FOR ANY DAMAGES OR + OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, ARISING + FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + OTHER DEALINGS IN THE SOFTWARE. + + 17. Portions of the Nsight Eclipse Edition is subject to the + following license: + + The Eclipse Foundation makes available all content in this plug-in + ("Content"). Unless otherwise indicated below, the Content is provided + to you under the terms and conditions of the Eclipse Public License + Version 1.0 ("EPL"). A copy of the EPL is available at http:// + www.eclipse.org/legal/epl-v10.html. For purposes of the EPL, "Program" + will mean the Content. + + If you did not receive this Content directly from the Eclipse + Foundation, the Content is being redistributed by another party + ("Redistributor") and different terms and conditions may apply to your + use of any object code in the Content. Check the Redistributor's + license that was provided with the Content. If no such license exists, + contact the Redistributor. Unless otherwise indicated below, the terms + and conditions of the EPL still apply to any source code in the + Content and such source code may be obtained at http://www.eclipse.org. + + 18. Some of the cuBLAS library routines uses code from + OpenAI, which is subject to the following license: + + License URL + https://github.com/openai/openai-gemm/blob/master/LICENSE + + License Text + The MIT License + + Copyright (c) 2016 OpenAI (http://openai.com), 2016 Google Inc. + + Permission is hereby granted, free of charge, to any person obtaining a copy + of this software and associated documentation files (the "Software"), to deal + in the Software without restriction, including without limitation the rights + to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + copies of the Software, and to permit persons to whom the Software is + furnished to do so, subject to the following conditions: + + The above copyright notice and this permission notice shall be included in + all copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + THE SOFTWARE. + + 19. Licensee's use of the Visual Studio Setup Configuration + Samples is subject to the following license: + + The MIT License (MIT) + Copyright (C) Microsoft Corporation. All rights reserved. + + Permission is hereby granted, free of charge, to any person + obtaining a copy of this software and associated documentation + files (the "Software"), to deal in the Software without restriction, + including without limitation the rights to use, copy, modify, merge, + publish, distribute, sublicense, and/or sell copies of the Software, + and to permit persons to whom the Software is furnished to do so, + subject to the following conditions: + + The above copyright notice and this permission notice shall be included + in all copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + + 20. Licensee's use of linmath.h header for CPU functions for + GL vector/matrix operations from lunarG is subject to the + Apache License Version 2.0. + + 21. The DX12-CUDA sample uses the d3dx12.h header, which is + subject to the MIT license . + +----------------- diff --git a/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/METADATA b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/METADATA new file mode 100644 index 0000000000000000000000000000000000000000..e8ef8db3f0102373f6e6897f513e5a80fb4e4dbd --- /dev/null +++ b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/METADATA @@ -0,0 +1,36 @@ +Metadata-Version: 2.1 +Name: nvidia-cusparse-cu12 +Version: 12.3.1.170 +Summary: CUSPARSE native runtime libraries +Home-page: https://developer.nvidia.com/cuda-zone +Author: Nvidia CUDA Installer Team +Author-email: cuda_installer@nvidia.com +License: NVIDIA Proprietary Software +Keywords: cuda,nvidia,runtime,machine learning,deep learning +Classifier: Development Status :: 4 - Beta +Classifier: Intended Audience :: Developers +Classifier: Intended Audience :: Education +Classifier: Intended Audience :: Science/Research +Classifier: License :: Other/Proprietary License +Classifier: Natural Language :: English +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.5 +Classifier: Programming Language :: Python :: 3.6 +Classifier: Programming Language :: Python :: 3.7 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Programming Language :: Python :: 3 :: Only +Classifier: Topic :: Scientific/Engineering +Classifier: Topic :: Scientific/Engineering :: Mathematics +Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence +Classifier: Topic :: Software Development +Classifier: Topic :: Software Development :: Libraries +Classifier: Operating System :: Microsoft :: Windows +Classifier: Operating System :: POSIX :: Linux +Requires-Python: >=3 +License-File: License.txt +Requires-Dist: nvidia-nvjitlink-cu12 + +CUSPARSE native runtime libraries diff --git a/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/RECORD b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/RECORD new file mode 100644 index 0000000000000000000000000000000000000000..6f5cf0c3f006211a8e4ea9c93b319b6e702c1aaf --- /dev/null +++ b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/RECORD @@ -0,0 +1,17 @@ +nvidia/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +nvidia/__pycache__/__init__.cpython-311.pyc,, +nvidia/cusparse/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +nvidia/cusparse/__pycache__/__init__.cpython-311.pyc,, +nvidia/cusparse/include/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +nvidia/cusparse/include/__pycache__/__init__.cpython-311.pyc,, +nvidia/cusparse/include/cusparse.h,sha256=g-jyaV5nBdnCExOVQOd7lf4lMBoR3KvriiNHUoXNgXw,295187 +nvidia/cusparse/include/cusparse_v2.h,sha256=jkH2A9hYc-TEF0vuQ_SurbhPNEHkYGUIRuxKXhFAqnw,2587 +nvidia/cusparse/lib/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +nvidia/cusparse/lib/__pycache__/__init__.cpython-311.pyc,, +nvidia/cusparse/lib/libcusparse.so.12,sha256=y0KIyWVFOoAgZTwP2SzfzRK7WmRoQ9Zj6dTl9SD4kVw,281313984 +nvidia_cusparse_cu12-12.3.1.170.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4 +nvidia_cusparse_cu12-12.3.1.170.dist-info/License.txt,sha256=rW9YU_ugyg0VnQ9Y1JrkmDDC-Mk_epJki5zpCttMbM0,59262 +nvidia_cusparse_cu12-12.3.1.170.dist-info/METADATA,sha256=ivWTVzZv8-4bxy3u1Ce2wd3qUZeGJ3jTuH3-8u7AILM,1550 +nvidia_cusparse_cu12-12.3.1.170.dist-info/RECORD,, +nvidia_cusparse_cu12-12.3.1.170.dist-info/WHEEL,sha256=XDTs3wIbcE-BcRO08VJlZpA6z9OaC1mOKPCGGGwuM2g,109 +nvidia_cusparse_cu12-12.3.1.170.dist-info/top_level.txt,sha256=fTkAtiFuL16nUrB9ytDDtpytz2t0B4NvYTnRzwAhO14,7 diff --git a/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/WHEEL b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/WHEEL new file mode 100644 index 0000000000000000000000000000000000000000..e6c30e957cfb045017a9fef3430bb8ee87c4a074 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/WHEEL @@ -0,0 +1,5 @@ +Wheel-Version: 1.0 +Generator: bdist_wheel (0.42.0) +Root-Is-Purelib: true +Tag: py3-none-manylinux2014_x86_64 + diff --git a/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/top_level.txt b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/top_level.txt new file mode 100644 index 0000000000000000000000000000000000000000..862f7abf232cdfbb928609856247292e81c9decb --- /dev/null +++ b/.venv/lib/python3.11/site-packages/nvidia_cusparse_cu12-12.3.1.170.dist-info/top_level.txt @@ -0,0 +1 @@ +nvidia diff --git a/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/INSTALLER b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/INSTALLER new file mode 100644 index 0000000000000000000000000000000000000000..a1b589e38a32041e49332e5e81c2d363dc418d68 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/INSTALLER @@ -0,0 +1 @@ +pip diff --git a/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/METADATA b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/METADATA new file mode 100644 index 0000000000000000000000000000000000000000..e119fbd23e75aa66a8cc2b7dd4cd535eeae4acfd --- /dev/null +++ b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/METADATA @@ -0,0 +1,359 @@ +Metadata-Version: 2.4 +Name: pydantic +Version: 2.10.6 +Summary: Data validation using Python type hints +Project-URL: Homepage, https://github.com/pydantic/pydantic +Project-URL: Documentation, https://docs.pydantic.dev +Project-URL: Funding, https://github.com/sponsors/samuelcolvin +Project-URL: Source, https://github.com/pydantic/pydantic +Project-URL: Changelog, https://docs.pydantic.dev/latest/changelog/ +Author-email: Samuel Colvin , Eric Jolibois , Hasan Ramezani , Adrian Garcia Badaracco <1755071+adriangb@users.noreply.github.com>, Terrence Dorsey , David Montague , Serge Matveenko , Marcelo Trylesinski , Sydney Runkle , David Hewitt , Alex Hall , Victorien Plot +License-Expression: MIT +License-File: LICENSE +Classifier: Development Status :: 5 - Production/Stable +Classifier: Framework :: Hypothesis +Classifier: Framework :: Pydantic +Classifier: Intended Audience :: Developers +Classifier: Intended Audience :: Information Technology +Classifier: License :: OSI Approved :: MIT License +Classifier: Operating System :: OS Independent +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3 :: Only +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Programming Language :: Python :: 3.12 +Classifier: Programming Language :: Python :: 3.13 +Classifier: Programming Language :: Python :: Implementation :: CPython +Classifier: Programming Language :: Python :: Implementation :: PyPy +Classifier: Topic :: Internet +Classifier: Topic :: Software Development :: Libraries :: Python Modules +Requires-Python: >=3.8 +Requires-Dist: annotated-types>=0.6.0 +Requires-Dist: pydantic-core==2.27.2 +Requires-Dist: typing-extensions>=4.12.2 +Provides-Extra: email +Requires-Dist: email-validator>=2.0.0; extra == 'email' +Provides-Extra: timezone +Requires-Dist: tzdata; (python_version >= '3.9' and platform_system == 'Windows') and extra == 'timezone' +Description-Content-Type: text/markdown + +# Pydantic +[![CI](https://img.shields.io/github/actions/workflow/status/pydantic/pydantic/ci.yml?branch=main&logo=github&label=CI)](https://github.com/pydantic/pydantic/actions?query=event%3Apush+branch%3Amain+workflow%3ACI) +[![Coverage](https://coverage-badge.samuelcolvin.workers.dev/pydantic/pydantic.svg)](https://coverage-badge.samuelcolvin.workers.dev/redirect/pydantic/pydantic) +[![pypi](https://img.shields.io/pypi/v/pydantic.svg)](https://pypi.python.org/pypi/pydantic) +[![CondaForge](https://img.shields.io/conda/v/conda-forge/pydantic.svg)](https://anaconda.org/conda-forge/pydantic) +[![downloads](https://static.pepy.tech/badge/pydantic/month)](https://pepy.tech/project/pydantic) +[![versions](https://img.shields.io/pypi/pyversions/pydantic.svg)](https://github.com/pydantic/pydantic) +[![license](https://img.shields.io/github/license/pydantic/pydantic.svg)](https://github.com/pydantic/pydantic/blob/main/LICENSE) +[![Pydantic v2](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/pydantic/pydantic/main/docs/badge/v2.json)](https://docs.pydantic.dev/latest/contributing/#badges) + +Data validation using Python type hints. + +Fast and extensible, Pydantic plays nicely with your linters/IDE/brain. +Define how data should be in pure, canonical Python 3.8+; validate it with Pydantic. + +## Pydantic Logfire :fire: + +We've recently launched Pydantic Logfire to help you monitor your applications. +[Learn more](https://pydantic.dev/articles/logfire-announcement) + +## Pydantic V1.10 vs. V2 + +Pydantic V2 is a ground-up rewrite that offers many new features, performance improvements, and some breaking changes compared to Pydantic V1. + +If you're using Pydantic V1 you may want to look at the +[pydantic V1.10 Documentation](https://docs.pydantic.dev/) or, +[`1.10.X-fixes` git branch](https://github.com/pydantic/pydantic/tree/1.10.X-fixes). Pydantic V2 also ships with the latest version of Pydantic V1 built in so that you can incrementally upgrade your code base and projects: `from pydantic import v1 as pydantic_v1`. + +## Help + +See [documentation](https://docs.pydantic.dev/) for more details. + +## Installation + +Install using `pip install -U pydantic` or `conda install pydantic -c conda-forge`. +For more installation options to make Pydantic even faster, +see the [Install](https://docs.pydantic.dev/install/) section in the documentation. + +## A Simple Example + +```python +from datetime import datetime +from typing import List, Optional +from pydantic import BaseModel + +class User(BaseModel): + id: int + name: str = 'John Doe' + signup_ts: Optional[datetime] = None + friends: List[int] = [] + +external_data = {'id': '123', 'signup_ts': '2017-06-01 12:22', 'friends': [1, '2', b'3']} +user = User(**external_data) +print(user) +#> User id=123 name='John Doe' signup_ts=datetime.datetime(2017, 6, 1, 12, 22) friends=[1, 2, 3] +print(user.id) +#> 123 +``` + +## Contributing + +For guidance on setting up a development environment and how to make a +contribution to Pydantic, see +[Contributing to Pydantic](https://docs.pydantic.dev/contributing/). + +## Reporting a Security Vulnerability + +See our [security policy](https://github.com/pydantic/pydantic/security/policy). + +## Changelog + +## v2.10.6 (2025-01-23) + +[GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.6) + +### What's Changed + +#### Fixes + +* Fix JSON Schema reference collection with `'examples'` keys by [@Viicos](https://github.com/Viicos) in [#11325](https://github.com/pydantic/pydantic/pull/11325) +* Fix url python serialization by [@sydney-runkle](https://github.com/sydney-runkle) in [#11331](https://github.com/pydantic/pydantic/pull/11331) + + +## v2.10.5 (2025-01-08) + +[GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.5) + +### What's Changed + +#### Fixes + +* Remove custom MRO implementation of Pydantic models by [@Viicos](https://github.com/Viicos) in [#11184](https://github.com/pydantic/pydantic/pull/11184) +* Fix URL serialization for unions by [@sydney-runkle](https://github.com/sydney-runkle) in [#11233](https://github.com/pydantic/pydantic/pull/11233) + +## v2.10.4 (2024-12-18) + +[GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.4) + +### What's Changed + +#### Packaging + +* Bump `pydantic-core` to v2.27.2 by [@davidhewitt](https://github.com/davidhewitt) in [#11138](https://github.com/pydantic/pydantic/pull/11138) + +#### Fixes + +* Fix for comparison of `AnyUrl` objects by [@alexprabhat99](https://github.com/alexprabhat99) in [#11082](https://github.com/pydantic/pydantic/pull/11082) +* Properly fetch PEP 695 type params for functions, do not fetch annotations from signature by [@Viicos](https://github.com/Viicos) in [#11093](https://github.com/pydantic/pydantic/pull/11093) +* Include JSON Schema input core schema in function schemas by [@Viicos](https://github.com/Viicos) in [#11085](https://github.com/pydantic/pydantic/pull/11085) +* Add `len` to `_BaseUrl` to avoid TypeError by [@Kharianne](https://github.com/Kharianne) in [#11111](https://github.com/pydantic/pydantic/pull/11111) +* Make sure the type reference is removed from the seen references by [@Viicos](https://github.com/Viicos) in [#11143](https://github.com/pydantic/pydantic/pull/11143) + +### New Contributors + +* [@alexprabhat99](https://github.com/alexprabhat99) made their first contribution in [#11082](https://github.com/pydantic/pydantic/pull/11082) +* [@Kharianne](https://github.com/Kharianne) made their first contribution in [#11111](https://github.com/pydantic/pydantic/pull/11111) + +## v2.10.3 (2024-12-03) + +[GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.3) + +### What's Changed + +#### Fixes + +* Set fields when `defer_build` is set on Pydantic dataclasses by [@Viicos](https://github.com/Viicos) in [#10984](https://github.com/pydantic/pydantic/pull/10984) +* Do not resolve the JSON Schema reference for `dict` core schema keys by [@Viicos](https://github.com/Viicos) in [#10989](https://github.com/pydantic/pydantic/pull/10989) +* Use the globals of the function when evaluating the return type for `PlainSerializer` and `WrapSerializer` functions by [@Viicos](https://github.com/Viicos) in [#11008](https://github.com/pydantic/pydantic/pull/11008) +* Fix host required enforcement for urls to be compatible with v2.9 behavior by [@sydney-runkle](https://github.com/sydney-runkle) in [#11027](https://github.com/pydantic/pydantic/pull/11027) +* Add a `default_factory_takes_validated_data` property to `FieldInfo` by [@Viicos](https://github.com/Viicos) in [#11034](https://github.com/pydantic/pydantic/pull/11034) +* Fix url json schema in `serialization` mode by [@sydney-runkle](https://github.com/sydney-runkle) in [#11035](https://github.com/pydantic/pydantic/pull/11035) + +## v2.10.2 (2024-11-25) + +[GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.2) + +### What's Changed + +#### Fixes + +* Only evaluate FieldInfo annotations if required during schema building by [@Viicos](https://github.com/Viicos) in [#10769](https://github.com/pydantic/pydantic/pull/10769) +* Do not evaluate annotations for private fields by [@Viicos](https://github.com/Viicos) in [#10962](https://github.com/pydantic/pydantic/pull/10962) +* Support serialization as any for `Secret` types and `Url` types by [@sydney-runkle](https://github.com/sydney-runkle) in [#10947](https://github.com/pydantic/pydantic/pull/10947) +* Fix type hint of `Field.default` to be compatible with Python 3.8 and 3.9 by [@Viicos](https://github.com/Viicos) in [#10972](https://github.com/pydantic/pydantic/pull/10972) +* Add hashing support for URL types by [@sydney-runkle](https://github.com/sydney-runkle) in [#10975](https://github.com/pydantic/pydantic/pull/10975) +* Hide `BaseModel.__replace__` definition from type checkers by [@Viicos](https://github.com/Viicos) in [10979](https://github.com/pydantic/pydantic/pull/10979) + +## v2.10.1 (2024-11-21) + +[GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.1) + +### What's Changed + +#### Packaging + +* Bump `pydantic-core` version to `v2.27.1` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10938](https://github.com/pydantic/pydantic/pull/10938) + +#### Fixes + +* Use the correct frame when instantiating a parametrized `TypeAdapter` by [@Viicos](https://github.com/Viicos) in [#10893](https://github.com/pydantic/pydantic/pull/10893) +* Relax check for validated data in `default_factory` utils by [@sydney-runkle](https://github.com/sydney-runkle) in [#10909](https://github.com/pydantic/pydantic/pull/10909) +* Fix type checking issue with `model_fields` and `model_computed_fields` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10911](https://github.com/pydantic/pydantic/pull/10911) +* Use the parent configuration during schema generation for stdlib `dataclass`es by [@sydney-runkle](https://github.com/sydney-runkle) in [#10928](https://github.com/pydantic/pydantic/pull/10928) +* Use the `globals` of the function when evaluating the return type of serializers and `computed_field`s by [@Viicos](https://github.com/Viicos) in [#10929](https://github.com/pydantic/pydantic/pull/10929) +* Fix URL constraint application by [@sydney-runkle](https://github.com/sydney-runkle) in [#10922](https://github.com/pydantic/pydantic/pull/10922) +* Fix URL equality with different validation methods by [@sydney-runkle](https://github.com/sydney-runkle) in [#10934](https://github.com/pydantic/pydantic/pull/10934) +* Fix JSON schema title when specified as `''` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10936](https://github.com/pydantic/pydantic/pull/10936) +* Fix `python` mode serialization for `complex` inference by [@sydney-runkle](https://github.com/sydney-runkle) in [pydantic-core#1549](https://github.com/pydantic/pydantic-core/pull/1549) + +## v2.10.0 (2024-11-20) + +The code released in v2.10.0 is practically identical to that of v2.10.0b2. + +[GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.0) + +See the [v2.10 release blog post](https://pydantic.dev/articles/pydantic-v2-10-release) for the highlights! + +### What's Changed + +#### Packaging + +* Bump `pydantic-core` to `v2.27.0` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10825](https://github.com/pydantic/pydantic/pull/10825) +* Replaced pdm with uv by [@frfahim](https://github.com/frfahim) in [#10727](https://github.com/pydantic/pydantic/pull/10727) + +#### New Features + +* Support `fractions.Fraction` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10318](https://github.com/pydantic/pydantic/pull/10318) +* Support `Hashable` for json validation by [@sydney-runkle](https://github.com/sydney-runkle) in [#10324](https://github.com/pydantic/pydantic/pull/10324) +* Add a `SocketPath` type for `linux` systems by [@theunkn0wn1](https://github.com/theunkn0wn1) in [#10378](https://github.com/pydantic/pydantic/pull/10378) +* Allow arbitrary refs in JSON schema `examples` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10417](https://github.com/pydantic/pydantic/pull/10417) +* Support `defer_build` for Pydantic dataclasses by [@Viicos](https://github.com/Viicos) in [#10313](https://github.com/pydantic/pydantic/pull/10313) +* Adding v1 / v2 incompatibility warning for nested v1 model by [@sydney-runkle](https://github.com/sydney-runkle) in [#10431](https://github.com/pydantic/pydantic/pull/10431) +* Add support for unpacked `TypedDict` to type hint variadic keyword arguments with `@validate_call` by [@Viicos](https://github.com/Viicos) in [#10416](https://github.com/pydantic/pydantic/pull/10416) +* Support compiled patterns in `protected_namespaces` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10522](https://github.com/pydantic/pydantic/pull/10522) +* Add support for `propertyNames` in JSON schema by [@FlorianSW](https://github.com/FlorianSW) in [#10478](https://github.com/pydantic/pydantic/pull/10478) +* Adding `__replace__` protocol for Python 3.13+ support by [@sydney-runkle](https://github.com/sydney-runkle) in [#10596](https://github.com/pydantic/pydantic/pull/10596) +* Expose public `sort` method for JSON schema generation by [@sydney-runkle](https://github.com/sydney-runkle) in [#10595](https://github.com/pydantic/pydantic/pull/10595) +* Add runtime validation of `@validate_call` callable argument by [@kc0506](https://github.com/kc0506) in [#10627](https://github.com/pydantic/pydantic/pull/10627) +* Add `experimental_allow_partial` support by [@samuelcolvin](https://github.com/samuelcolvin) in [#10748](https://github.com/pydantic/pydantic/pull/10748) +* Support default factories taking validated data as an argument by [@Viicos](https://github.com/Viicos) in [#10678](https://github.com/pydantic/pydantic/pull/10678) +* Allow subclassing `ValidationError` and `PydanticCustomError` by [@Youssefares](https://github.com/Youssefares) in [pydantic/pydantic-core#1413](https://github.com/pydantic/pydantic-core/pull/1413) +* Add `trailing-strings` support to `experimental_allow_partial` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10825](https://github.com/pydantic/pydantic/pull/10825) +* Add `rebuild()` method for `TypeAdapter` and simplify `defer_build` patterns by [@sydney-runkle](https://github.com/sydney-runkle) in [#10537](https://github.com/pydantic/pydantic/pull/10537) +* Improve `TypeAdapter` instance repr by [@sydney-runkle](https://github.com/sydney-runkle) in [#10872](https://github.com/pydantic/pydantic/pull/10872) + +#### Changes + +* Don't allow customization of `SchemaGenerator` until interface is more stable by [@sydney-runkle](https://github.com/sydney-runkle) in [#10303](https://github.com/pydantic/pydantic/pull/10303) +* Cleanly `defer_build` on `TypeAdapters`, removing experimental flag by [@sydney-runkle](https://github.com/sydney-runkle) in [#10329](https://github.com/pydantic/pydantic/pull/10329) +* Fix `mro` of generic subclass by [@kc0506](https://github.com/kc0506) in [#10100](https://github.com/pydantic/pydantic/pull/10100) +* Strip whitespaces on JSON Schema title generation by [@sydney-runkle](https://github.com/sydney-runkle) in [#10404](https://github.com/pydantic/pydantic/pull/10404) +* Use `b64decode` and `b64encode` for `Base64Bytes` type by [@sydney-runkle](https://github.com/sydney-runkle) in [#10486](https://github.com/pydantic/pydantic/pull/10486) +* Relax protected namespace config default by [@sydney-runkle](https://github.com/sydney-runkle) in [#10441](https://github.com/pydantic/pydantic/pull/10441) +* Revalidate parametrized generics if instance's origin is subclass of OG class by [@sydney-runkle](https://github.com/sydney-runkle) in [#10666](https://github.com/pydantic/pydantic/pull/10666) +* Warn if configuration is specified on the `@dataclass` decorator and with the `__pydantic_config__` attribute by [@sydney-runkle](https://github.com/sydney-runkle) in [#10406](https://github.com/pydantic/pydantic/pull/10406) +* Recommend against using `Ellipsis` (...) with `Field` by [@Viicos](https://github.com/Viicos) in [#10661](https://github.com/pydantic/pydantic/pull/10661) +* Migrate to subclassing instead of annotated approach for pydantic url types by [@sydney-runkle](https://github.com/sydney-runkle) in [#10662](https://github.com/pydantic/pydantic/pull/10662) +* Change JSON schema generation of `Literal`s and `Enums` by [@Viicos](https://github.com/Viicos) in [#10692](https://github.com/pydantic/pydantic/pull/10692) +* Simplify unions involving `Any` or `Never` when replacing type variables by [@Viicos](https://github.com/Viicos) in [#10338](https://github.com/pydantic/pydantic/pull/10338) +* Do not require padding when decoding `base64` bytes by [@bschoenmaeckers](https://github.com/bschoenmaeckers) in [pydantic/pydantic-core#1448](https://github.com/pydantic/pydantic-core/pull/1448) +* Support dates all the way to 1BC by [@changhc](https://github.com/changhc) in [pydantic/speedate#77](https://github.com/pydantic/speedate/pull/77) + +#### Performance + +* Schema cleaning: skip unnecessary copies during schema walking by [@Viicos](https://github.com/Viicos) in [#10286](https://github.com/pydantic/pydantic/pull/10286) +* Refactor namespace logic for annotations evaluation by [@Viicos](https://github.com/Viicos) in [#10530](https://github.com/pydantic/pydantic/pull/10530) +* Improve email regexp on edge cases by [@AlekseyLobanov](https://github.com/AlekseyLobanov) in [#10601](https://github.com/pydantic/pydantic/pull/10601) +* `CoreMetadata` refactor with an emphasis on documentation, schema build time performance, and reducing complexity by [@sydney-runkle](https://github.com/sydney-runkle) in [#10675](https://github.com/pydantic/pydantic/pull/10675) + +#### Fixes + +* Remove guarding check on `computed_field` with `field_serializer` by [@nix010](https://github.com/nix010) in [#10390](https://github.com/pydantic/pydantic/pull/10390) +* Fix `Predicate` issue in `v2.9.0` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10321](https://github.com/pydantic/pydantic/pull/10321) +* Fixing `annotated-types` bound by [@sydney-runkle](https://github.com/sydney-runkle) in [#10327](https://github.com/pydantic/pydantic/pull/10327) +* Turn `tzdata` install requirement into optional `timezone` dependency by [@jakob-keller](https://github.com/jakob-keller) in [#10331](https://github.com/pydantic/pydantic/pull/10331) +* Use correct types namespace when building `namedtuple` core schemas by [@Viicos](https://github.com/Viicos) in [#10337](https://github.com/pydantic/pydantic/pull/10337) +* Fix evaluation of stringified annotations during namespace inspection by [@Viicos](https://github.com/Viicos) in [#10347](https://github.com/pydantic/pydantic/pull/10347) +* Fix `IncEx` type alias definition by [@Viicos](https://github.com/Viicos) in [#10339](https://github.com/pydantic/pydantic/pull/10339) +* Do not error when trying to evaluate annotations of private attributes by [@Viicos](https://github.com/Viicos) in [#10358](https://github.com/pydantic/pydantic/pull/10358) +* Fix nested type statement by [@kc0506](https://github.com/kc0506) in [#10369](https://github.com/pydantic/pydantic/pull/10369) +* Improve typing of `ModelMetaclass.mro` by [@Viicos](https://github.com/Viicos) in [#10372](https://github.com/pydantic/pydantic/pull/10372) +* Fix class access of deprecated `computed_field`s by [@Viicos](https://github.com/Viicos) in [#10391](https://github.com/pydantic/pydantic/pull/10391) +* Make sure `inspect.iscoroutinefunction` works on coroutines decorated with `@validate_call` by [@MovisLi](https://github.com/MovisLi) in [#10374](https://github.com/pydantic/pydantic/pull/10374) +* Fix `NameError` when using `validate_call` with PEP 695 on a class by [@kc0506](https://github.com/kc0506) in [#10380](https://github.com/pydantic/pydantic/pull/10380) +* Fix `ZoneInfo` with various invalid types by [@sydney-runkle](https://github.com/sydney-runkle) in [#10408](https://github.com/pydantic/pydantic/pull/10408) +* Fix `PydanticUserError` on empty `model_config` with annotations by [@cdwilson](https://github.com/cdwilson) in [#10412](https://github.com/pydantic/pydantic/pull/10412) +* Fix variance issue in `_IncEx` type alias, only allow `True` by [@Viicos](https://github.com/Viicos) in [#10414](https://github.com/pydantic/pydantic/pull/10414) +* Fix serialization schema generation when using `PlainValidator` by [@Viicos](https://github.com/Viicos) in [#10427](https://github.com/pydantic/pydantic/pull/10427) +* Fix schema generation error when serialization schema holds references by [@Viicos](https://github.com/Viicos) in [#10444](https://github.com/pydantic/pydantic/pull/10444) +* Inline references if possible when generating schema for `json_schema_input_type` by [@Viicos](https://github.com/Viicos) in [#10439](https://github.com/pydantic/pydantic/pull/10439) +* Fix recursive arguments in `Representation` by [@Viicos](https://github.com/Viicos) in [#10480](https://github.com/pydantic/pydantic/pull/10480) +* Fix representation for builtin function types by [@kschwab](https://github.com/kschwab) in [#10479](https://github.com/pydantic/pydantic/pull/10479) +* Add python validators for decimal constraints (`max_digits` and `decimal_places`) by [@sydney-runkle](https://github.com/sydney-runkle) in [#10506](https://github.com/pydantic/pydantic/pull/10506) +* Only fetch `__pydantic_core_schema__` from the current class during schema generation by [@Viicos](https://github.com/Viicos) in [#10518](https://github.com/pydantic/pydantic/pull/10518) +* Fix `stacklevel` on deprecation warnings for `BaseModel` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10520](https://github.com/pydantic/pydantic/pull/10520) +* Fix warning `stacklevel` in `BaseModel.__init__` by [@Viicos](https://github.com/Viicos) in [#10526](https://github.com/pydantic/pydantic/pull/10526) +* Improve error handling for in-evaluable refs for discriminator application by [@sydney-runkle](https://github.com/sydney-runkle) in [#10440](https://github.com/pydantic/pydantic/pull/10440) +* Change the signature of `ConfigWrapper.core_config` to take the title directly by [@Viicos](https://github.com/Viicos) in [#10562](https://github.com/pydantic/pydantic/pull/10562) +* Do not use the previous config from the stack for dataclasses without config by [@Viicos](https://github.com/Viicos) in [#10576](https://github.com/pydantic/pydantic/pull/10576) +* Fix serialization for IP types with `mode='python'` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10594](https://github.com/pydantic/pydantic/pull/10594) +* Support constraint application for `Base64Etc` types by [@sydney-runkle](https://github.com/sydney-runkle) in [#10584](https://github.com/pydantic/pydantic/pull/10584) +* Fix `validate_call` ignoring `Field` in `Annotated` by [@kc0506](https://github.com/kc0506) in [#10610](https://github.com/pydantic/pydantic/pull/10610) +* Raise an error when `Self` is invalid by [@kc0506](https://github.com/kc0506) in [#10609](https://github.com/pydantic/pydantic/pull/10609) +* Using `core_schema.InvalidSchema` instead of metadata injection + checks by [@sydney-runkle](https://github.com/sydney-runkle) in [#10523](https://github.com/pydantic/pydantic/pull/10523) +* Tweak type alias logic by [@kc0506](https://github.com/kc0506) in [#10643](https://github.com/pydantic/pydantic/pull/10643) +* Support usage of `type` with `typing.Self` and type aliases by [@kc0506](https://github.com/kc0506) in [#10621](https://github.com/pydantic/pydantic/pull/10621) +* Use overloads for `Field` and `PrivateAttr` functions by [@Viicos](https://github.com/Viicos) in [#10651](https://github.com/pydantic/pydantic/pull/10651) +* Clean up the `mypy` plugin implementation by [@Viicos](https://github.com/Viicos) in [#10669](https://github.com/pydantic/pydantic/pull/10669) +* Properly check for `typing_extensions` variant of `TypeAliasType` by [@Daraan](https://github.com/Daraan) in [#10713](https://github.com/pydantic/pydantic/pull/10713) +* Allow any mapping in `BaseModel.model_copy()` by [@Viicos](https://github.com/Viicos) in [#10751](https://github.com/pydantic/pydantic/pull/10751) +* Fix `isinstance` behavior for urls by [@sydney-runkle](https://github.com/sydney-runkle) in [#10766](https://github.com/pydantic/pydantic/pull/10766) +* Ensure `cached_property` can be set on Pydantic models by [@Viicos](https://github.com/Viicos) in [#10774](https://github.com/pydantic/pydantic/pull/10774) +* Fix equality checks for primitives in literals by [@sydney-runkle](https://github.com/sydney-runkle) in [pydantic/pydantic-core#1459](https://github.com/pydantic/pydantic-core/pull/1459) +* Properly enforce `host_required` for URLs by [@Viicos](https://github.com/Viicos) in [pydantic/pydantic-core#1488](https://github.com/pydantic/pydantic-core/pull/1488) +* Fix when `coerce_numbers_to_str` enabled and string has invalid Unicode character by [@andrey-berenda](https://github.com/andrey-berenda) in [pydantic/pydantic-core#1515](https://github.com/pydantic/pydantic-core/pull/1515) +* Fix serializing `complex` values in `Enum`s by [@changhc](https://github.com/changhc) in [pydantic/pydantic-core#1524](https://github.com/pydantic/pydantic-core/pull/1524) +* Refactor `_typing_extra` module by [@Viicos](https://github.com/Viicos) in [#10725](https://github.com/pydantic/pydantic/pull/10725) +* Support intuitive equality for urls by [@sydney-runkle](https://github.com/sydney-runkle) in [#10798](https://github.com/pydantic/pydantic/pull/10798) +* Add `bytearray` to `TypeAdapter.validate_json` signature by [@samuelcolvin](https://github.com/samuelcolvin) in [#10802](https://github.com/pydantic/pydantic/pull/10802) +* Ensure class access of method descriptors is performed when used as a default with `Field` by [@Viicos](https://github.com/Viicos) in [#10816](https://github.com/pydantic/pydantic/pull/10816) +* Fix circular import with `validate_call` by [@sydney-runkle](https://github.com/sydney-runkle) in [#10807](https://github.com/pydantic/pydantic/pull/10807) +* Fix error when using type aliases referencing other type aliases by [@Viicos](https://github.com/Viicos) in [#10809](https://github.com/pydantic/pydantic/pull/10809) +* Fix `IncEx` type alias to be compatible with mypy by [@Viicos](https://github.com/Viicos) in [#10813](https://github.com/pydantic/pydantic/pull/10813) +* Make `__signature__` a lazy property, do not deepcopy defaults by [@Viicos](https://github.com/Viicos) in [#10818](https://github.com/pydantic/pydantic/pull/10818) +* Make `__signature__` lazy for dataclasses, too by [@sydney-runkle](https://github.com/sydney-runkle) in [#10832](https://github.com/pydantic/pydantic/pull/10832) +* Subclass all single host url classes from `AnyUrl` to preserve behavior from v2.9 by [@sydney-runkle](https://github.com/sydney-runkle) in [#10856](https://github.com/pydantic/pydantic/pull/10856) + +### New Contributors + +* [@jakob-keller](https://github.com/jakob-keller) made their first contribution in [#10331](https://github.com/pydantic/pydantic/pull/10331) +* [@MovisLi](https://github.com/MovisLi) made their first contribution in [#10374](https://github.com/pydantic/pydantic/pull/10374) +* [@joaopalmeiro](https://github.com/joaopalmeiro) made their first contribution in [#10405](https://github.com/pydantic/pydantic/pull/10405) +* [@theunkn0wn1](https://github.com/theunkn0wn1) made their first contribution in [#10378](https://github.com/pydantic/pydantic/pull/10378) +* [@cdwilson](https://github.com/cdwilson) made their first contribution in [#10412](https://github.com/pydantic/pydantic/pull/10412) +* [@dlax](https://github.com/dlax) made their first contribution in [#10421](https://github.com/pydantic/pydantic/pull/10421) +* [@kschwab](https://github.com/kschwab) made their first contribution in [#10479](https://github.com/pydantic/pydantic/pull/10479) +* [@santibreo](https://github.com/santibreo) made their first contribution in [#10453](https://github.com/pydantic/pydantic/pull/10453) +* [@FlorianSW](https://github.com/FlorianSW) made their first contribution in [#10478](https://github.com/pydantic/pydantic/pull/10478) +* [@tkasuz](https://github.com/tkasuz) made their first contribution in [#10555](https://github.com/pydantic/pydantic/pull/10555) +* [@AlekseyLobanov](https://github.com/AlekseyLobanov) made their first contribution in [#10601](https://github.com/pydantic/pydantic/pull/10601) +* [@NiclasvanEyk](https://github.com/NiclasvanEyk) made their first contribution in [#10667](https://github.com/pydantic/pydantic/pull/10667) +* [@mschoettle](https://github.com/mschoettle) made their first contribution in [#10677](https://github.com/pydantic/pydantic/pull/10677) +* [@Daraan](https://github.com/Daraan) made their first contribution in [#10713](https://github.com/pydantic/pydantic/pull/10713) +* [@k4nar](https://github.com/k4nar) made their first contribution in [#10736](https://github.com/pydantic/pydantic/pull/10736) +* [@UriyaHarpeness](https://github.com/UriyaHarpeness) made their first contribution in [#10740](https://github.com/pydantic/pydantic/pull/10740) +* [@frfahim](https://github.com/frfahim) made their first contribution in [#10727](https://github.com/pydantic/pydantic/pull/10727) + +## v2.10.0b2 (2024-11-13) + +Pre-release, see [the GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.10.0b2) for details. + +## v2.10.0b1 (2024-11-06) + +Pre-release, see [the GitHub release](https://github.com/pydantic/pydantic/releases/tag/v2.9.0b1) for details. + + +... see [here](https://docs.pydantic.dev/changelog/#v0322-2019-08-17) for earlier changes. diff --git a/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/RECORD b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/RECORD new file mode 100644 index 0000000000000000000000000000000000000000..74a02865e1841ddd3fdfb5114dc1964884554f31 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/RECORD @@ -0,0 +1,213 @@ +pydantic-2.10.6.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4 +pydantic-2.10.6.dist-info/METADATA,sha256=foVlFz39Sq5b57ca5MBqkdoknQXuG2SoWL6630aoDZc,30869 +pydantic-2.10.6.dist-info/RECORD,, +pydantic-2.10.6.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87 +pydantic-2.10.6.dist-info/licenses/LICENSE,sha256=qeGG88oWte74QxjnpwFyE1GgDLe4rjpDlLZ7SeNSnvM,1129 +pydantic/__init__.py,sha256=kJyWIdXaLtG7NZb6f0Phm86MuNA1E8Xdv0cHvNCx9_A,14860 +pydantic/__pycache__/__init__.cpython-311.pyc,, +pydantic/__pycache__/_migration.cpython-311.pyc,, +pydantic/__pycache__/alias_generators.cpython-311.pyc,, +pydantic/__pycache__/aliases.cpython-311.pyc,, +pydantic/__pycache__/annotated_handlers.cpython-311.pyc,, +pydantic/__pycache__/class_validators.cpython-311.pyc,, +pydantic/__pycache__/color.cpython-311.pyc,, +pydantic/__pycache__/config.cpython-311.pyc,, +pydantic/__pycache__/dataclasses.cpython-311.pyc,, +pydantic/__pycache__/datetime_parse.cpython-311.pyc,, +pydantic/__pycache__/decorator.cpython-311.pyc,, +pydantic/__pycache__/env_settings.cpython-311.pyc,, +pydantic/__pycache__/error_wrappers.cpython-311.pyc,, +pydantic/__pycache__/errors.cpython-311.pyc,, +pydantic/__pycache__/fields.cpython-311.pyc,, +pydantic/__pycache__/functional_serializers.cpython-311.pyc,, +pydantic/__pycache__/functional_validators.cpython-311.pyc,, +pydantic/__pycache__/generics.cpython-311.pyc,, +pydantic/__pycache__/json.cpython-311.pyc,, +pydantic/__pycache__/json_schema.cpython-311.pyc,, +pydantic/__pycache__/main.cpython-311.pyc,, +pydantic/__pycache__/mypy.cpython-311.pyc,, +pydantic/__pycache__/networks.cpython-311.pyc,, +pydantic/__pycache__/parse.cpython-311.pyc,, +pydantic/__pycache__/root_model.cpython-311.pyc,, +pydantic/__pycache__/schema.cpython-311.pyc,, +pydantic/__pycache__/tools.cpython-311.pyc,, +pydantic/__pycache__/type_adapter.cpython-311.pyc,, +pydantic/__pycache__/types.cpython-311.pyc,, +pydantic/__pycache__/typing.cpython-311.pyc,, +pydantic/__pycache__/utils.cpython-311.pyc,, +pydantic/__pycache__/validate_call_decorator.cpython-311.pyc,, +pydantic/__pycache__/validators.cpython-311.pyc,, +pydantic/__pycache__/version.cpython-311.pyc,, +pydantic/__pycache__/warnings.cpython-311.pyc,, +pydantic/_internal/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +pydantic/_internal/__pycache__/__init__.cpython-311.pyc,, +pydantic/_internal/__pycache__/_config.cpython-311.pyc,, +pydantic/_internal/__pycache__/_core_metadata.cpython-311.pyc,, +pydantic/_internal/__pycache__/_core_utils.cpython-311.pyc,, +pydantic/_internal/__pycache__/_dataclasses.cpython-311.pyc,, +pydantic/_internal/__pycache__/_decorators.cpython-311.pyc,, +pydantic/_internal/__pycache__/_decorators_v1.cpython-311.pyc,, +pydantic/_internal/__pycache__/_discriminated_union.cpython-311.pyc,, +pydantic/_internal/__pycache__/_docs_extraction.cpython-311.pyc,, +pydantic/_internal/__pycache__/_fields.cpython-311.pyc,, +pydantic/_internal/__pycache__/_forward_ref.cpython-311.pyc,, +pydantic/_internal/__pycache__/_generate_schema.cpython-311.pyc,, +pydantic/_internal/__pycache__/_generics.cpython-311.pyc,, +pydantic/_internal/__pycache__/_git.cpython-311.pyc,, +pydantic/_internal/__pycache__/_import_utils.cpython-311.pyc,, +pydantic/_internal/__pycache__/_internal_dataclass.cpython-311.pyc,, +pydantic/_internal/__pycache__/_known_annotated_metadata.cpython-311.pyc,, +pydantic/_internal/__pycache__/_mock_val_ser.cpython-311.pyc,, +pydantic/_internal/__pycache__/_model_construction.cpython-311.pyc,, +pydantic/_internal/__pycache__/_namespace_utils.cpython-311.pyc,, +pydantic/_internal/__pycache__/_repr.cpython-311.pyc,, +pydantic/_internal/__pycache__/_schema_generation_shared.cpython-311.pyc,, +pydantic/_internal/__pycache__/_serializers.cpython-311.pyc,, +pydantic/_internal/__pycache__/_signature.cpython-311.pyc,, +pydantic/_internal/__pycache__/_std_types_schema.cpython-311.pyc,, +pydantic/_internal/__pycache__/_typing_extra.cpython-311.pyc,, +pydantic/_internal/__pycache__/_utils.cpython-311.pyc,, +pydantic/_internal/__pycache__/_validate_call.cpython-311.pyc,, +pydantic/_internal/__pycache__/_validators.cpython-311.pyc,, +pydantic/_internal/_config.py,sha256=HsKc9guTGHyRdseuTA1gIBSfwAV_psyhEh5M_vzCPP0,12548 +pydantic/_internal/_core_metadata.py,sha256=S5T76SgR9rJWzsZDTLBKx6LoTnIK5MRbsty0VttgB7M,4655 +pydantic/_internal/_core_utils.py,sha256=MKWMqlEdKTYdNB_hMfpEGTCAKQ1L9DVdXMB1s_KKueQ,26572 +pydantic/_internal/_dataclasses.py,sha256=x6sVAFvWfTpyePPd5gflPsedM0kIB-r4TX7M_UEBh3A,9486 +pydantic/_internal/_decorators.py,sha256=xWTj0zYYpk2KTuWeWjTIGDEwDqSS7rYDoU9k8dQ4lis,32304 +pydantic/_internal/_decorators_v1.py,sha256=bkjIhVdah-M9V3Y4laICdD1B4cFwS7f03ewvvf5JnFY,6198 +pydantic/_internal/_discriminated_union.py,sha256=eUu06ze9wOIe3WQmilssyoOGsdJ_L8I3E1owpeod5ls,26446 +pydantic/_internal/_docs_extraction.py,sha256=bIWhw7nFKFt-qD-txtKRAb5VqGlAD0H9YzEEXE3PHj4,3791 +pydantic/_internal/_fields.py,sha256=xgz-p1rNwxhjbblXNZmMCXH_OsbKKbcmaurDdVIOHN8,17060 +pydantic/_internal/_forward_ref.py,sha256=5n3Y7-3AKLn8_FS3Yc7KutLiPUhyXmAtkEZOaFnonwM,611 +pydantic/_internal/_generate_schema.py,sha256=oIRpGlPWbHcy1e1YtIGUEhrOinW0ZnDAuEKdo1VY6iM,114118 +pydantic/_internal/_generics.py,sha256=RTf16ckBDtdwwX5zeFz7bTj4CxurtWC6DGQgfGMy9e8,22746 +pydantic/_internal/_git.py,sha256=lN6QlZm8RNSuNsUdHHC1F5a4VToe8vu5tlUxAfaJgGE,784 +pydantic/_internal/_import_utils.py,sha256=eLe9Aa1GMQB4TD9sKz42yyJBTiTAjBfCuua_rYFTrQo,448 +pydantic/_internal/_internal_dataclass.py,sha256=_bedc1XbuuygRGiLZqkUkwwFpQaoR1hKLlR501nyySY,144 +pydantic/_internal/_known_annotated_metadata.py,sha256=EhAj2V2SMkOvTRII01dJDeSasUY2JTaEELJIoxG23YM,16186 +pydantic/_internal/_mock_val_ser.py,sha256=Z0ipRGcU_EhVyiJjkAhX2YS6ibRNbuIf2Ng-Y8BpIT0,9166 +pydantic/_internal/_model_construction.py,sha256=J5DsSj-fhXg6kZR-xm4aoDeu6t3xF557-O6qd6uuBJY,34589 +pydantic/_internal/_namespace_utils.py,sha256=ls9Tp10aYV6Ce49kuaz5SyOry_7YL21K45_ay_RqPXw,12024 +pydantic/_internal/_repr.py,sha256=AP0I2216KzlbKR47WxjTarBitIWWvbj4hN6yfXYznkU,4960 +pydantic/_internal/_schema_generation_shared.py,sha256=fgjsYtE0mxRN6Ip7wt02V7mMSHcmD9zNHPoplGRtUrk,4897 +pydantic/_internal/_serializers.py,sha256=s0peFJTaHTfdmr4V-l6n3Dh-57ZHeUo8oJcq3NOjJNI,1356 +pydantic/_internal/_signature.py,sha256=6kwSolb53Y6NMq2RPRWMLwa1eBpnza6duJQqGC0Sea4,6779 +pydantic/_internal/_std_types_schema.py,sha256=XrPtBN0H9iBfkRDDallczOyrvBnM0pOa2CxEkt6KnWw,16163 +pydantic/_internal/_typing_extra.py,sha256=lBl9wTpPdgn-xfTH7y0rX-E9sLPDnp0CHN6rUo-Bk10,33396 +pydantic/_internal/_utils.py,sha256=BAFf8ql0hnTeVhH1L87amvS6L-DjsjXWfiCqfuD4uAI,13537 +pydantic/_internal/_validate_call.py,sha256=6EuHEQeYzr7A1oH_DYl68RpkkhQ9Rz0PpgsgF-HbH_0,4536 +pydantic/_internal/_validators.py,sha256=YpveqmG9FkuRedOaVhUaEzXsJsf39GHCeNhMNDdsZC4,15933 +pydantic/_migration.py,sha256=j6TbRpJofjAX8lr-k2nVnQcBR9RD2B91I7Ulcw_ZzEo,11913 +pydantic/alias_generators.py,sha256=KM1n3u4JfLSBl1UuYg3hoYHzXJD-yvgrnq8u1ccwh_A,2124 +pydantic/aliases.py,sha256=lzfmL-dZUh9b9kGXqk0ai3yviUZzY61rs2sgDOeWrnQ,4822 +pydantic/annotated_handlers.py,sha256=WfyFSqwoEIFXBh7T73PycKloI1DiX45GWi0-JOsCR4Y,4407 +pydantic/class_validators.py,sha256=i_V3j-PYdGLSLmj_IJZekTRjunO8SIVz8LMlquPyP7E,148 +pydantic/color.py,sha256=4GrtPvFCBKdM-1NpLVFOC7KkLejyZd1BiELfCKvT2yw,21494 +pydantic/config.py,sha256=iI9sEiVQPwW6LshLUzbxTcNyaDyeOBm7KsmgNhyll7U,35611 +pydantic/dataclasses.py,sha256=S7MOr6CzM_yhx9tlK8rsuGzTIy2OcFdJSPe5c6g0vEg,15900 +pydantic/datetime_parse.py,sha256=QC-WgMxMr_wQ_mNXUS7AVf-2hLEhvvsPY1PQyhSGOdk,150 +pydantic/decorator.py,sha256=YX-jUApu5AKaVWKPoaV-n-4l7UbS69GEt9Ra3hszmKI,145 +pydantic/deprecated/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +pydantic/deprecated/__pycache__/__init__.cpython-311.pyc,, +pydantic/deprecated/__pycache__/class_validators.cpython-311.pyc,, +pydantic/deprecated/__pycache__/config.cpython-311.pyc,, +pydantic/deprecated/__pycache__/copy_internals.cpython-311.pyc,, +pydantic/deprecated/__pycache__/decorator.cpython-311.pyc,, +pydantic/deprecated/__pycache__/json.cpython-311.pyc,, +pydantic/deprecated/__pycache__/parse.cpython-311.pyc,, +pydantic/deprecated/__pycache__/tools.cpython-311.pyc,, +pydantic/deprecated/class_validators.py,sha256=IARV4v0NuHCZVVc_ItfaVNJOk1Vful7iv8bawj9o4Uw,10245 +pydantic/deprecated/config.py,sha256=eKhnG--ZQtJ4A7KA3xeF76E15-4pVau3B5T8D39ptFs,2663 +pydantic/deprecated/copy_internals.py,sha256=2y3G0pAJMuahxIvN4IJzaE9hzN1C81h2KQspcWB4RuU,7630 +pydantic/deprecated/decorator.py,sha256=Fa9ou1uUmR3NvyJsPSFGfQLI4n27y4Bs2LhlkDPmNmA,10843 +pydantic/deprecated/json.py,sha256=PIzc-zAZxL-7O2gW7tY986XASvgsgD-J1D5pmhd-JLM,4669 +pydantic/deprecated/parse.py,sha256=Gzd6b_g8zJXcuE7QRq5adhx_EMJahXfcpXCF0RgrqqI,2511 +pydantic/deprecated/tools.py,sha256=XUoIW9W4sgOUWQ6Xzf-Z_NukUC1l_yUwz2_n0fE3MEI,3336 +pydantic/env_settings.py,sha256=6IHeeWEqlUPRUv3V-AXiF_W91fg2Jw_M3O0l34J_eyA,148 +pydantic/error_wrappers.py,sha256=RK6mqATc9yMD-KBD9IJS9HpKCprWHd8wo84Bnm-3fR8,150 +pydantic/errors.py,sha256=axy4Uzfdqme7zjLXmqUM-Q3Z8lpit7MNn7PsZ7WLu80,5009 +pydantic/experimental/__init__.py,sha256=j08eROfz-xW4k_X9W4m2AW26IVdyF3Eg1OzlIGA11vk,328 +pydantic/experimental/__pycache__/__init__.cpython-311.pyc,, +pydantic/experimental/__pycache__/pipeline.cpython-311.pyc,, +pydantic/experimental/pipeline.py,sha256=PjiTnmaN5XwLUk-Nlzfr2C2R-xrMz7Nz-XVC8kiFGyU,23979 +pydantic/fields.py,sha256=8HbAEdQv6DMV7DH6oj5erOLpLPORf0rTbKLWd83FLQI,62631 +pydantic/functional_serializers.py,sha256=gqOVOETSVZfQqZg6jrkFSA1yMXCqVwpkaVy-oDNsKJA,17005 +pydantic/functional_validators.py,sha256=fAacaUMaJRKCbZfsGXtmkg469PpLNTLEITJLwUHnMmc,29430 +pydantic/generics.py,sha256=0ZqZ9O9annIj_3mGBRqps4htey3b5lV1-d2tUxPMMnA,144 +pydantic/json.py,sha256=ZH8RkI7h4Bz-zp8OdTAxbJUoVvcoU-jhMdRZ0B-k0xc,140 +pydantic/json_schema.py,sha256=W9Ae4B2cGDpBJCyzCOm6-l6eUHHrVRjU4ZoSAiasNao,113117 +pydantic/main.py,sha256=gwm18587VN9pS2GSDeiwzl2VNSpuhxzPM-D-07vI_BI,76200 +pydantic/mypy.py,sha256=70iw65os6WJuuVQvNMqzZDehRAuVRL-79NVr0B8NuEc,56366 +pydantic/networks.py,sha256=wjJtwXBZOTVt5uRNQncB5MMGLL_n7oyo5U-EzYaeu0U,40855 +pydantic/parse.py,sha256=wkd82dgtvWtD895U_I6E1htqMlGhBSYEV39cuBSeo3A,141 +pydantic/plugin/__init__.py,sha256=63AkDEwbqLQKtN6Pj9BSTnRp1iOyTNv0KrXP3N0dHZ0,6117 +pydantic/plugin/__pycache__/__init__.cpython-311.pyc,, +pydantic/plugin/__pycache__/_loader.cpython-311.pyc,, +pydantic/plugin/__pycache__/_schema_validator.cpython-311.pyc,, +pydantic/plugin/_loader.py,sha256=rmLbIwThDmVR1JwFVi_XvrLH7b1A5teMED-O3pr6Gk4,2140 +pydantic/plugin/_schema_validator.py,sha256=VFaNQpVNSuI2ymRDkTwBGaMKeKmySk1TbW-3rQeozxk,5240 +pydantic/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +pydantic/root_model.py,sha256=g4SGnHIlXlQZAJG-udwt7BFjq82_D_2U3W_LlkCcCCI,6231 +pydantic/schema.py,sha256=Vqqjvq_LnapVknebUd3Bp_J1p2gXZZnZRgL48bVEG7o,142 +pydantic/tools.py,sha256=iHQpd8SJ5DCTtPV5atAV06T89bjSaMFeZZ2LX9lasZY,141 +pydantic/type_adapter.py,sha256=j7-FPYjWJXFVYozJ5qwOmmBGO349gVbnQ8aIdvf_8sU,28730 +pydantic/types.py,sha256=UMUaWU3O_bdrH57JDSN1G34BvqVpAudW6Y7GoN1u7P8,103958 +pydantic/typing.py,sha256=P7feA35MwTcLsR1uL7db0S-oydBxobmXa55YDoBgajQ,138 +pydantic/utils.py,sha256=15nR2QpqTBFlQV4TNtTItMyTJx_fbyV-gPmIEY1Gooc,141 +pydantic/v1/__init__.py,sha256=SxQPklgBs4XHJwE6BZ9qoewYoGiNyYUnmHzEFCZbfnI,2946 +pydantic/v1/__pycache__/__init__.cpython-311.pyc,, +pydantic/v1/__pycache__/_hypothesis_plugin.cpython-311.pyc,, +pydantic/v1/__pycache__/annotated_types.cpython-311.pyc,, +pydantic/v1/__pycache__/class_validators.cpython-311.pyc,, +pydantic/v1/__pycache__/color.cpython-311.pyc,, +pydantic/v1/__pycache__/config.cpython-311.pyc,, +pydantic/v1/__pycache__/dataclasses.cpython-311.pyc,, +pydantic/v1/__pycache__/datetime_parse.cpython-311.pyc,, +pydantic/v1/__pycache__/decorator.cpython-311.pyc,, +pydantic/v1/__pycache__/env_settings.cpython-311.pyc,, +pydantic/v1/__pycache__/error_wrappers.cpython-311.pyc,, +pydantic/v1/__pycache__/errors.cpython-311.pyc,, +pydantic/v1/__pycache__/fields.cpython-311.pyc,, +pydantic/v1/__pycache__/generics.cpython-311.pyc,, +pydantic/v1/__pycache__/json.cpython-311.pyc,, +pydantic/v1/__pycache__/main.cpython-311.pyc,, +pydantic/v1/__pycache__/mypy.cpython-311.pyc,, +pydantic/v1/__pycache__/networks.cpython-311.pyc,, +pydantic/v1/__pycache__/parse.cpython-311.pyc,, +pydantic/v1/__pycache__/schema.cpython-311.pyc,, +pydantic/v1/__pycache__/tools.cpython-311.pyc,, +pydantic/v1/__pycache__/types.cpython-311.pyc,, +pydantic/v1/__pycache__/typing.cpython-311.pyc,, +pydantic/v1/__pycache__/utils.cpython-311.pyc,, +pydantic/v1/__pycache__/validators.cpython-311.pyc,, +pydantic/v1/__pycache__/version.cpython-311.pyc,, +pydantic/v1/_hypothesis_plugin.py,sha256=5ES5xWuw1FQAsymLezy8QgnVz0ZpVfU3jkmT74H27VQ,14847 +pydantic/v1/annotated_types.py,sha256=uk2NAAxqiNELKjiHhyhxKaIOh8F1lYW_LzrW3X7oZBc,3157 +pydantic/v1/class_validators.py,sha256=ULOaIUgYUDBsHL7EEVEarcM-UubKUggoN8hSbDonsFE,14672 +pydantic/v1/color.py,sha256=iZABLYp6OVoo2AFkP9Ipri_wSc6-Kklu8YuhSartd5g,16844 +pydantic/v1/config.py,sha256=a6P0Wer9x4cbwKW7Xv8poSUqM4WP-RLWwX6YMpYq9AA,6532 +pydantic/v1/dataclasses.py,sha256=784cqvInbwIPWr9usfpX3ch7z4t3J2tTK6N067_wk1o,18172 +pydantic/v1/datetime_parse.py,sha256=4Qy1kQpq3rNVZJeIHeSPDpuS2Bvhp1KPtzJG1xu-H00,7724 +pydantic/v1/decorator.py,sha256=zaaxxxoWPCm818D1bs0yhapRjXm32V8G0ZHWCdM1uXA,10339 +pydantic/v1/env_settings.py,sha256=A9VXwtRl02AY-jH0C0ouy5VNw3fi6F_pkzuHDjgAAOM,14105 +pydantic/v1/error_wrappers.py,sha256=6625Mfw9qkC2NwitB_JFAWe8B-Xv6zBU7rL9k28tfyo,5196 +pydantic/v1/errors.py,sha256=mIwPED5vGM5Q5v4C4Z1JPldTRH-omvEylH6ksMhOmPw,17726 +pydantic/v1/fields.py,sha256=VqWJCriUNiEyptXroDVJ501JpVA0en2VANcksqXL2b8,50649 +pydantic/v1/generics.py,sha256=VzC9YUV-EbPpQ3aAfk1cNFej79_IzznkQ7WrmTTZS9E,17871 +pydantic/v1/json.py,sha256=WQ5Hy_hIpfdR3YS8k6N2E6KMJzsdbBi_ldWOPJaV81M,3390 +pydantic/v1/main.py,sha256=nFvpoUH2UKAe2vzw7wVUNkXBeIUe1Xs4EIv-JBqSo2k,44555 +pydantic/v1/mypy.py,sha256=AiZYkv127-WsgL9vwvLqj0dS8dz-HUMbH9Yvvlq4bfE,38949 +pydantic/v1/networks.py,sha256=HYNtKAfOmOnKJpsDg1g6SIkj9WPhU_-i8l5e2JKBpG4,22124 +pydantic/v1/parse.py,sha256=BJtdqiZRtav9VRFCmOxoY-KImQmjPy-A_NoojiFUZxY,1821 +pydantic/v1/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +pydantic/v1/schema.py,sha256=aqBuA--cq8gAVkim5BJPFASHzOZ8dFtmFX_fNGr6ip4,47801 +pydantic/v1/tools.py,sha256=1lDdXHk0jL5uP3u5RCYAvUAlGClgAO-45lkq9j7fyBA,2881 +pydantic/v1/types.py,sha256=Fltx5GoP_qaUmAktlGz7nFeJa13yNy3FY1-RcMzEVt8,35455 +pydantic/v1/typing.py,sha256=GjThObaqHMhLaECzYUrDk0X-RHjo7x6vsv4Z4qUYV8I,19387 +pydantic/v1/utils.py,sha256=fvjXCNeaU1jB-_mi0SUk4fzN4OpvX-SKTu4yJMPRIlQ,25941 +pydantic/v1/validators.py,sha256=lyUkn1MWhHxlCX5ZfEgFj_CAHojoiPcaQeMdEM9XviU,22187 +pydantic/v1/version.py,sha256=IHqnBuD5DuqB3Pxzw1L8EJsYheQBzxaIZeCZkPvI7Rk,1039 +pydantic/validate_call_decorator.py,sha256=ebcbCp_DXffYnfrBFWTv6ygbOuGBwLzcj_JupG8Eut8,4360 +pydantic/validators.py,sha256=pwbIJXVb1CV2mAE4w_EGfNj7DwzsKaWw_tTL6cviTus,146 +pydantic/version.py,sha256=ZNnF4reJoKeuyPxDYIKibiY9a-9I_yRp-0fYEJD_mqg,2498 +pydantic/warnings.py,sha256=5Ok3GxyP1LkcEvONxqjQkuzFaOuly7QdCUhmEpFM5Lk,3350 diff --git a/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/WHEEL b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/WHEEL new file mode 100644 index 0000000000000000000000000000000000000000..12228d414b6cfed7c39d3781c85c63256a1d7fb5 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/WHEEL @@ -0,0 +1,4 @@ +Wheel-Version: 1.0 +Generator: hatchling 1.27.0 +Root-Is-Purelib: true +Tag: py3-none-any diff --git a/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/licenses/LICENSE b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/licenses/LICENSE new file mode 100644 index 0000000000000000000000000000000000000000..488c6260c10f2e88fa1fae58a63fccec8d600cd1 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/pydantic-2.10.6.dist-info/licenses/LICENSE @@ -0,0 +1,21 @@ +The MIT License (MIT) + +Copyright (c) 2017 to present Pydantic Services Inc. and individual contributors. + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/.venv/lib/python3.11/site-packages/urllib3/_collections.py b/.venv/lib/python3.11/site-packages/urllib3/_collections.py new file mode 100644 index 0000000000000000000000000000000000000000..1b6c1364213b990c716c39d7cc6427cb319cb948 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/_collections.py @@ -0,0 +1,479 @@ +from __future__ import annotations + +import typing +from collections import OrderedDict +from enum import Enum, auto +from threading import RLock + +if typing.TYPE_CHECKING: + # We can only import Protocol if TYPE_CHECKING because it's a development + # dependency, and is not available at runtime. + from typing import Protocol + + from typing_extensions import Self + + class HasGettableStringKeys(Protocol): + def keys(self) -> typing.Iterator[str]: ... + + def __getitem__(self, key: str) -> str: ... + + +__all__ = ["RecentlyUsedContainer", "HTTPHeaderDict"] + + +# Key type +_KT = typing.TypeVar("_KT") +# Value type +_VT = typing.TypeVar("_VT") +# Default type +_DT = typing.TypeVar("_DT") + +ValidHTTPHeaderSource = typing.Union[ + "HTTPHeaderDict", + typing.Mapping[str, str], + typing.Iterable[tuple[str, str]], + "HasGettableStringKeys", +] + + +class _Sentinel(Enum): + not_passed = auto() + + +def ensure_can_construct_http_header_dict( + potential: object, +) -> ValidHTTPHeaderSource | None: + if isinstance(potential, HTTPHeaderDict): + return potential + elif isinstance(potential, typing.Mapping): + # Full runtime checking of the contents of a Mapping is expensive, so for the + # purposes of typechecking, we assume that any Mapping is the right shape. + return typing.cast(typing.Mapping[str, str], potential) + elif isinstance(potential, typing.Iterable): + # Similarly to Mapping, full runtime checking of the contents of an Iterable is + # expensive, so for the purposes of typechecking, we assume that any Iterable + # is the right shape. + return typing.cast(typing.Iterable[tuple[str, str]], potential) + elif hasattr(potential, "keys") and hasattr(potential, "__getitem__"): + return typing.cast("HasGettableStringKeys", potential) + else: + return None + + +class RecentlyUsedContainer(typing.Generic[_KT, _VT], typing.MutableMapping[_KT, _VT]): + """ + Provides a thread-safe dict-like container which maintains up to + ``maxsize`` keys while throwing away the least-recently-used keys beyond + ``maxsize``. + + :param maxsize: + Maximum number of recent elements to retain. + + :param dispose_func: + Every time an item is evicted from the container, + ``dispose_func(value)`` is called. Callback which will get called + """ + + _container: typing.OrderedDict[_KT, _VT] + _maxsize: int + dispose_func: typing.Callable[[_VT], None] | None + lock: RLock + + def __init__( + self, + maxsize: int = 10, + dispose_func: typing.Callable[[_VT], None] | None = None, + ) -> None: + super().__init__() + self._maxsize = maxsize + self.dispose_func = dispose_func + self._container = OrderedDict() + self.lock = RLock() + + def __getitem__(self, key: _KT) -> _VT: + # Re-insert the item, moving it to the end of the eviction line. + with self.lock: + item = self._container.pop(key) + self._container[key] = item + return item + + def __setitem__(self, key: _KT, value: _VT) -> None: + evicted_item = None + with self.lock: + # Possibly evict the existing value of 'key' + try: + # If the key exists, we'll overwrite it, which won't change the + # size of the pool. Because accessing a key should move it to + # the end of the eviction line, we pop it out first. + evicted_item = key, self._container.pop(key) + self._container[key] = value + except KeyError: + # When the key does not exist, we insert the value first so that + # evicting works in all cases, including when self._maxsize is 0 + self._container[key] = value + if len(self._container) > self._maxsize: + # If we didn't evict an existing value, and we've hit our maximum + # size, then we have to evict the least recently used item from + # the beginning of the container. + evicted_item = self._container.popitem(last=False) + + # After releasing the lock on the pool, dispose of any evicted value. + if evicted_item is not None and self.dispose_func: + _, evicted_value = evicted_item + self.dispose_func(evicted_value) + + def __delitem__(self, key: _KT) -> None: + with self.lock: + value = self._container.pop(key) + + if self.dispose_func: + self.dispose_func(value) + + def __len__(self) -> int: + with self.lock: + return len(self._container) + + def __iter__(self) -> typing.NoReturn: + raise NotImplementedError( + "Iteration over this class is unlikely to be threadsafe." + ) + + def clear(self) -> None: + with self.lock: + # Copy pointers to all values, then wipe the mapping + values = list(self._container.values()) + self._container.clear() + + if self.dispose_func: + for value in values: + self.dispose_func(value) + + def keys(self) -> set[_KT]: # type: ignore[override] + with self.lock: + return set(self._container.keys()) + + +class HTTPHeaderDictItemView(set[tuple[str, str]]): + """ + HTTPHeaderDict is unusual for a Mapping[str, str] in that it has two modes of + address. + + If we directly try to get an item with a particular name, we will get a string + back that is the concatenated version of all the values: + + >>> d['X-Header-Name'] + 'Value1, Value2, Value3' + + However, if we iterate over an HTTPHeaderDict's items, we will optionally combine + these values based on whether combine=True was called when building up the dictionary + + >>> d = HTTPHeaderDict({"A": "1", "B": "foo"}) + >>> d.add("A", "2", combine=True) + >>> d.add("B", "bar") + >>> list(d.items()) + [ + ('A', '1, 2'), + ('B', 'foo'), + ('B', 'bar'), + ] + + This class conforms to the interface required by the MutableMapping ABC while + also giving us the nonstandard iteration behavior we want; items with duplicate + keys, ordered by time of first insertion. + """ + + _headers: HTTPHeaderDict + + def __init__(self, headers: HTTPHeaderDict) -> None: + self._headers = headers + + def __len__(self) -> int: + return len(list(self._headers.iteritems())) + + def __iter__(self) -> typing.Iterator[tuple[str, str]]: + return self._headers.iteritems() + + def __contains__(self, item: object) -> bool: + if isinstance(item, tuple) and len(item) == 2: + passed_key, passed_val = item + if isinstance(passed_key, str) and isinstance(passed_val, str): + return self._headers._has_value_for_header(passed_key, passed_val) + return False + + +class HTTPHeaderDict(typing.MutableMapping[str, str]): + """ + :param headers: + An iterable of field-value pairs. Must not contain multiple field names + when compared case-insensitively. + + :param kwargs: + Additional field-value pairs to pass in to ``dict.update``. + + A ``dict`` like container for storing HTTP Headers. + + Field names are stored and compared case-insensitively in compliance with + RFC 7230. Iteration provides the first case-sensitive key seen for each + case-insensitive pair. + + Using ``__setitem__`` syntax overwrites fields that compare equal + case-insensitively in order to maintain ``dict``'s api. For fields that + compare equal, instead create a new ``HTTPHeaderDict`` and use ``.add`` + in a loop. + + If multiple fields that are equal case-insensitively are passed to the + constructor or ``.update``, the behavior is undefined and some will be + lost. + + >>> headers = HTTPHeaderDict() + >>> headers.add('Set-Cookie', 'foo=bar') + >>> headers.add('set-cookie', 'baz=quxx') + >>> headers['content-length'] = '7' + >>> headers['SET-cookie'] + 'foo=bar, baz=quxx' + >>> headers['Content-Length'] + '7' + """ + + _container: typing.MutableMapping[str, list[str]] + + def __init__(self, headers: ValidHTTPHeaderSource | None = None, **kwargs: str): + super().__init__() + self._container = {} # 'dict' is insert-ordered + if headers is not None: + if isinstance(headers, HTTPHeaderDict): + self._copy_from(headers) + else: + self.extend(headers) + if kwargs: + self.extend(kwargs) + + def __setitem__(self, key: str, val: str) -> None: + # avoid a bytes/str comparison by decoding before httplib + if isinstance(key, bytes): + key = key.decode("latin-1") + self._container[key.lower()] = [key, val] + + def __getitem__(self, key: str) -> str: + val = self._container[key.lower()] + return ", ".join(val[1:]) + + def __delitem__(self, key: str) -> None: + del self._container[key.lower()] + + def __contains__(self, key: object) -> bool: + if isinstance(key, str): + return key.lower() in self._container + return False + + def setdefault(self, key: str, default: str = "") -> str: + return super().setdefault(key, default) + + def __eq__(self, other: object) -> bool: + maybe_constructable = ensure_can_construct_http_header_dict(other) + if maybe_constructable is None: + return False + else: + other_as_http_header_dict = type(self)(maybe_constructable) + + return {k.lower(): v for k, v in self.itermerged()} == { + k.lower(): v for k, v in other_as_http_header_dict.itermerged() + } + + def __ne__(self, other: object) -> bool: + return not self.__eq__(other) + + def __len__(self) -> int: + return len(self._container) + + def __iter__(self) -> typing.Iterator[str]: + # Only provide the originally cased names + for vals in self._container.values(): + yield vals[0] + + def discard(self, key: str) -> None: + try: + del self[key] + except KeyError: + pass + + def add(self, key: str, val: str, *, combine: bool = False) -> None: + """Adds a (name, value) pair, doesn't overwrite the value if it already + exists. + + If this is called with combine=True, instead of adding a new header value + as a distinct item during iteration, this will instead append the value to + any existing header value with a comma. If no existing header value exists + for the key, then the value will simply be added, ignoring the combine parameter. + + >>> headers = HTTPHeaderDict(foo='bar') + >>> headers.add('Foo', 'baz') + >>> headers['foo'] + 'bar, baz' + >>> list(headers.items()) + [('foo', 'bar'), ('foo', 'baz')] + >>> headers.add('foo', 'quz', combine=True) + >>> list(headers.items()) + [('foo', 'bar, baz, quz')] + """ + # avoid a bytes/str comparison by decoding before httplib + if isinstance(key, bytes): + key = key.decode("latin-1") + key_lower = key.lower() + new_vals = [key, val] + # Keep the common case aka no item present as fast as possible + vals = self._container.setdefault(key_lower, new_vals) + if new_vals is not vals: + # if there are values here, then there is at least the initial + # key/value pair + assert len(vals) >= 2 + if combine: + vals[-1] = vals[-1] + ", " + val + else: + vals.append(val) + + def extend(self, *args: ValidHTTPHeaderSource, **kwargs: str) -> None: + """Generic import function for any type of header-like object. + Adapted version of MutableMapping.update in order to insert items + with self.add instead of self.__setitem__ + """ + if len(args) > 1: + raise TypeError( + f"extend() takes at most 1 positional arguments ({len(args)} given)" + ) + other = args[0] if len(args) >= 1 else () + + if isinstance(other, HTTPHeaderDict): + for key, val in other.iteritems(): + self.add(key, val) + elif isinstance(other, typing.Mapping): + for key, val in other.items(): + self.add(key, val) + elif isinstance(other, typing.Iterable): + other = typing.cast(typing.Iterable[tuple[str, str]], other) + for key, value in other: + self.add(key, value) + elif hasattr(other, "keys") and hasattr(other, "__getitem__"): + # THIS IS NOT A TYPESAFE BRANCH + # In this branch, the object has a `keys` attr but is not a Mapping or any of + # the other types indicated in the method signature. We do some stuff with + # it as though it partially implements the Mapping interface, but we're not + # doing that stuff safely AT ALL. + for key in other.keys(): + self.add(key, other[key]) + + for key, value in kwargs.items(): + self.add(key, value) + + @typing.overload + def getlist(self, key: str) -> list[str]: ... + + @typing.overload + def getlist(self, key: str, default: _DT) -> list[str] | _DT: ... + + def getlist( + self, key: str, default: _Sentinel | _DT = _Sentinel.not_passed + ) -> list[str] | _DT: + """Returns a list of all the values for the named field. Returns an + empty list if the key doesn't exist.""" + try: + vals = self._container[key.lower()] + except KeyError: + if default is _Sentinel.not_passed: + # _DT is unbound; empty list is instance of List[str] + return [] + # _DT is bound; default is instance of _DT + return default + else: + # _DT may or may not be bound; vals[1:] is instance of List[str], which + # meets our external interface requirement of `Union[List[str], _DT]`. + return vals[1:] + + def _prepare_for_method_change(self) -> Self: + """ + Remove content-specific header fields before changing the request + method to GET or HEAD according to RFC 9110, Section 15.4. + """ + content_specific_headers = [ + "Content-Encoding", + "Content-Language", + "Content-Location", + "Content-Type", + "Content-Length", + "Digest", + "Last-Modified", + ] + for header in content_specific_headers: + self.discard(header) + return self + + # Backwards compatibility for httplib + getheaders = getlist + getallmatchingheaders = getlist + iget = getlist + + # Backwards compatibility for http.cookiejar + get_all = getlist + + def __repr__(self) -> str: + return f"{type(self).__name__}({dict(self.itermerged())})" + + def _copy_from(self, other: HTTPHeaderDict) -> None: + for key in other: + val = other.getlist(key) + self._container[key.lower()] = [key, *val] + + def copy(self) -> Self: + clone = type(self)() + clone._copy_from(self) + return clone + + def iteritems(self) -> typing.Iterator[tuple[str, str]]: + """Iterate over all header lines, including duplicate ones.""" + for key in self: + vals = self._container[key.lower()] + for val in vals[1:]: + yield vals[0], val + + def itermerged(self) -> typing.Iterator[tuple[str, str]]: + """Iterate over all headers, merging duplicate ones together.""" + for key in self: + val = self._container[key.lower()] + yield val[0], ", ".join(val[1:]) + + def items(self) -> HTTPHeaderDictItemView: # type: ignore[override] + return HTTPHeaderDictItemView(self) + + def _has_value_for_header(self, header_name: str, potential_value: str) -> bool: + if header_name in self: + return potential_value in self._container[header_name.lower()][1:] + return False + + def __ior__(self, other: object) -> HTTPHeaderDict: + # Supports extending a header dict in-place using operator |= + # combining items with add instead of __setitem__ + maybe_constructable = ensure_can_construct_http_header_dict(other) + if maybe_constructable is None: + return NotImplemented + self.extend(maybe_constructable) + return self + + def __or__(self, other: object) -> Self: + # Supports merging header dicts using operator | + # combining items with add instead of __setitem__ + maybe_constructable = ensure_can_construct_http_header_dict(other) + if maybe_constructable is None: + return NotImplemented + result = self.copy() + result.extend(maybe_constructable) + return result + + def __ror__(self, other: object) -> Self: + # Supports merging header dicts using operator | when other is on left side + # combining items with add instead of __setitem__ + maybe_constructable = ensure_can_construct_http_header_dict(other) + if maybe_constructable is None: + return NotImplemented + result = type(self)(maybe_constructable) + result.extend(self) + return result diff --git a/.venv/lib/python3.11/site-packages/urllib3/_request_methods.py b/.venv/lib/python3.11/site-packages/urllib3/_request_methods.py new file mode 100644 index 0000000000000000000000000000000000000000..297c271bf401c1cb48c6225f8822e78f58c3ca56 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/_request_methods.py @@ -0,0 +1,278 @@ +from __future__ import annotations + +import json as _json +import typing +from urllib.parse import urlencode + +from ._base_connection import _TYPE_BODY +from ._collections import HTTPHeaderDict +from .filepost import _TYPE_FIELDS, encode_multipart_formdata +from .response import BaseHTTPResponse + +__all__ = ["RequestMethods"] + +_TYPE_ENCODE_URL_FIELDS = typing.Union[ + typing.Sequence[tuple[str, typing.Union[str, bytes]]], + typing.Mapping[str, typing.Union[str, bytes]], +] + + +class RequestMethods: + """ + Convenience mixin for classes who implement a :meth:`urlopen` method, such + as :class:`urllib3.HTTPConnectionPool` and + :class:`urllib3.PoolManager`. + + Provides behavior for making common types of HTTP request methods and + decides which type of request field encoding to use. + + Specifically, + + :meth:`.request_encode_url` is for sending requests whose fields are + encoded in the URL (such as GET, HEAD, DELETE). + + :meth:`.request_encode_body` is for sending requests whose fields are + encoded in the *body* of the request using multipart or www-form-urlencoded + (such as for POST, PUT, PATCH). + + :meth:`.request` is for making any kind of request, it will look up the + appropriate encoding format and use one of the above two methods to make + the request. + + Initializer parameters: + + :param headers: + Headers to include with all requests, unless other headers are given + explicitly. + """ + + _encode_url_methods = {"DELETE", "GET", "HEAD", "OPTIONS"} + + def __init__(self, headers: typing.Mapping[str, str] | None = None) -> None: + self.headers = headers or {} + + def urlopen( + self, + method: str, + url: str, + body: _TYPE_BODY | None = None, + headers: typing.Mapping[str, str] | None = None, + encode_multipart: bool = True, + multipart_boundary: str | None = None, + **kw: typing.Any, + ) -> BaseHTTPResponse: # Abstract + raise NotImplementedError( + "Classes extending RequestMethods must implement " + "their own ``urlopen`` method." + ) + + def request( + self, + method: str, + url: str, + body: _TYPE_BODY | None = None, + fields: _TYPE_FIELDS | None = None, + headers: typing.Mapping[str, str] | None = None, + json: typing.Any | None = None, + **urlopen_kw: typing.Any, + ) -> BaseHTTPResponse: + """ + Make a request using :meth:`urlopen` with the appropriate encoding of + ``fields`` based on the ``method`` used. + + This is a convenience method that requires the least amount of manual + effort. It can be used in most situations, while still having the + option to drop down to more specific methods when necessary, such as + :meth:`request_encode_url`, :meth:`request_encode_body`, + or even the lowest level :meth:`urlopen`. + + :param method: + HTTP request method (such as GET, POST, PUT, etc.) + + :param url: + The URL to perform the request on. + + :param body: + Data to send in the request body, either :class:`str`, :class:`bytes`, + an iterable of :class:`str`/:class:`bytes`, or a file-like object. + + :param fields: + Data to encode and send in the URL or request body, depending on ``method``. + + :param headers: + Dictionary of custom headers to send, such as User-Agent, + If-None-Match, etc. If None, pool headers are used. If provided, + these headers completely replace any pool-specific headers. + + :param json: + Data to encode and send as JSON with UTF-encoded in the request body. + The ``"Content-Type"`` header will be set to ``"application/json"`` + unless specified otherwise. + """ + method = method.upper() + + if json is not None and body is not None: + raise TypeError( + "request got values for both 'body' and 'json' parameters which are mutually exclusive" + ) + + if json is not None: + if headers is None: + headers = self.headers + + if not ("content-type" in map(str.lower, headers.keys())): + headers = HTTPHeaderDict(headers) + headers["Content-Type"] = "application/json" + + body = _json.dumps(json, separators=(",", ":"), ensure_ascii=False).encode( + "utf-8" + ) + + if body is not None: + urlopen_kw["body"] = body + + if method in self._encode_url_methods: + return self.request_encode_url( + method, + url, + fields=fields, # type: ignore[arg-type] + headers=headers, + **urlopen_kw, + ) + else: + return self.request_encode_body( + method, url, fields=fields, headers=headers, **urlopen_kw + ) + + def request_encode_url( + self, + method: str, + url: str, + fields: _TYPE_ENCODE_URL_FIELDS | None = None, + headers: typing.Mapping[str, str] | None = None, + **urlopen_kw: str, + ) -> BaseHTTPResponse: + """ + Make a request using :meth:`urlopen` with the ``fields`` encoded in + the url. This is useful for request methods like GET, HEAD, DELETE, etc. + + :param method: + HTTP request method (such as GET, POST, PUT, etc.) + + :param url: + The URL to perform the request on. + + :param fields: + Data to encode and send in the URL. + + :param headers: + Dictionary of custom headers to send, such as User-Agent, + If-None-Match, etc. If None, pool headers are used. If provided, + these headers completely replace any pool-specific headers. + """ + if headers is None: + headers = self.headers + + extra_kw: dict[str, typing.Any] = {"headers": headers} + extra_kw.update(urlopen_kw) + + if fields: + url += "?" + urlencode(fields) + + return self.urlopen(method, url, **extra_kw) + + def request_encode_body( + self, + method: str, + url: str, + fields: _TYPE_FIELDS | None = None, + headers: typing.Mapping[str, str] | None = None, + encode_multipart: bool = True, + multipart_boundary: str | None = None, + **urlopen_kw: str, + ) -> BaseHTTPResponse: + """ + Make a request using :meth:`urlopen` with the ``fields`` encoded in + the body. This is useful for request methods like POST, PUT, PATCH, etc. + + When ``encode_multipart=True`` (default), then + :func:`urllib3.encode_multipart_formdata` is used to encode + the payload with the appropriate content type. Otherwise + :func:`urllib.parse.urlencode` is used with the + 'application/x-www-form-urlencoded' content type. + + Multipart encoding must be used when posting files, and it's reasonably + safe to use it in other times too. However, it may break request + signing, such as with OAuth. + + Supports an optional ``fields`` parameter of key/value strings AND + key/filetuple. A filetuple is a (filename, data, MIME type) tuple where + the MIME type is optional. For example:: + + fields = { + 'foo': 'bar', + 'fakefile': ('foofile.txt', 'contents of foofile'), + 'realfile': ('barfile.txt', open('realfile').read()), + 'typedfile': ('bazfile.bin', open('bazfile').read(), + 'image/jpeg'), + 'nonamefile': 'contents of nonamefile field', + } + + When uploading a file, providing a filename (the first parameter of the + tuple) is optional but recommended to best mimic behavior of browsers. + + Note that if ``headers`` are supplied, the 'Content-Type' header will + be overwritten because it depends on the dynamic random boundary string + which is used to compose the body of the request. The random boundary + string can be explicitly set with the ``multipart_boundary`` parameter. + + :param method: + HTTP request method (such as GET, POST, PUT, etc.) + + :param url: + The URL to perform the request on. + + :param fields: + Data to encode and send in the request body. + + :param headers: + Dictionary of custom headers to send, such as User-Agent, + If-None-Match, etc. If None, pool headers are used. If provided, + these headers completely replace any pool-specific headers. + + :param encode_multipart: + If True, encode the ``fields`` using the multipart/form-data MIME + format. + + :param multipart_boundary: + If not specified, then a random boundary will be generated using + :func:`urllib3.filepost.choose_boundary`. + """ + if headers is None: + headers = self.headers + + extra_kw: dict[str, typing.Any] = {"headers": HTTPHeaderDict(headers)} + body: bytes | str + + if fields: + if "body" in urlopen_kw: + raise TypeError( + "request got values for both 'fields' and 'body', can only specify one." + ) + + if encode_multipart: + body, content_type = encode_multipart_formdata( + fields, boundary=multipart_boundary + ) + else: + body, content_type = ( + urlencode(fields), # type: ignore[arg-type] + "application/x-www-form-urlencoded", + ) + + extra_kw["body"] = body + extra_kw["headers"].setdefault("Content-Type", content_type) + + extra_kw.update(urlopen_kw) + + return self.urlopen(method, url, **extra_kw) diff --git a/.venv/lib/python3.11/site-packages/urllib3/connection.py b/.venv/lib/python3.11/site-packages/urllib3/connection.py new file mode 100644 index 0000000000000000000000000000000000000000..591ac407bcf655610ca59e4aab4f87a9e32022bf --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/connection.py @@ -0,0 +1,1044 @@ +from __future__ import annotations + +import datetime +import http.client +import logging +import os +import re +import socket +import sys +import threading +import typing +import warnings +from http.client import HTTPConnection as _HTTPConnection +from http.client import HTTPException as HTTPException # noqa: F401 +from http.client import ResponseNotReady +from socket import timeout as SocketTimeout + +if typing.TYPE_CHECKING: + from .response import HTTPResponse + from .util.ssl_ import _TYPE_PEER_CERT_RET_DICT + from .util.ssltransport import SSLTransport + +from ._collections import HTTPHeaderDict +from .http2 import probe as http2_probe +from .util.response import assert_header_parsing +from .util.timeout import _DEFAULT_TIMEOUT, _TYPE_TIMEOUT, Timeout +from .util.util import to_str +from .util.wait import wait_for_read + +try: # Compiled with SSL? + import ssl + + BaseSSLError = ssl.SSLError +except (ImportError, AttributeError): + ssl = None # type: ignore[assignment] + + class BaseSSLError(BaseException): # type: ignore[no-redef] + pass + + +from ._base_connection import _TYPE_BODY +from ._base_connection import ProxyConfig as ProxyConfig +from ._base_connection import _ResponseOptions as _ResponseOptions +from ._version import __version__ +from .exceptions import ( + ConnectTimeoutError, + HeaderParsingError, + NameResolutionError, + NewConnectionError, + ProxyError, + SystemTimeWarning, +) +from .util import SKIP_HEADER, SKIPPABLE_HEADERS, connection, ssl_ +from .util.request import body_to_chunks +from .util.ssl_ import assert_fingerprint as _assert_fingerprint +from .util.ssl_ import ( + create_urllib3_context, + is_ipaddress, + resolve_cert_reqs, + resolve_ssl_version, + ssl_wrap_socket, +) +from .util.ssl_match_hostname import CertificateError, match_hostname +from .util.url import Url + +# Not a no-op, we're adding this to the namespace so it can be imported. +ConnectionError = ConnectionError +BrokenPipeError = BrokenPipeError + + +log = logging.getLogger(__name__) + +port_by_scheme = {"http": 80, "https": 443} + +# When it comes time to update this value as a part of regular maintenance +# (ie test_recent_date is failing) update it to ~6 months before the current date. +RECENT_DATE = datetime.date(2023, 6, 1) + +_CONTAINS_CONTROL_CHAR_RE = re.compile(r"[^-!#$%&'*+.^_`|~0-9a-zA-Z]") + + +class HTTPConnection(_HTTPConnection): + """ + Based on :class:`http.client.HTTPConnection` but provides an extra constructor + backwards-compatibility layer between older and newer Pythons. + + Additional keyword parameters are used to configure attributes of the connection. + Accepted parameters include: + + - ``source_address``: Set the source address for the current connection. + - ``socket_options``: Set specific options on the underlying socket. If not specified, then + defaults are loaded from ``HTTPConnection.default_socket_options`` which includes disabling + Nagle's algorithm (sets TCP_NODELAY to 1) unless the connection is behind a proxy. + + For example, if you wish to enable TCP Keep Alive in addition to the defaults, + you might pass: + + .. code-block:: python + + HTTPConnection.default_socket_options + [ + (socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1), + ] + + Or you may want to disable the defaults by passing an empty list (e.g., ``[]``). + """ + + default_port: typing.ClassVar[int] = port_by_scheme["http"] # type: ignore[misc] + + #: Disable Nagle's algorithm by default. + #: ``[(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)]`` + default_socket_options: typing.ClassVar[connection._TYPE_SOCKET_OPTIONS] = [ + (socket.IPPROTO_TCP, socket.TCP_NODELAY, 1) + ] + + #: Whether this connection verifies the host's certificate. + is_verified: bool = False + + #: Whether this proxy connection verified the proxy host's certificate. + # If no proxy is currently connected to the value will be ``None``. + proxy_is_verified: bool | None = None + + blocksize: int + source_address: tuple[str, int] | None + socket_options: connection._TYPE_SOCKET_OPTIONS | None + + _has_connected_to_proxy: bool + _response_options: _ResponseOptions | None + _tunnel_host: str | None + _tunnel_port: int | None + _tunnel_scheme: str | None + + def __init__( + self, + host: str, + port: int | None = None, + *, + timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT, + source_address: tuple[str, int] | None = None, + blocksize: int = 16384, + socket_options: None | ( + connection._TYPE_SOCKET_OPTIONS + ) = default_socket_options, + proxy: Url | None = None, + proxy_config: ProxyConfig | None = None, + ) -> None: + super().__init__( + host=host, + port=port, + timeout=Timeout.resolve_default_timeout(timeout), + source_address=source_address, + blocksize=blocksize, + ) + self.socket_options = socket_options + self.proxy = proxy + self.proxy_config = proxy_config + + self._has_connected_to_proxy = False + self._response_options = None + self._tunnel_host: str | None = None + self._tunnel_port: int | None = None + self._tunnel_scheme: str | None = None + + @property + def host(self) -> str: + """ + Getter method to remove any trailing dots that indicate the hostname is an FQDN. + + In general, SSL certificates don't include the trailing dot indicating a + fully-qualified domain name, and thus, they don't validate properly when + checked against a domain name that includes the dot. In addition, some + servers may not expect to receive the trailing dot when provided. + + However, the hostname with trailing dot is critical to DNS resolution; doing a + lookup with the trailing dot will properly only resolve the appropriate FQDN, + whereas a lookup without a trailing dot will search the system's search domain + list. Thus, it's important to keep the original host around for use only in + those cases where it's appropriate (i.e., when doing DNS lookup to establish the + actual TCP connection across which we're going to send HTTP requests). + """ + return self._dns_host.rstrip(".") + + @host.setter + def host(self, value: str) -> None: + """ + Setter for the `host` property. + + We assume that only urllib3 uses the _dns_host attribute; httplib itself + only uses `host`, and it seems reasonable that other libraries follow suit. + """ + self._dns_host = value + + def _new_conn(self) -> socket.socket: + """Establish a socket connection and set nodelay settings on it. + + :return: New socket connection. + """ + try: + sock = connection.create_connection( + (self._dns_host, self.port), + self.timeout, + source_address=self.source_address, + socket_options=self.socket_options, + ) + except socket.gaierror as e: + raise NameResolutionError(self.host, self, e) from e + except SocketTimeout as e: + raise ConnectTimeoutError( + self, + f"Connection to {self.host} timed out. (connect timeout={self.timeout})", + ) from e + + except OSError as e: + raise NewConnectionError( + self, f"Failed to establish a new connection: {e}" + ) from e + + sys.audit("http.client.connect", self, self.host, self.port) + + return sock + + def set_tunnel( + self, + host: str, + port: int | None = None, + headers: typing.Mapping[str, str] | None = None, + scheme: str = "http", + ) -> None: + if scheme not in ("http", "https"): + raise ValueError( + f"Invalid proxy scheme for tunneling: {scheme!r}, must be either 'http' or 'https'" + ) + super().set_tunnel(host, port=port, headers=headers) + self._tunnel_scheme = scheme + + if sys.version_info < (3, 11, 4): + + def _tunnel(self) -> None: + _MAXLINE = http.client._MAXLINE # type: ignore[attr-defined] + connect = b"CONNECT %s:%d HTTP/1.0\r\n" % ( # type: ignore[str-format] + self._tunnel_host.encode("ascii"), # type: ignore[union-attr] + self._tunnel_port, + ) + headers = [connect] + for header, value in self._tunnel_headers.items(): # type: ignore[attr-defined] + headers.append(f"{header}: {value}\r\n".encode("latin-1")) + headers.append(b"\r\n") + # Making a single send() call instead of one per line encourages + # the host OS to use a more optimal packet size instead of + # potentially emitting a series of small packets. + self.send(b"".join(headers)) + del headers + + response = self.response_class(self.sock, method=self._method) # type: ignore[attr-defined] + try: + (version, code, message) = response._read_status() # type: ignore[attr-defined] + + if code != http.HTTPStatus.OK: + self.close() + raise OSError(f"Tunnel connection failed: {code} {message.strip()}") + while True: + line = response.fp.readline(_MAXLINE + 1) + if len(line) > _MAXLINE: + raise http.client.LineTooLong("header line") + if not line: + # for sites which EOF without sending a trailer + break + if line in (b"\r\n", b"\n", b""): + break + + if self.debuglevel > 0: + print("header:", line.decode()) + finally: + response.close() + + def connect(self) -> None: + self.sock = self._new_conn() + if self._tunnel_host: + # If we're tunneling it means we're connected to our proxy. + self._has_connected_to_proxy = True + + # TODO: Fix tunnel so it doesn't depend on self.sock state. + self._tunnel() + + # If there's a proxy to be connected to we are fully connected. + # This is set twice (once above and here) due to forwarding proxies + # not using tunnelling. + self._has_connected_to_proxy = bool(self.proxy) + + if self._has_connected_to_proxy: + self.proxy_is_verified = False + + @property + def is_closed(self) -> bool: + return self.sock is None + + @property + def is_connected(self) -> bool: + if self.sock is None: + return False + return not wait_for_read(self.sock, timeout=0.0) + + @property + def has_connected_to_proxy(self) -> bool: + return self._has_connected_to_proxy + + @property + def proxy_is_forwarding(self) -> bool: + """ + Return True if a forwarding proxy is configured, else return False + """ + return bool(self.proxy) and self._tunnel_host is None + + @property + def proxy_is_tunneling(self) -> bool: + """ + Return True if a tunneling proxy is configured, else return False + """ + return self._tunnel_host is not None + + def close(self) -> None: + try: + super().close() + finally: + # Reset all stateful properties so connection + # can be re-used without leaking prior configs. + self.sock = None + self.is_verified = False + self.proxy_is_verified = None + self._has_connected_to_proxy = False + self._response_options = None + self._tunnel_host = None + self._tunnel_port = None + self._tunnel_scheme = None + + def putrequest( + self, + method: str, + url: str, + skip_host: bool = False, + skip_accept_encoding: bool = False, + ) -> None: + """""" + # Empty docstring because the indentation of CPython's implementation + # is broken but we don't want this method in our documentation. + match = _CONTAINS_CONTROL_CHAR_RE.search(method) + if match: + raise ValueError( + f"Method cannot contain non-token characters {method!r} (found at least {match.group()!r})" + ) + + return super().putrequest( + method, url, skip_host=skip_host, skip_accept_encoding=skip_accept_encoding + ) + + def putheader(self, header: str, *values: str) -> None: # type: ignore[override] + """""" + if not any(isinstance(v, str) and v == SKIP_HEADER for v in values): + super().putheader(header, *values) + elif to_str(header.lower()) not in SKIPPABLE_HEADERS: + skippable_headers = "', '".join( + [str.title(header) for header in sorted(SKIPPABLE_HEADERS)] + ) + raise ValueError( + f"urllib3.util.SKIP_HEADER only supports '{skippable_headers}'" + ) + + # `request` method's signature intentionally violates LSP. + # urllib3's API is different from `http.client.HTTPConnection` and the subclassing is only incidental. + def request( # type: ignore[override] + self, + method: str, + url: str, + body: _TYPE_BODY | None = None, + headers: typing.Mapping[str, str] | None = None, + *, + chunked: bool = False, + preload_content: bool = True, + decode_content: bool = True, + enforce_content_length: bool = True, + ) -> None: + # Update the inner socket's timeout value to send the request. + # This only triggers if the connection is re-used. + if self.sock is not None: + self.sock.settimeout(self.timeout) + + # Store these values to be fed into the HTTPResponse + # object later. TODO: Remove this in favor of a real + # HTTP lifecycle mechanism. + + # We have to store these before we call .request() + # because sometimes we can still salvage a response + # off the wire even if we aren't able to completely + # send the request body. + self._response_options = _ResponseOptions( + request_method=method, + request_url=url, + preload_content=preload_content, + decode_content=decode_content, + enforce_content_length=enforce_content_length, + ) + + if headers is None: + headers = {} + header_keys = frozenset(to_str(k.lower()) for k in headers) + skip_accept_encoding = "accept-encoding" in header_keys + skip_host = "host" in header_keys + self.putrequest( + method, url, skip_accept_encoding=skip_accept_encoding, skip_host=skip_host + ) + + # Transform the body into an iterable of sendall()-able chunks + # and detect if an explicit Content-Length is doable. + chunks_and_cl = body_to_chunks(body, method=method, blocksize=self.blocksize) + chunks = chunks_and_cl.chunks + content_length = chunks_and_cl.content_length + + # When chunked is explicit set to 'True' we respect that. + if chunked: + if "transfer-encoding" not in header_keys: + self.putheader("Transfer-Encoding", "chunked") + else: + # Detect whether a framing mechanism is already in use. If so + # we respect that value, otherwise we pick chunked vs content-length + # depending on the type of 'body'. + if "content-length" in header_keys: + chunked = False + elif "transfer-encoding" in header_keys: + chunked = True + + # Otherwise we go off the recommendation of 'body_to_chunks()'. + else: + chunked = False + if content_length is None: + if chunks is not None: + chunked = True + self.putheader("Transfer-Encoding", "chunked") + else: + self.putheader("Content-Length", str(content_length)) + + # Now that framing headers are out of the way we send all the other headers. + if "user-agent" not in header_keys: + self.putheader("User-Agent", _get_default_user_agent()) + for header, value in headers.items(): + self.putheader(header, value) + self.endheaders() + + # If we're given a body we start sending that in chunks. + if chunks is not None: + for chunk in chunks: + # Sending empty chunks isn't allowed for TE: chunked + # as it indicates the end of the body. + if not chunk: + continue + if isinstance(chunk, str): + chunk = chunk.encode("utf-8") + if chunked: + self.send(b"%x\r\n%b\r\n" % (len(chunk), chunk)) + else: + self.send(chunk) + + # Regardless of whether we have a body or not, if we're in + # chunked mode we want to send an explicit empty chunk. + if chunked: + self.send(b"0\r\n\r\n") + + def request_chunked( + self, + method: str, + url: str, + body: _TYPE_BODY | None = None, + headers: typing.Mapping[str, str] | None = None, + ) -> None: + """ + Alternative to the common request method, which sends the + body with chunked encoding and not as one block + """ + warnings.warn( + "HTTPConnection.request_chunked() is deprecated and will be removed " + "in urllib3 v2.1.0. Instead use HTTPConnection.request(..., chunked=True).", + category=DeprecationWarning, + stacklevel=2, + ) + self.request(method, url, body=body, headers=headers, chunked=True) + + def getresponse( # type: ignore[override] + self, + ) -> HTTPResponse: + """ + Get the response from the server. + + If the HTTPConnection is in the correct state, returns an instance of HTTPResponse or of whatever object is returned by the response_class variable. + + If a request has not been sent or if a previous response has not be handled, ResponseNotReady is raised. If the HTTP response indicates that the connection should be closed, then it will be closed before the response is returned. When the connection is closed, the underlying socket is closed. + """ + # Raise the same error as http.client.HTTPConnection + if self._response_options is None: + raise ResponseNotReady() + + # Reset this attribute for being used again. + resp_options = self._response_options + self._response_options = None + + # Since the connection's timeout value may have been updated + # we need to set the timeout on the socket. + self.sock.settimeout(self.timeout) + + # This is needed here to avoid circular import errors + from .response import HTTPResponse + + # Save a reference to the shutdown function before ownership is passed + # to httplib_response + # TODO should we implement it everywhere? + _shutdown = getattr(self.sock, "shutdown", None) + + # Get the response from http.client.HTTPConnection + httplib_response = super().getresponse() + + try: + assert_header_parsing(httplib_response.msg) + except (HeaderParsingError, TypeError) as hpe: + log.warning( + "Failed to parse headers (url=%s): %s", + _url_from_connection(self, resp_options.request_url), + hpe, + exc_info=True, + ) + + headers = HTTPHeaderDict(httplib_response.msg.items()) + + response = HTTPResponse( + body=httplib_response, + headers=headers, + status=httplib_response.status, + version=httplib_response.version, + version_string=getattr(self, "_http_vsn_str", "HTTP/?"), + reason=httplib_response.reason, + preload_content=resp_options.preload_content, + decode_content=resp_options.decode_content, + original_response=httplib_response, + enforce_content_length=resp_options.enforce_content_length, + request_method=resp_options.request_method, + request_url=resp_options.request_url, + sock_shutdown=_shutdown, + ) + return response + + +class HTTPSConnection(HTTPConnection): + """ + Many of the parameters to this constructor are passed to the underlying SSL + socket by means of :py:func:`urllib3.util.ssl_wrap_socket`. + """ + + default_port = port_by_scheme["https"] # type: ignore[misc] + + cert_reqs: int | str | None = None + ca_certs: str | None = None + ca_cert_dir: str | None = None + ca_cert_data: None | str | bytes = None + ssl_version: int | str | None = None + ssl_minimum_version: int | None = None + ssl_maximum_version: int | None = None + assert_fingerprint: str | None = None + _connect_callback: typing.Callable[..., None] | None = None + + def __init__( + self, + host: str, + port: int | None = None, + *, + timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT, + source_address: tuple[str, int] | None = None, + blocksize: int = 16384, + socket_options: None | ( + connection._TYPE_SOCKET_OPTIONS + ) = HTTPConnection.default_socket_options, + proxy: Url | None = None, + proxy_config: ProxyConfig | None = None, + cert_reqs: int | str | None = None, + assert_hostname: None | str | typing.Literal[False] = None, + assert_fingerprint: str | None = None, + server_hostname: str | None = None, + ssl_context: ssl.SSLContext | None = None, + ca_certs: str | None = None, + ca_cert_dir: str | None = None, + ca_cert_data: None | str | bytes = None, + ssl_minimum_version: int | None = None, + ssl_maximum_version: int | None = None, + ssl_version: int | str | None = None, # Deprecated + cert_file: str | None = None, + key_file: str | None = None, + key_password: str | None = None, + ) -> None: + super().__init__( + host, + port=port, + timeout=timeout, + source_address=source_address, + blocksize=blocksize, + socket_options=socket_options, + proxy=proxy, + proxy_config=proxy_config, + ) + + self.key_file = key_file + self.cert_file = cert_file + self.key_password = key_password + self.ssl_context = ssl_context + self.server_hostname = server_hostname + self.assert_hostname = assert_hostname + self.assert_fingerprint = assert_fingerprint + self.ssl_version = ssl_version + self.ssl_minimum_version = ssl_minimum_version + self.ssl_maximum_version = ssl_maximum_version + self.ca_certs = ca_certs and os.path.expanduser(ca_certs) + self.ca_cert_dir = ca_cert_dir and os.path.expanduser(ca_cert_dir) + self.ca_cert_data = ca_cert_data + + # cert_reqs depends on ssl_context so calculate last. + if cert_reqs is None: + if self.ssl_context is not None: + cert_reqs = self.ssl_context.verify_mode + else: + cert_reqs = resolve_cert_reqs(None) + self.cert_reqs = cert_reqs + self._connect_callback = None + + def set_cert( + self, + key_file: str | None = None, + cert_file: str | None = None, + cert_reqs: int | str | None = None, + key_password: str | None = None, + ca_certs: str | None = None, + assert_hostname: None | str | typing.Literal[False] = None, + assert_fingerprint: str | None = None, + ca_cert_dir: str | None = None, + ca_cert_data: None | str | bytes = None, + ) -> None: + """ + This method should only be called once, before the connection is used. + """ + warnings.warn( + "HTTPSConnection.set_cert() is deprecated and will be removed " + "in urllib3 v2.1.0. Instead provide the parameters to the " + "HTTPSConnection constructor.", + category=DeprecationWarning, + stacklevel=2, + ) + + # If cert_reqs is not provided we'll assume CERT_REQUIRED unless we also + # have an SSLContext object in which case we'll use its verify_mode. + if cert_reqs is None: + if self.ssl_context is not None: + cert_reqs = self.ssl_context.verify_mode + else: + cert_reqs = resolve_cert_reqs(None) + + self.key_file = key_file + self.cert_file = cert_file + self.cert_reqs = cert_reqs + self.key_password = key_password + self.assert_hostname = assert_hostname + self.assert_fingerprint = assert_fingerprint + self.ca_certs = ca_certs and os.path.expanduser(ca_certs) + self.ca_cert_dir = ca_cert_dir and os.path.expanduser(ca_cert_dir) + self.ca_cert_data = ca_cert_data + + def connect(self) -> None: + # Today we don't need to be doing this step before the /actual/ socket + # connection, however in the future we'll need to decide whether to + # create a new socket or re-use an existing "shared" socket as a part + # of the HTTP/2 handshake dance. + if self._tunnel_host is not None and self._tunnel_port is not None: + probe_http2_host = self._tunnel_host + probe_http2_port = self._tunnel_port + else: + probe_http2_host = self.host + probe_http2_port = self.port + + # Check if the target origin supports HTTP/2. + # If the value comes back as 'None' it means that the current thread + # is probing for HTTP/2 support. Otherwise, we're waiting for another + # probe to complete, or we get a value right away. + target_supports_http2: bool | None + if "h2" in ssl_.ALPN_PROTOCOLS: + target_supports_http2 = http2_probe.acquire_and_get( + host=probe_http2_host, port=probe_http2_port + ) + else: + # If HTTP/2 isn't going to be offered it doesn't matter if + # the target supports HTTP/2. Don't want to make a probe. + target_supports_http2 = False + + if self._connect_callback is not None: + self._connect_callback( + "before connect", + thread_id=threading.get_ident(), + target_supports_http2=target_supports_http2, + ) + + try: + sock: socket.socket | ssl.SSLSocket + self.sock = sock = self._new_conn() + server_hostname: str = self.host + tls_in_tls = False + + # Do we need to establish a tunnel? + if self.proxy_is_tunneling: + # We're tunneling to an HTTPS origin so need to do TLS-in-TLS. + if self._tunnel_scheme == "https": + # _connect_tls_proxy will verify and assign proxy_is_verified + self.sock = sock = self._connect_tls_proxy(self.host, sock) + tls_in_tls = True + elif self._tunnel_scheme == "http": + self.proxy_is_verified = False + + # If we're tunneling it means we're connected to our proxy. + self._has_connected_to_proxy = True + + self._tunnel() + # Override the host with the one we're requesting data from. + server_hostname = typing.cast(str, self._tunnel_host) + + if self.server_hostname is not None: + server_hostname = self.server_hostname + + is_time_off = datetime.date.today() < RECENT_DATE + if is_time_off: + warnings.warn( + ( + f"System time is way off (before {RECENT_DATE}). This will probably " + "lead to SSL verification errors" + ), + SystemTimeWarning, + ) + + # Remove trailing '.' from fqdn hostnames to allow certificate validation + server_hostname_rm_dot = server_hostname.rstrip(".") + + sock_and_verified = _ssl_wrap_socket_and_match_hostname( + sock=sock, + cert_reqs=self.cert_reqs, + ssl_version=self.ssl_version, + ssl_minimum_version=self.ssl_minimum_version, + ssl_maximum_version=self.ssl_maximum_version, + ca_certs=self.ca_certs, + ca_cert_dir=self.ca_cert_dir, + ca_cert_data=self.ca_cert_data, + cert_file=self.cert_file, + key_file=self.key_file, + key_password=self.key_password, + server_hostname=server_hostname_rm_dot, + ssl_context=self.ssl_context, + tls_in_tls=tls_in_tls, + assert_hostname=self.assert_hostname, + assert_fingerprint=self.assert_fingerprint, + ) + self.sock = sock_and_verified.socket + + # If an error occurs during connection/handshake we may need to release + # our lock so another connection can probe the origin. + except BaseException: + if self._connect_callback is not None: + self._connect_callback( + "after connect failure", + thread_id=threading.get_ident(), + target_supports_http2=target_supports_http2, + ) + + if target_supports_http2 is None: + http2_probe.set_and_release( + host=probe_http2_host, port=probe_http2_port, supports_http2=None + ) + raise + + # If this connection doesn't know if the origin supports HTTP/2 + # we report back to the HTTP/2 probe our result. + if target_supports_http2 is None: + supports_http2 = sock_and_verified.socket.selected_alpn_protocol() == "h2" + http2_probe.set_and_release( + host=probe_http2_host, + port=probe_http2_port, + supports_http2=supports_http2, + ) + + # Forwarding proxies can never have a verified target since + # the proxy is the one doing the verification. Should instead + # use a CONNECT tunnel in order to verify the target. + # See: https://github.com/urllib3/urllib3/issues/3267. + if self.proxy_is_forwarding: + self.is_verified = False + else: + self.is_verified = sock_and_verified.is_verified + + # If there's a proxy to be connected to we are fully connected. + # This is set twice (once above and here) due to forwarding proxies + # not using tunnelling. + self._has_connected_to_proxy = bool(self.proxy) + + # Set `self.proxy_is_verified` unless it's already set while + # establishing a tunnel. + if self._has_connected_to_proxy and self.proxy_is_verified is None: + self.proxy_is_verified = sock_and_verified.is_verified + + def _connect_tls_proxy(self, hostname: str, sock: socket.socket) -> ssl.SSLSocket: + """ + Establish a TLS connection to the proxy using the provided SSL context. + """ + # `_connect_tls_proxy` is called when self._tunnel_host is truthy. + proxy_config = typing.cast(ProxyConfig, self.proxy_config) + ssl_context = proxy_config.ssl_context + sock_and_verified = _ssl_wrap_socket_and_match_hostname( + sock, + cert_reqs=self.cert_reqs, + ssl_version=self.ssl_version, + ssl_minimum_version=self.ssl_minimum_version, + ssl_maximum_version=self.ssl_maximum_version, + ca_certs=self.ca_certs, + ca_cert_dir=self.ca_cert_dir, + ca_cert_data=self.ca_cert_data, + server_hostname=hostname, + ssl_context=ssl_context, + assert_hostname=proxy_config.assert_hostname, + assert_fingerprint=proxy_config.assert_fingerprint, + # Features that aren't implemented for proxies yet: + cert_file=None, + key_file=None, + key_password=None, + tls_in_tls=False, + ) + self.proxy_is_verified = sock_and_verified.is_verified + return sock_and_verified.socket # type: ignore[return-value] + + +class _WrappedAndVerifiedSocket(typing.NamedTuple): + """ + Wrapped socket and whether the connection is + verified after the TLS handshake + """ + + socket: ssl.SSLSocket | SSLTransport + is_verified: bool + + +def _ssl_wrap_socket_and_match_hostname( + sock: socket.socket, + *, + cert_reqs: None | str | int, + ssl_version: None | str | int, + ssl_minimum_version: int | None, + ssl_maximum_version: int | None, + cert_file: str | None, + key_file: str | None, + key_password: str | None, + ca_certs: str | None, + ca_cert_dir: str | None, + ca_cert_data: None | str | bytes, + assert_hostname: None | str | typing.Literal[False], + assert_fingerprint: str | None, + server_hostname: str | None, + ssl_context: ssl.SSLContext | None, + tls_in_tls: bool = False, +) -> _WrappedAndVerifiedSocket: + """Logic for constructing an SSLContext from all TLS parameters, passing + that down into ssl_wrap_socket, and then doing certificate verification + either via hostname or fingerprint. This function exists to guarantee + that both proxies and targets have the same behavior when connecting via TLS. + """ + default_ssl_context = False + if ssl_context is None: + default_ssl_context = True + context = create_urllib3_context( + ssl_version=resolve_ssl_version(ssl_version), + ssl_minimum_version=ssl_minimum_version, + ssl_maximum_version=ssl_maximum_version, + cert_reqs=resolve_cert_reqs(cert_reqs), + ) + else: + context = ssl_context + + context.verify_mode = resolve_cert_reqs(cert_reqs) + + # In some cases, we want to verify hostnames ourselves + if ( + # `ssl` can't verify fingerprints or alternate hostnames + assert_fingerprint + or assert_hostname + # assert_hostname can be set to False to disable hostname checking + or assert_hostname is False + # We still support OpenSSL 1.0.2, which prevents us from verifying + # hostnames easily: https://github.com/pyca/pyopenssl/pull/933 + or ssl_.IS_PYOPENSSL + or not ssl_.HAS_NEVER_CHECK_COMMON_NAME + ): + context.check_hostname = False + + # Try to load OS default certs if none are given. We need to do the hasattr() check + # for custom pyOpenSSL SSLContext objects because they don't support + # load_default_certs(). + if ( + not ca_certs + and not ca_cert_dir + and not ca_cert_data + and default_ssl_context + and hasattr(context, "load_default_certs") + ): + context.load_default_certs() + + # Ensure that IPv6 addresses are in the proper format and don't have a + # scope ID. Python's SSL module fails to recognize scoped IPv6 addresses + # and interprets them as DNS hostnames. + if server_hostname is not None: + normalized = server_hostname.strip("[]") + if "%" in normalized: + normalized = normalized[: normalized.rfind("%")] + if is_ipaddress(normalized): + server_hostname = normalized + + ssl_sock = ssl_wrap_socket( + sock=sock, + keyfile=key_file, + certfile=cert_file, + key_password=key_password, + ca_certs=ca_certs, + ca_cert_dir=ca_cert_dir, + ca_cert_data=ca_cert_data, + server_hostname=server_hostname, + ssl_context=context, + tls_in_tls=tls_in_tls, + ) + + try: + if assert_fingerprint: + _assert_fingerprint( + ssl_sock.getpeercert(binary_form=True), assert_fingerprint + ) + elif ( + context.verify_mode != ssl.CERT_NONE + and not context.check_hostname + and assert_hostname is not False + ): + cert: _TYPE_PEER_CERT_RET_DICT = ssl_sock.getpeercert() # type: ignore[assignment] + + # Need to signal to our match_hostname whether to use 'commonName' or not. + # If we're using our own constructed SSLContext we explicitly set 'False' + # because PyPy hard-codes 'True' from SSLContext.hostname_checks_common_name. + if default_ssl_context: + hostname_checks_common_name = False + else: + hostname_checks_common_name = ( + getattr(context, "hostname_checks_common_name", False) or False + ) + + _match_hostname( + cert, + assert_hostname or server_hostname, # type: ignore[arg-type] + hostname_checks_common_name, + ) + + return _WrappedAndVerifiedSocket( + socket=ssl_sock, + is_verified=context.verify_mode == ssl.CERT_REQUIRED + or bool(assert_fingerprint), + ) + except BaseException: + ssl_sock.close() + raise + + +def _match_hostname( + cert: _TYPE_PEER_CERT_RET_DICT | None, + asserted_hostname: str, + hostname_checks_common_name: bool = False, +) -> None: + # Our upstream implementation of ssl.match_hostname() + # only applies this normalization to IP addresses so it doesn't + # match DNS SANs so we do the same thing! + stripped_hostname = asserted_hostname.strip("[]") + if is_ipaddress(stripped_hostname): + asserted_hostname = stripped_hostname + + try: + match_hostname(cert, asserted_hostname, hostname_checks_common_name) + except CertificateError as e: + log.warning( + "Certificate did not match expected hostname: %s. Certificate: %s", + asserted_hostname, + cert, + ) + # Add cert to exception and reraise so client code can inspect + # the cert when catching the exception, if they want to + e._peer_cert = cert # type: ignore[attr-defined] + raise + + +def _wrap_proxy_error(err: Exception, proxy_scheme: str | None) -> ProxyError: + # Look for the phrase 'wrong version number', if found + # then we should warn the user that we're very sure that + # this proxy is HTTP-only and they have a configuration issue. + error_normalized = " ".join(re.split("[^a-z]", str(err).lower())) + is_likely_http_proxy = ( + "wrong version number" in error_normalized + or "unknown protocol" in error_normalized + or "record layer failure" in error_normalized + ) + http_proxy_warning = ( + ". Your proxy appears to only use HTTP and not HTTPS, " + "try changing your proxy URL to be HTTP. See: " + "https://urllib3.readthedocs.io/en/latest/advanced-usage.html" + "#https-proxy-error-http-proxy" + ) + new_err = ProxyError( + f"Unable to connect to proxy" + f"{http_proxy_warning if is_likely_http_proxy and proxy_scheme == 'https' else ''}", + err, + ) + new_err.__cause__ = err + return new_err + + +def _get_default_user_agent() -> str: + return f"python-urllib3/{__version__}" + + +class DummyConnection: + """Used to detect a failed ConnectionCls import.""" + + +if not ssl: + HTTPSConnection = DummyConnection # type: ignore[misc, assignment] # noqa: F811 + + +VerifiedHTTPSConnection = HTTPSConnection + + +def _url_from_connection( + conn: HTTPConnection | HTTPSConnection, path: str | None = None +) -> str: + """Returns the URL from a given connection. This is mainly used for testing and logging.""" + + scheme = "https" if isinstance(conn, HTTPSConnection) else "http" + + return Url(scheme=scheme, host=conn.host, port=conn.port, path=path).url diff --git a/.venv/lib/python3.11/site-packages/urllib3/connectionpool.py b/.venv/lib/python3.11/site-packages/urllib3/connectionpool.py new file mode 100644 index 0000000000000000000000000000000000000000..3a0685b4cdd0562e508b9dd032765b5c759ea61e --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/connectionpool.py @@ -0,0 +1,1178 @@ +from __future__ import annotations + +import errno +import logging +import queue +import sys +import typing +import warnings +import weakref +from socket import timeout as SocketTimeout +from types import TracebackType + +from ._base_connection import _TYPE_BODY +from ._collections import HTTPHeaderDict +from ._request_methods import RequestMethods +from .connection import ( + BaseSSLError, + BrokenPipeError, + DummyConnection, + HTTPConnection, + HTTPException, + HTTPSConnection, + ProxyConfig, + _wrap_proxy_error, +) +from .connection import port_by_scheme as port_by_scheme +from .exceptions import ( + ClosedPoolError, + EmptyPoolError, + FullPoolError, + HostChangedError, + InsecureRequestWarning, + LocationValueError, + MaxRetryError, + NewConnectionError, + ProtocolError, + ProxyError, + ReadTimeoutError, + SSLError, + TimeoutError, +) +from .response import BaseHTTPResponse +from .util.connection import is_connection_dropped +from .util.proxy import connection_requires_http_tunnel +from .util.request import _TYPE_BODY_POSITION, set_file_position +from .util.retry import Retry +from .util.ssl_match_hostname import CertificateError +from .util.timeout import _DEFAULT_TIMEOUT, _TYPE_DEFAULT, Timeout +from .util.url import Url, _encode_target +from .util.url import _normalize_host as normalize_host +from .util.url import parse_url +from .util.util import to_str + +if typing.TYPE_CHECKING: + import ssl + + from typing_extensions import Self + + from ._base_connection import BaseHTTPConnection, BaseHTTPSConnection + +log = logging.getLogger(__name__) + +_TYPE_TIMEOUT = typing.Union[Timeout, float, _TYPE_DEFAULT, None] + + +# Pool objects +class ConnectionPool: + """ + Base class for all connection pools, such as + :class:`.HTTPConnectionPool` and :class:`.HTTPSConnectionPool`. + + .. note:: + ConnectionPool.urlopen() does not normalize or percent-encode target URIs + which is useful if your target server doesn't support percent-encoded + target URIs. + """ + + scheme: str | None = None + QueueCls = queue.LifoQueue + + def __init__(self, host: str, port: int | None = None) -> None: + if not host: + raise LocationValueError("No host specified.") + + self.host = _normalize_host(host, scheme=self.scheme) + self.port = port + + # This property uses 'normalize_host()' (not '_normalize_host()') + # to avoid removing square braces around IPv6 addresses. + # This value is sent to `HTTPConnection.set_tunnel()` if called + # because square braces are required for HTTP CONNECT tunneling. + self._tunnel_host = normalize_host(host, scheme=self.scheme).lower() + + def __str__(self) -> str: + return f"{type(self).__name__}(host={self.host!r}, port={self.port!r})" + + def __enter__(self) -> Self: + return self + + def __exit__( + self, + exc_type: type[BaseException] | None, + exc_val: BaseException | None, + exc_tb: TracebackType | None, + ) -> typing.Literal[False]: + self.close() + # Return False to re-raise any potential exceptions + return False + + def close(self) -> None: + """ + Close all pooled connections and disable the pool. + """ + + +# This is taken from http://hg.python.org/cpython/file/7aaba721ebc0/Lib/socket.py#l252 +_blocking_errnos = {errno.EAGAIN, errno.EWOULDBLOCK} + + +class HTTPConnectionPool(ConnectionPool, RequestMethods): + """ + Thread-safe connection pool for one host. + + :param host: + Host used for this HTTP Connection (e.g. "localhost"), passed into + :class:`http.client.HTTPConnection`. + + :param port: + Port used for this HTTP Connection (None is equivalent to 80), passed + into :class:`http.client.HTTPConnection`. + + :param timeout: + Socket timeout in seconds for each individual connection. This can + be a float or integer, which sets the timeout for the HTTP request, + or an instance of :class:`urllib3.util.Timeout` which gives you more + fine-grained control over request timeouts. After the constructor has + been parsed, this is always a `urllib3.util.Timeout` object. + + :param maxsize: + Number of connections to save that can be reused. More than 1 is useful + in multithreaded situations. If ``block`` is set to False, more + connections will be created but they will not be saved once they've + been used. + + :param block: + If set to True, no more than ``maxsize`` connections will be used at + a time. When no free connections are available, the call will block + until a connection has been released. This is a useful side effect for + particular multithreaded situations where one does not want to use more + than maxsize connections per host to prevent flooding. + + :param headers: + Headers to include with all requests, unless other headers are given + explicitly. + + :param retries: + Retry configuration to use by default with requests in this pool. + + :param _proxy: + Parsed proxy URL, should not be used directly, instead, see + :class:`urllib3.ProxyManager` + + :param _proxy_headers: + A dictionary with proxy headers, should not be used directly, + instead, see :class:`urllib3.ProxyManager` + + :param \\**conn_kw: + Additional parameters are used to create fresh :class:`urllib3.connection.HTTPConnection`, + :class:`urllib3.connection.HTTPSConnection` instances. + """ + + scheme = "http" + ConnectionCls: type[BaseHTTPConnection] | type[BaseHTTPSConnection] = HTTPConnection + + def __init__( + self, + host: str, + port: int | None = None, + timeout: _TYPE_TIMEOUT | None = _DEFAULT_TIMEOUT, + maxsize: int = 1, + block: bool = False, + headers: typing.Mapping[str, str] | None = None, + retries: Retry | bool | int | None = None, + _proxy: Url | None = None, + _proxy_headers: typing.Mapping[str, str] | None = None, + _proxy_config: ProxyConfig | None = None, + **conn_kw: typing.Any, + ): + ConnectionPool.__init__(self, host, port) + RequestMethods.__init__(self, headers) + + if not isinstance(timeout, Timeout): + timeout = Timeout.from_float(timeout) + + if retries is None: + retries = Retry.DEFAULT + + self.timeout = timeout + self.retries = retries + + self.pool: queue.LifoQueue[typing.Any] | None = self.QueueCls(maxsize) + self.block = block + + self.proxy = _proxy + self.proxy_headers = _proxy_headers or {} + self.proxy_config = _proxy_config + + # Fill the queue up so that doing get() on it will block properly + for _ in range(maxsize): + self.pool.put(None) + + # These are mostly for testing and debugging purposes. + self.num_connections = 0 + self.num_requests = 0 + self.conn_kw = conn_kw + + if self.proxy: + # Enable Nagle's algorithm for proxies, to avoid packet fragmentation. + # We cannot know if the user has added default socket options, so we cannot replace the + # list. + self.conn_kw.setdefault("socket_options", []) + + self.conn_kw["proxy"] = self.proxy + self.conn_kw["proxy_config"] = self.proxy_config + + # Do not pass 'self' as callback to 'finalize'. + # Then the 'finalize' would keep an endless living (leak) to self. + # By just passing a reference to the pool allows the garbage collector + # to free self if nobody else has a reference to it. + pool = self.pool + + # Close all the HTTPConnections in the pool before the + # HTTPConnectionPool object is garbage collected. + weakref.finalize(self, _close_pool_connections, pool) + + def _new_conn(self) -> BaseHTTPConnection: + """ + Return a fresh :class:`HTTPConnection`. + """ + self.num_connections += 1 + log.debug( + "Starting new HTTP connection (%d): %s:%s", + self.num_connections, + self.host, + self.port or "80", + ) + + conn = self.ConnectionCls( + host=self.host, + port=self.port, + timeout=self.timeout.connect_timeout, + **self.conn_kw, + ) + return conn + + def _get_conn(self, timeout: float | None = None) -> BaseHTTPConnection: + """ + Get a connection. Will return a pooled connection if one is available. + + If no connections are available and :prop:`.block` is ``False``, then a + fresh connection is returned. + + :param timeout: + Seconds to wait before giving up and raising + :class:`urllib3.exceptions.EmptyPoolError` if the pool is empty and + :prop:`.block` is ``True``. + """ + conn = None + + if self.pool is None: + raise ClosedPoolError(self, "Pool is closed.") + + try: + conn = self.pool.get(block=self.block, timeout=timeout) + + except AttributeError: # self.pool is None + raise ClosedPoolError(self, "Pool is closed.") from None # Defensive: + + except queue.Empty: + if self.block: + raise EmptyPoolError( + self, + "Pool is empty and a new connection can't be opened due to blocking mode.", + ) from None + pass # Oh well, we'll create a new connection then + + # If this is a persistent connection, check if it got disconnected + if conn and is_connection_dropped(conn): + log.debug("Resetting dropped connection: %s", self.host) + conn.close() + + return conn or self._new_conn() + + def _put_conn(self, conn: BaseHTTPConnection | None) -> None: + """ + Put a connection back into the pool. + + :param conn: + Connection object for the current host and port as returned by + :meth:`._new_conn` or :meth:`._get_conn`. + + If the pool is already full, the connection is closed and discarded + because we exceeded maxsize. If connections are discarded frequently, + then maxsize should be increased. + + If the pool is closed, then the connection will be closed and discarded. + """ + if self.pool is not None: + try: + self.pool.put(conn, block=False) + return # Everything is dandy, done. + except AttributeError: + # self.pool is None. + pass + except queue.Full: + # Connection never got put back into the pool, close it. + if conn: + conn.close() + + if self.block: + # This should never happen if you got the conn from self._get_conn + raise FullPoolError( + self, + "Pool reached maximum size and no more connections are allowed.", + ) from None + + log.warning( + "Connection pool is full, discarding connection: %s. Connection pool size: %s", + self.host, + self.pool.qsize(), + ) + + # Connection never got put back into the pool, close it. + if conn: + conn.close() + + def _validate_conn(self, conn: BaseHTTPConnection) -> None: + """ + Called right before a request is made, after the socket is created. + """ + + def _prepare_proxy(self, conn: BaseHTTPConnection) -> None: + # Nothing to do for HTTP connections. + pass + + def _get_timeout(self, timeout: _TYPE_TIMEOUT) -> Timeout: + """Helper that always returns a :class:`urllib3.util.Timeout`""" + if timeout is _DEFAULT_TIMEOUT: + return self.timeout.clone() + + if isinstance(timeout, Timeout): + return timeout.clone() + else: + # User passed us an int/float. This is for backwards compatibility, + # can be removed later + return Timeout.from_float(timeout) + + def _raise_timeout( + self, + err: BaseSSLError | OSError | SocketTimeout, + url: str, + timeout_value: _TYPE_TIMEOUT | None, + ) -> None: + """Is the error actually a timeout? Will raise a ReadTimeout or pass""" + + if isinstance(err, SocketTimeout): + raise ReadTimeoutError( + self, url, f"Read timed out. (read timeout={timeout_value})" + ) from err + + # See the above comment about EAGAIN in Python 3. + if hasattr(err, "errno") and err.errno in _blocking_errnos: + raise ReadTimeoutError( + self, url, f"Read timed out. (read timeout={timeout_value})" + ) from err + + def _make_request( + self, + conn: BaseHTTPConnection, + method: str, + url: str, + body: _TYPE_BODY | None = None, + headers: typing.Mapping[str, str] | None = None, + retries: Retry | None = None, + timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT, + chunked: bool = False, + response_conn: BaseHTTPConnection | None = None, + preload_content: bool = True, + decode_content: bool = True, + enforce_content_length: bool = True, + ) -> BaseHTTPResponse: + """ + Perform a request on a given urllib connection object taken from our + pool. + + :param conn: + a connection from one of our connection pools + + :param method: + HTTP request method (such as GET, POST, PUT, etc.) + + :param url: + The URL to perform the request on. + + :param body: + Data to send in the request body, either :class:`str`, :class:`bytes`, + an iterable of :class:`str`/:class:`bytes`, or a file-like object. + + :param headers: + Dictionary of custom headers to send, such as User-Agent, + If-None-Match, etc. If None, pool headers are used. If provided, + these headers completely replace any pool-specific headers. + + :param retries: + Configure the number of retries to allow before raising a + :class:`~urllib3.exceptions.MaxRetryError` exception. + + Pass ``None`` to retry until you receive a response. Pass a + :class:`~urllib3.util.retry.Retry` object for fine-grained control + over different types of retries. + Pass an integer number to retry connection errors that many times, + but no other types of errors. Pass zero to never retry. + + If ``False``, then retries are disabled and any exception is raised + immediately. Also, instead of raising a MaxRetryError on redirects, + the redirect response will be returned. + + :type retries: :class:`~urllib3.util.retry.Retry`, False, or an int. + + :param timeout: + If specified, overrides the default timeout for this one + request. It may be a float (in seconds) or an instance of + :class:`urllib3.util.Timeout`. + + :param chunked: + If True, urllib3 will send the body using chunked transfer + encoding. Otherwise, urllib3 will send the body using the standard + content-length form. Defaults to False. + + :param response_conn: + Set this to ``None`` if you will handle releasing the connection or + set the connection to have the response release it. + + :param preload_content: + If True, the response's body will be preloaded during construction. + + :param decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. + + :param enforce_content_length: + Enforce content length checking. Body returned by server must match + value of Content-Length header, if present. Otherwise, raise error. + """ + self.num_requests += 1 + + timeout_obj = self._get_timeout(timeout) + timeout_obj.start_connect() + conn.timeout = Timeout.resolve_default_timeout(timeout_obj.connect_timeout) + + try: + # Trigger any extra validation we need to do. + try: + self._validate_conn(conn) + except (SocketTimeout, BaseSSLError) as e: + self._raise_timeout(err=e, url=url, timeout_value=conn.timeout) + raise + + # _validate_conn() starts the connection to an HTTPS proxy + # so we need to wrap errors with 'ProxyError' here too. + except ( + OSError, + NewConnectionError, + TimeoutError, + BaseSSLError, + CertificateError, + SSLError, + ) as e: + new_e: Exception = e + if isinstance(e, (BaseSSLError, CertificateError)): + new_e = SSLError(e) + # If the connection didn't successfully connect to it's proxy + # then there + if isinstance( + new_e, (OSError, NewConnectionError, TimeoutError, SSLError) + ) and (conn and conn.proxy and not conn.has_connected_to_proxy): + new_e = _wrap_proxy_error(new_e, conn.proxy.scheme) + raise new_e + + # conn.request() calls http.client.*.request, not the method in + # urllib3.request. It also calls makefile (recv) on the socket. + try: + conn.request( + method, + url, + body=body, + headers=headers, + chunked=chunked, + preload_content=preload_content, + decode_content=decode_content, + enforce_content_length=enforce_content_length, + ) + + # We are swallowing BrokenPipeError (errno.EPIPE) since the server is + # legitimately able to close the connection after sending a valid response. + # With this behaviour, the received response is still readable. + except BrokenPipeError: + pass + except OSError as e: + # MacOS/Linux + # EPROTOTYPE and ECONNRESET are needed on macOS + # https://erickt.github.io/blog/2014/11/19/adventures-in-debugging-a-potential-osx-kernel-bug/ + # Condition changed later to emit ECONNRESET instead of only EPROTOTYPE. + if e.errno != errno.EPROTOTYPE and e.errno != errno.ECONNRESET: + raise + + # Reset the timeout for the recv() on the socket + read_timeout = timeout_obj.read_timeout + + if not conn.is_closed: + # In Python 3 socket.py will catch EAGAIN and return None when you + # try and read into the file pointer created by http.client, which + # instead raises a BadStatusLine exception. Instead of catching + # the exception and assuming all BadStatusLine exceptions are read + # timeouts, check for a zero timeout before making the request. + if read_timeout == 0: + raise ReadTimeoutError( + self, url, f"Read timed out. (read timeout={read_timeout})" + ) + conn.timeout = read_timeout + + # Receive the response from the server + try: + response = conn.getresponse() + except (BaseSSLError, OSError) as e: + self._raise_timeout(err=e, url=url, timeout_value=read_timeout) + raise + + # Set properties that are used by the pooling layer. + response.retries = retries + response._connection = response_conn # type: ignore[attr-defined] + response._pool = self # type: ignore[attr-defined] + + log.debug( + '%s://%s:%s "%s %s %s" %s %s', + self.scheme, + self.host, + self.port, + method, + url, + response.version_string, + response.status, + response.length_remaining, + ) + + return response + + def close(self) -> None: + """ + Close all pooled connections and disable the pool. + """ + if self.pool is None: + return + # Disable access to the pool + old_pool, self.pool = self.pool, None + + # Close all the HTTPConnections in the pool. + _close_pool_connections(old_pool) + + def is_same_host(self, url: str) -> bool: + """ + Check if the given ``url`` is a member of the same host as this + connection pool. + """ + if url.startswith("/"): + return True + + # TODO: Add optional support for socket.gethostbyname checking. + scheme, _, host, port, *_ = parse_url(url) + scheme = scheme or "http" + if host is not None: + host = _normalize_host(host, scheme=scheme) + + # Use explicit default port for comparison when none is given + if self.port and not port: + port = port_by_scheme.get(scheme) + elif not self.port and port == port_by_scheme.get(scheme): + port = None + + return (scheme, host, port) == (self.scheme, self.host, self.port) + + def urlopen( # type: ignore[override] + self, + method: str, + url: str, + body: _TYPE_BODY | None = None, + headers: typing.Mapping[str, str] | None = None, + retries: Retry | bool | int | None = None, + redirect: bool = True, + assert_same_host: bool = True, + timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT, + pool_timeout: int | None = None, + release_conn: bool | None = None, + chunked: bool = False, + body_pos: _TYPE_BODY_POSITION | None = None, + preload_content: bool = True, + decode_content: bool = True, + **response_kw: typing.Any, + ) -> BaseHTTPResponse: + """ + Get a connection from the pool and perform an HTTP request. This is the + lowest level call for making a request, so you'll need to specify all + the raw details. + + .. note:: + + More commonly, it's appropriate to use a convenience method + such as :meth:`request`. + + .. note:: + + `release_conn` will only behave as expected if + `preload_content=False` because we want to make + `preload_content=False` the default behaviour someday soon without + breaking backwards compatibility. + + :param method: + HTTP request method (such as GET, POST, PUT, etc.) + + :param url: + The URL to perform the request on. + + :param body: + Data to send in the request body, either :class:`str`, :class:`bytes`, + an iterable of :class:`str`/:class:`bytes`, or a file-like object. + + :param headers: + Dictionary of custom headers to send, such as User-Agent, + If-None-Match, etc. If None, pool headers are used. If provided, + these headers completely replace any pool-specific headers. + + :param retries: + Configure the number of retries to allow before raising a + :class:`~urllib3.exceptions.MaxRetryError` exception. + + If ``None`` (default) will retry 3 times, see ``Retry.DEFAULT``. Pass a + :class:`~urllib3.util.retry.Retry` object for fine-grained control + over different types of retries. + Pass an integer number to retry connection errors that many times, + but no other types of errors. Pass zero to never retry. + + If ``False``, then retries are disabled and any exception is raised + immediately. Also, instead of raising a MaxRetryError on redirects, + the redirect response will be returned. + + :type retries: :class:`~urllib3.util.retry.Retry`, False, or an int. + + :param redirect: + If True, automatically handle redirects (status codes 301, 302, + 303, 307, 308). Each redirect counts as a retry. Disabling retries + will disable redirect, too. + + :param assert_same_host: + If ``True``, will make sure that the host of the pool requests is + consistent else will raise HostChangedError. When ``False``, you can + use the pool on an HTTP proxy and request foreign hosts. + + :param timeout: + If specified, overrides the default timeout for this one + request. It may be a float (in seconds) or an instance of + :class:`urllib3.util.Timeout`. + + :param pool_timeout: + If set and the pool is set to block=True, then this method will + block for ``pool_timeout`` seconds and raise EmptyPoolError if no + connection is available within the time period. + + :param bool preload_content: + If True, the response's body will be preloaded into memory. + + :param bool decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. + + :param release_conn: + If False, then the urlopen call will not release the connection + back into the pool once a response is received (but will release if + you read the entire contents of the response such as when + `preload_content=True`). This is useful if you're not preloading + the response's content immediately. You will need to call + ``r.release_conn()`` on the response ``r`` to return the connection + back into the pool. If None, it takes the value of ``preload_content`` + which defaults to ``True``. + + :param bool chunked: + If True, urllib3 will send the body using chunked transfer + encoding. Otherwise, urllib3 will send the body using the standard + content-length form. Defaults to False. + + :param int body_pos: + Position to seek to in file-like body in the event of a retry or + redirect. Typically this won't need to be set because urllib3 will + auto-populate the value when needed. + """ + parsed_url = parse_url(url) + destination_scheme = parsed_url.scheme + + if headers is None: + headers = self.headers + + if not isinstance(retries, Retry): + retries = Retry.from_int(retries, redirect=redirect, default=self.retries) + + if release_conn is None: + release_conn = preload_content + + # Check host + if assert_same_host and not self.is_same_host(url): + raise HostChangedError(self, url, retries) + + # Ensure that the URL we're connecting to is properly encoded + if url.startswith("/"): + url = to_str(_encode_target(url)) + else: + url = to_str(parsed_url.url) + + conn = None + + # Track whether `conn` needs to be released before + # returning/raising/recursing. Update this variable if necessary, and + # leave `release_conn` constant throughout the function. That way, if + # the function recurses, the original value of `release_conn` will be + # passed down into the recursive call, and its value will be respected. + # + # See issue #651 [1] for details. + # + # [1] + release_this_conn = release_conn + + http_tunnel_required = connection_requires_http_tunnel( + self.proxy, self.proxy_config, destination_scheme + ) + + # Merge the proxy headers. Only done when not using HTTP CONNECT. We + # have to copy the headers dict so we can safely change it without those + # changes being reflected in anyone else's copy. + if not http_tunnel_required: + headers = headers.copy() # type: ignore[attr-defined] + headers.update(self.proxy_headers) # type: ignore[union-attr] + + # Must keep the exception bound to a separate variable or else Python 3 + # complains about UnboundLocalError. + err = None + + # Keep track of whether we cleanly exited the except block. This + # ensures we do proper cleanup in finally. + clean_exit = False + + # Rewind body position, if needed. Record current position + # for future rewinds in the event of a redirect/retry. + body_pos = set_file_position(body, body_pos) + + try: + # Request a connection from the queue. + timeout_obj = self._get_timeout(timeout) + conn = self._get_conn(timeout=pool_timeout) + + conn.timeout = timeout_obj.connect_timeout # type: ignore[assignment] + + # Is this a closed/new connection that requires CONNECT tunnelling? + if self.proxy is not None and http_tunnel_required and conn.is_closed: + try: + self._prepare_proxy(conn) + except (BaseSSLError, OSError, SocketTimeout) as e: + self._raise_timeout( + err=e, url=self.proxy.url, timeout_value=conn.timeout + ) + raise + + # If we're going to release the connection in ``finally:``, then + # the response doesn't need to know about the connection. Otherwise + # it will also try to release it and we'll have a double-release + # mess. + response_conn = conn if not release_conn else None + + # Make the request on the HTTPConnection object + response = self._make_request( + conn, + method, + url, + timeout=timeout_obj, + body=body, + headers=headers, + chunked=chunked, + retries=retries, + response_conn=response_conn, + preload_content=preload_content, + decode_content=decode_content, + **response_kw, + ) + + # Everything went great! + clean_exit = True + + except EmptyPoolError: + # Didn't get a connection from the pool, no need to clean up + clean_exit = True + release_this_conn = False + raise + + except ( + TimeoutError, + HTTPException, + OSError, + ProtocolError, + BaseSSLError, + SSLError, + CertificateError, + ProxyError, + ) as e: + # Discard the connection for these exceptions. It will be + # replaced during the next _get_conn() call. + clean_exit = False + new_e: Exception = e + if isinstance(e, (BaseSSLError, CertificateError)): + new_e = SSLError(e) + if isinstance( + new_e, + ( + OSError, + NewConnectionError, + TimeoutError, + SSLError, + HTTPException, + ), + ) and (conn and conn.proxy and not conn.has_connected_to_proxy): + new_e = _wrap_proxy_error(new_e, conn.proxy.scheme) + elif isinstance(new_e, (OSError, HTTPException)): + new_e = ProtocolError("Connection aborted.", new_e) + + retries = retries.increment( + method, url, error=new_e, _pool=self, _stacktrace=sys.exc_info()[2] + ) + retries.sleep() + + # Keep track of the error for the retry warning. + err = e + + finally: + if not clean_exit: + # We hit some kind of exception, handled or otherwise. We need + # to throw the connection away unless explicitly told not to. + # Close the connection, set the variable to None, and make sure + # we put the None back in the pool to avoid leaking it. + if conn: + conn.close() + conn = None + release_this_conn = True + + if release_this_conn: + # Put the connection back to be reused. If the connection is + # expired then it will be None, which will get replaced with a + # fresh connection during _get_conn. + self._put_conn(conn) + + if not conn: + # Try again + log.warning( + "Retrying (%r) after connection broken by '%r': %s", retries, err, url + ) + return self.urlopen( + method, + url, + body, + headers, + retries, + redirect, + assert_same_host, + timeout=timeout, + pool_timeout=pool_timeout, + release_conn=release_conn, + chunked=chunked, + body_pos=body_pos, + preload_content=preload_content, + decode_content=decode_content, + **response_kw, + ) + + # Handle redirect? + redirect_location = redirect and response.get_redirect_location() + if redirect_location: + if response.status == 303: + # Change the method according to RFC 9110, Section 15.4.4. + method = "GET" + # And lose the body not to transfer anything sensitive. + body = None + headers = HTTPHeaderDict(headers)._prepare_for_method_change() + + try: + retries = retries.increment(method, url, response=response, _pool=self) + except MaxRetryError: + if retries.raise_on_redirect: + response.drain_conn() + raise + return response + + response.drain_conn() + retries.sleep_for_retry(response) + log.debug("Redirecting %s -> %s", url, redirect_location) + return self.urlopen( + method, + redirect_location, + body, + headers, + retries=retries, + redirect=redirect, + assert_same_host=assert_same_host, + timeout=timeout, + pool_timeout=pool_timeout, + release_conn=release_conn, + chunked=chunked, + body_pos=body_pos, + preload_content=preload_content, + decode_content=decode_content, + **response_kw, + ) + + # Check if we should retry the HTTP response. + has_retry_after = bool(response.headers.get("Retry-After")) + if retries.is_retry(method, response.status, has_retry_after): + try: + retries = retries.increment(method, url, response=response, _pool=self) + except MaxRetryError: + if retries.raise_on_status: + response.drain_conn() + raise + return response + + response.drain_conn() + retries.sleep(response) + log.debug("Retry: %s", url) + return self.urlopen( + method, + url, + body, + headers, + retries=retries, + redirect=redirect, + assert_same_host=assert_same_host, + timeout=timeout, + pool_timeout=pool_timeout, + release_conn=release_conn, + chunked=chunked, + body_pos=body_pos, + preload_content=preload_content, + decode_content=decode_content, + **response_kw, + ) + + return response + + +class HTTPSConnectionPool(HTTPConnectionPool): + """ + Same as :class:`.HTTPConnectionPool`, but HTTPS. + + :class:`.HTTPSConnection` uses one of ``assert_fingerprint``, + ``assert_hostname`` and ``host`` in this order to verify connections. + If ``assert_hostname`` is False, no verification is done. + + The ``key_file``, ``cert_file``, ``cert_reqs``, ``ca_certs``, + ``ca_cert_dir``, ``ssl_version``, ``key_password`` are only used if :mod:`ssl` + is available and are fed into :meth:`urllib3.util.ssl_wrap_socket` to upgrade + the connection socket into an SSL socket. + """ + + scheme = "https" + ConnectionCls: type[BaseHTTPSConnection] = HTTPSConnection + + def __init__( + self, + host: str, + port: int | None = None, + timeout: _TYPE_TIMEOUT | None = _DEFAULT_TIMEOUT, + maxsize: int = 1, + block: bool = False, + headers: typing.Mapping[str, str] | None = None, + retries: Retry | bool | int | None = None, + _proxy: Url | None = None, + _proxy_headers: typing.Mapping[str, str] | None = None, + key_file: str | None = None, + cert_file: str | None = None, + cert_reqs: int | str | None = None, + key_password: str | None = None, + ca_certs: str | None = None, + ssl_version: int | str | None = None, + ssl_minimum_version: ssl.TLSVersion | None = None, + ssl_maximum_version: ssl.TLSVersion | None = None, + assert_hostname: str | typing.Literal[False] | None = None, + assert_fingerprint: str | None = None, + ca_cert_dir: str | None = None, + **conn_kw: typing.Any, + ) -> None: + super().__init__( + host, + port, + timeout, + maxsize, + block, + headers, + retries, + _proxy, + _proxy_headers, + **conn_kw, + ) + + self.key_file = key_file + self.cert_file = cert_file + self.cert_reqs = cert_reqs + self.key_password = key_password + self.ca_certs = ca_certs + self.ca_cert_dir = ca_cert_dir + self.ssl_version = ssl_version + self.ssl_minimum_version = ssl_minimum_version + self.ssl_maximum_version = ssl_maximum_version + self.assert_hostname = assert_hostname + self.assert_fingerprint = assert_fingerprint + + def _prepare_proxy(self, conn: HTTPSConnection) -> None: # type: ignore[override] + """Establishes a tunnel connection through HTTP CONNECT.""" + if self.proxy and self.proxy.scheme == "https": + tunnel_scheme = "https" + else: + tunnel_scheme = "http" + + conn.set_tunnel( + scheme=tunnel_scheme, + host=self._tunnel_host, + port=self.port, + headers=self.proxy_headers, + ) + conn.connect() + + def _new_conn(self) -> BaseHTTPSConnection: + """ + Return a fresh :class:`urllib3.connection.HTTPConnection`. + """ + self.num_connections += 1 + log.debug( + "Starting new HTTPS connection (%d): %s:%s", + self.num_connections, + self.host, + self.port or "443", + ) + + if not self.ConnectionCls or self.ConnectionCls is DummyConnection: # type: ignore[comparison-overlap] + raise ImportError( + "Can't connect to HTTPS URL because the SSL module is not available." + ) + + actual_host: str = self.host + actual_port = self.port + if self.proxy is not None and self.proxy.host is not None: + actual_host = self.proxy.host + actual_port = self.proxy.port + + return self.ConnectionCls( + host=actual_host, + port=actual_port, + timeout=self.timeout.connect_timeout, + cert_file=self.cert_file, + key_file=self.key_file, + key_password=self.key_password, + cert_reqs=self.cert_reqs, + ca_certs=self.ca_certs, + ca_cert_dir=self.ca_cert_dir, + assert_hostname=self.assert_hostname, + assert_fingerprint=self.assert_fingerprint, + ssl_version=self.ssl_version, + ssl_minimum_version=self.ssl_minimum_version, + ssl_maximum_version=self.ssl_maximum_version, + **self.conn_kw, + ) + + def _validate_conn(self, conn: BaseHTTPConnection) -> None: + """ + Called right before a request is made, after the socket is created. + """ + super()._validate_conn(conn) + + # Force connect early to allow us to validate the connection. + if conn.is_closed: + conn.connect() + + # TODO revise this, see https://github.com/urllib3/urllib3/issues/2791 + if not conn.is_verified and not conn.proxy_is_verified: + warnings.warn( + ( + f"Unverified HTTPS request is being made to host '{conn.host}'. " + "Adding certificate verification is strongly advised. See: " + "https://urllib3.readthedocs.io/en/latest/advanced-usage.html" + "#tls-warnings" + ), + InsecureRequestWarning, + ) + + +def connection_from_url(url: str, **kw: typing.Any) -> HTTPConnectionPool: + """ + Given a url, return an :class:`.ConnectionPool` instance of its host. + + This is a shortcut for not having to parse out the scheme, host, and port + of the url before creating an :class:`.ConnectionPool` instance. + + :param url: + Absolute URL string that must include the scheme. Port is optional. + + :param \\**kw: + Passes additional parameters to the constructor of the appropriate + :class:`.ConnectionPool`. Useful for specifying things like + timeout, maxsize, headers, etc. + + Example:: + + >>> conn = connection_from_url('http://google.com/') + >>> r = conn.request('GET', '/') + """ + scheme, _, host, port, *_ = parse_url(url) + scheme = scheme or "http" + port = port or port_by_scheme.get(scheme, 80) + if scheme == "https": + return HTTPSConnectionPool(host, port=port, **kw) # type: ignore[arg-type] + else: + return HTTPConnectionPool(host, port=port, **kw) # type: ignore[arg-type] + + +@typing.overload +def _normalize_host(host: None, scheme: str | None) -> None: ... + + +@typing.overload +def _normalize_host(host: str, scheme: str | None) -> str: ... + + +def _normalize_host(host: str | None, scheme: str | None) -> str | None: + """ + Normalize hosts for comparisons and use with sockets. + """ + + host = normalize_host(host, scheme) + + # httplib doesn't like it when we include brackets in IPv6 addresses + # Specifically, if we include brackets but also pass the port then + # httplib crazily doubles up the square brackets on the Host header. + # Instead, we need to make sure we never pass ``None`` as the port. + # However, for backward compatibility reasons we can't actually + # *assert* that. See http://bugs.python.org/issue28539 + if host and host.startswith("[") and host.endswith("]"): + host = host[1:-1] + return host + + +def _url_from_pool( + pool: HTTPConnectionPool | HTTPSConnectionPool, path: str | None = None +) -> str: + """Returns the URL from a given connection pool. This is mainly used for testing and logging.""" + return Url(scheme=pool.scheme, host=pool.host, port=pool.port, path=path).url + + +def _close_pool_connections(pool: queue.LifoQueue[typing.Any]) -> None: + """Drains a queue of connections and closes each one.""" + try: + while True: + conn = pool.get(block=False) + if conn: + conn.close() + except queue.Empty: + pass # Done. diff --git a/.venv/lib/python3.11/site-packages/urllib3/exceptions.py b/.venv/lib/python3.11/site-packages/urllib3/exceptions.py new file mode 100644 index 0000000000000000000000000000000000000000..0394578287c19042344f338c896c1178ba81fbd6 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/exceptions.py @@ -0,0 +1,327 @@ +from __future__ import annotations + +import socket +import typing +import warnings +from email.errors import MessageDefect +from http.client import IncompleteRead as httplib_IncompleteRead + +if typing.TYPE_CHECKING: + from .connection import HTTPConnection + from .connectionpool import ConnectionPool + from .response import HTTPResponse + from .util.retry import Retry + +# Base Exceptions + + +class HTTPError(Exception): + """Base exception used by this module.""" + + +class HTTPWarning(Warning): + """Base warning used by this module.""" + + +_TYPE_REDUCE_RESULT = tuple[typing.Callable[..., object], tuple[object, ...]] + + +class PoolError(HTTPError): + """Base exception for errors caused within a pool.""" + + def __init__(self, pool: ConnectionPool, message: str) -> None: + self.pool = pool + super().__init__(f"{pool}: {message}") + + def __reduce__(self) -> _TYPE_REDUCE_RESULT: + # For pickling purposes. + return self.__class__, (None, None) + + +class RequestError(PoolError): + """Base exception for PoolErrors that have associated URLs.""" + + def __init__(self, pool: ConnectionPool, url: str, message: str) -> None: + self.url = url + super().__init__(pool, message) + + def __reduce__(self) -> _TYPE_REDUCE_RESULT: + # For pickling purposes. + return self.__class__, (None, self.url, None) + + +class SSLError(HTTPError): + """Raised when SSL certificate fails in an HTTPS connection.""" + + +class ProxyError(HTTPError): + """Raised when the connection to a proxy fails.""" + + # The original error is also available as __cause__. + original_error: Exception + + def __init__(self, message: str, error: Exception) -> None: + super().__init__(message, error) + self.original_error = error + + +class DecodeError(HTTPError): + """Raised when automatic decoding based on Content-Type fails.""" + + +class ProtocolError(HTTPError): + """Raised when something unexpected happens mid-request/response.""" + + +#: Renamed to ProtocolError but aliased for backwards compatibility. +ConnectionError = ProtocolError + + +# Leaf Exceptions + + +class MaxRetryError(RequestError): + """Raised when the maximum number of retries is exceeded. + + :param pool: The connection pool + :type pool: :class:`~urllib3.connectionpool.HTTPConnectionPool` + :param str url: The requested Url + :param reason: The underlying error + :type reason: :class:`Exception` + + """ + + def __init__( + self, pool: ConnectionPool, url: str, reason: Exception | None = None + ) -> None: + self.reason = reason + + message = f"Max retries exceeded with url: {url} (Caused by {reason!r})" + + super().__init__(pool, url, message) + + +class HostChangedError(RequestError): + """Raised when an existing pool gets a request for a foreign host.""" + + def __init__( + self, pool: ConnectionPool, url: str, retries: Retry | int = 3 + ) -> None: + message = f"Tried to open a foreign host with url: {url}" + super().__init__(pool, url, message) + self.retries = retries + + +class TimeoutStateError(HTTPError): + """Raised when passing an invalid state to a timeout""" + + +class TimeoutError(HTTPError): + """Raised when a socket timeout error occurs. + + Catching this error will catch both :exc:`ReadTimeoutErrors + ` and :exc:`ConnectTimeoutErrors `. + """ + + +class ReadTimeoutError(TimeoutError, RequestError): + """Raised when a socket timeout occurs while receiving data from a server""" + + +# This timeout error does not have a URL attached and needs to inherit from the +# base HTTPError +class ConnectTimeoutError(TimeoutError): + """Raised when a socket timeout occurs while connecting to a server""" + + +class NewConnectionError(ConnectTimeoutError, HTTPError): + """Raised when we fail to establish a new connection. Usually ECONNREFUSED.""" + + def __init__(self, conn: HTTPConnection, message: str) -> None: + self.conn = conn + super().__init__(f"{conn}: {message}") + + def __reduce__(self) -> _TYPE_REDUCE_RESULT: + # For pickling purposes. + return self.__class__, (None, None) + + @property + def pool(self) -> HTTPConnection: + warnings.warn( + "The 'pool' property is deprecated and will be removed " + "in urllib3 v2.1.0. Use 'conn' instead.", + DeprecationWarning, + stacklevel=2, + ) + + return self.conn + + +class NameResolutionError(NewConnectionError): + """Raised when host name resolution fails.""" + + def __init__(self, host: str, conn: HTTPConnection, reason: socket.gaierror): + message = f"Failed to resolve '{host}' ({reason})" + super().__init__(conn, message) + + def __reduce__(self) -> _TYPE_REDUCE_RESULT: + # For pickling purposes. + return self.__class__, (None, None, None) + + +class EmptyPoolError(PoolError): + """Raised when a pool runs out of connections and no more are allowed.""" + + +class FullPoolError(PoolError): + """Raised when we try to add a connection to a full pool in blocking mode.""" + + +class ClosedPoolError(PoolError): + """Raised when a request enters a pool after the pool has been closed.""" + + +class LocationValueError(ValueError, HTTPError): + """Raised when there is something wrong with a given URL input.""" + + +class LocationParseError(LocationValueError): + """Raised when get_host or similar fails to parse the URL input.""" + + def __init__(self, location: str) -> None: + message = f"Failed to parse: {location}" + super().__init__(message) + + self.location = location + + +class URLSchemeUnknown(LocationValueError): + """Raised when a URL input has an unsupported scheme.""" + + def __init__(self, scheme: str): + message = f"Not supported URL scheme {scheme}" + super().__init__(message) + + self.scheme = scheme + + +class ResponseError(HTTPError): + """Used as a container for an error reason supplied in a MaxRetryError.""" + + GENERIC_ERROR = "too many error responses" + SPECIFIC_ERROR = "too many {status_code} error responses" + + +class SecurityWarning(HTTPWarning): + """Warned when performing security reducing actions""" + + +class InsecureRequestWarning(SecurityWarning): + """Warned when making an unverified HTTPS request.""" + + +class NotOpenSSLWarning(SecurityWarning): + """Warned when using unsupported SSL library""" + + +class SystemTimeWarning(SecurityWarning): + """Warned when system time is suspected to be wrong""" + + +class InsecurePlatformWarning(SecurityWarning): + """Warned when certain TLS/SSL configuration is not available on a platform.""" + + +class DependencyWarning(HTTPWarning): + """ + Warned when an attempt is made to import a module with missing optional + dependencies. + """ + + +class ResponseNotChunked(ProtocolError, ValueError): + """Response needs to be chunked in order to read it as chunks.""" + + +class BodyNotHttplibCompatible(HTTPError): + """ + Body should be :class:`http.client.HTTPResponse` like + (have an fp attribute which returns raw chunks) for read_chunked(). + """ + + +class IncompleteRead(HTTPError, httplib_IncompleteRead): + """ + Response length doesn't match expected Content-Length + + Subclass of :class:`http.client.IncompleteRead` to allow int value + for ``partial`` to avoid creating large objects on streamed reads. + """ + + partial: int # type: ignore[assignment] + expected: int + + def __init__(self, partial: int, expected: int) -> None: + self.partial = partial + self.expected = expected + + def __repr__(self) -> str: + return "IncompleteRead(%i bytes read, %i more expected)" % ( + self.partial, + self.expected, + ) + + +class InvalidChunkLength(HTTPError, httplib_IncompleteRead): + """Invalid chunk length in a chunked response.""" + + def __init__(self, response: HTTPResponse, length: bytes) -> None: + self.partial: int = response.tell() # type: ignore[assignment] + self.expected: int | None = response.length_remaining + self.response = response + self.length = length + + def __repr__(self) -> str: + return "InvalidChunkLength(got length %r, %i bytes read)" % ( + self.length, + self.partial, + ) + + +class InvalidHeader(HTTPError): + """The header provided was somehow invalid.""" + + +class ProxySchemeUnknown(AssertionError, URLSchemeUnknown): + """ProxyManager does not support the supplied scheme""" + + # TODO(t-8ch): Stop inheriting from AssertionError in v2.0. + + def __init__(self, scheme: str | None) -> None: + # 'localhost' is here because our URL parser parses + # localhost:8080 -> scheme=localhost, remove if we fix this. + if scheme == "localhost": + scheme = None + if scheme is None: + message = "Proxy URL had no scheme, should start with http:// or https://" + else: + message = f"Proxy URL had unsupported scheme {scheme}, should use http:// or https://" + super().__init__(message) + + +class ProxySchemeUnsupported(ValueError): + """Fetching HTTPS resources through HTTPS proxies is unsupported""" + + +class HeaderParsingError(HTTPError): + """Raised by assert_header_parsing, but we convert it to a log.warning statement.""" + + def __init__( + self, defects: list[MessageDefect], unparsed_data: bytes | str | None + ) -> None: + message = f"{defects or 'Unknown'}, unparsed data: {unparsed_data!r}" + super().__init__(message) + + +class UnrewindableBodyError(HTTPError): + """urllib3 encountered an error when trying to rewind a body""" diff --git a/.venv/lib/python3.11/site-packages/urllib3/fields.py b/.venv/lib/python3.11/site-packages/urllib3/fields.py new file mode 100644 index 0000000000000000000000000000000000000000..97c4730cff0df570e1ab47f77e6aa879ec3c36e7 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/fields.py @@ -0,0 +1,341 @@ +from __future__ import annotations + +import email.utils +import mimetypes +import typing + +_TYPE_FIELD_VALUE = typing.Union[str, bytes] +_TYPE_FIELD_VALUE_TUPLE = typing.Union[ + _TYPE_FIELD_VALUE, + tuple[str, _TYPE_FIELD_VALUE], + tuple[str, _TYPE_FIELD_VALUE, str], +] + + +def guess_content_type( + filename: str | None, default: str = "application/octet-stream" +) -> str: + """ + Guess the "Content-Type" of a file. + + :param filename: + The filename to guess the "Content-Type" of using :mod:`mimetypes`. + :param default: + If no "Content-Type" can be guessed, default to `default`. + """ + if filename: + return mimetypes.guess_type(filename)[0] or default + return default + + +def format_header_param_rfc2231(name: str, value: _TYPE_FIELD_VALUE) -> str: + """ + Helper function to format and quote a single header parameter using the + strategy defined in RFC 2231. + + Particularly useful for header parameters which might contain + non-ASCII values, like file names. This follows + `RFC 2388 Section 4.4 `_. + + :param name: + The name of the parameter, a string expected to be ASCII only. + :param value: + The value of the parameter, provided as ``bytes`` or `str``. + :returns: + An RFC-2231-formatted unicode string. + + .. deprecated:: 2.0.0 + Will be removed in urllib3 v2.1.0. This is not valid for + ``multipart/form-data`` header parameters. + """ + import warnings + + warnings.warn( + "'format_header_param_rfc2231' is deprecated and will be " + "removed in urllib3 v2.1.0. This is not valid for " + "multipart/form-data header parameters.", + DeprecationWarning, + stacklevel=2, + ) + + if isinstance(value, bytes): + value = value.decode("utf-8") + + if not any(ch in value for ch in '"\\\r\n'): + result = f'{name}="{value}"' + try: + result.encode("ascii") + except (UnicodeEncodeError, UnicodeDecodeError): + pass + else: + return result + + value = email.utils.encode_rfc2231(value, "utf-8") + value = f"{name}*={value}" + + return value + + +def format_multipart_header_param(name: str, value: _TYPE_FIELD_VALUE) -> str: + """ + Format and quote a single multipart header parameter. + + This follows the `WHATWG HTML Standard`_ as of 2021/06/10, matching + the behavior of current browser and curl versions. Values are + assumed to be UTF-8. The ``\\n``, ``\\r``, and ``"`` characters are + percent encoded. + + .. _WHATWG HTML Standard: + https://html.spec.whatwg.org/multipage/ + form-control-infrastructure.html#multipart-form-data + + :param name: + The name of the parameter, an ASCII-only ``str``. + :param value: + The value of the parameter, a ``str`` or UTF-8 encoded + ``bytes``. + :returns: + A string ``name="value"`` with the escaped value. + + .. versionchanged:: 2.0.0 + Matches the WHATWG HTML Standard as of 2021/06/10. Control + characters are no longer percent encoded. + + .. versionchanged:: 2.0.0 + Renamed from ``format_header_param_html5`` and + ``format_header_param``. The old names will be removed in + urllib3 v2.1.0. + """ + if isinstance(value, bytes): + value = value.decode("utf-8") + + # percent encode \n \r " + value = value.translate({10: "%0A", 13: "%0D", 34: "%22"}) + return f'{name}="{value}"' + + +def format_header_param_html5(name: str, value: _TYPE_FIELD_VALUE) -> str: + """ + .. deprecated:: 2.0.0 + Renamed to :func:`format_multipart_header_param`. Will be + removed in urllib3 v2.1.0. + """ + import warnings + + warnings.warn( + "'format_header_param_html5' has been renamed to " + "'format_multipart_header_param'. The old name will be " + "removed in urllib3 v2.1.0.", + DeprecationWarning, + stacklevel=2, + ) + return format_multipart_header_param(name, value) + + +def format_header_param(name: str, value: _TYPE_FIELD_VALUE) -> str: + """ + .. deprecated:: 2.0.0 + Renamed to :func:`format_multipart_header_param`. Will be + removed in urllib3 v2.1.0. + """ + import warnings + + warnings.warn( + "'format_header_param' has been renamed to " + "'format_multipart_header_param'. The old name will be " + "removed in urllib3 v2.1.0.", + DeprecationWarning, + stacklevel=2, + ) + return format_multipart_header_param(name, value) + + +class RequestField: + """ + A data container for request body parameters. + + :param name: + The name of this request field. Must be unicode. + :param data: + The data/value body. + :param filename: + An optional filename of the request field. Must be unicode. + :param headers: + An optional dict-like object of headers to initially use for the field. + + .. versionchanged:: 2.0.0 + The ``header_formatter`` parameter is deprecated and will + be removed in urllib3 v2.1.0. + """ + + def __init__( + self, + name: str, + data: _TYPE_FIELD_VALUE, + filename: str | None = None, + headers: typing.Mapping[str, str] | None = None, + header_formatter: typing.Callable[[str, _TYPE_FIELD_VALUE], str] | None = None, + ): + self._name = name + self._filename = filename + self.data = data + self.headers: dict[str, str | None] = {} + if headers: + self.headers = dict(headers) + + if header_formatter is not None: + import warnings + + warnings.warn( + "The 'header_formatter' parameter is deprecated and " + "will be removed in urllib3 v2.1.0.", + DeprecationWarning, + stacklevel=2, + ) + self.header_formatter = header_formatter + else: + self.header_formatter = format_multipart_header_param + + @classmethod + def from_tuples( + cls, + fieldname: str, + value: _TYPE_FIELD_VALUE_TUPLE, + header_formatter: typing.Callable[[str, _TYPE_FIELD_VALUE], str] | None = None, + ) -> RequestField: + """ + A :class:`~urllib3.fields.RequestField` factory from old-style tuple parameters. + + Supports constructing :class:`~urllib3.fields.RequestField` from + parameter of key/value strings AND key/filetuple. A filetuple is a + (filename, data, MIME type) tuple where the MIME type is optional. + For example:: + + 'foo': 'bar', + 'fakefile': ('foofile.txt', 'contents of foofile'), + 'realfile': ('barfile.txt', open('realfile').read()), + 'typedfile': ('bazfile.bin', open('bazfile').read(), 'image/jpeg'), + 'nonamefile': 'contents of nonamefile field', + + Field names and filenames must be unicode. + """ + filename: str | None + content_type: str | None + data: _TYPE_FIELD_VALUE + + if isinstance(value, tuple): + if len(value) == 3: + filename, data, content_type = value + else: + filename, data = value + content_type = guess_content_type(filename) + else: + filename = None + content_type = None + data = value + + request_param = cls( + fieldname, data, filename=filename, header_formatter=header_formatter + ) + request_param.make_multipart(content_type=content_type) + + return request_param + + def _render_part(self, name: str, value: _TYPE_FIELD_VALUE) -> str: + """ + Override this method to change how each multipart header + parameter is formatted. By default, this calls + :func:`format_multipart_header_param`. + + :param name: + The name of the parameter, an ASCII-only ``str``. + :param value: + The value of the parameter, a ``str`` or UTF-8 encoded + ``bytes``. + + :meta public: + """ + return self.header_formatter(name, value) + + def _render_parts( + self, + header_parts: ( + dict[str, _TYPE_FIELD_VALUE | None] + | typing.Sequence[tuple[str, _TYPE_FIELD_VALUE | None]] + ), + ) -> str: + """ + Helper function to format and quote a single header. + + Useful for single headers that are composed of multiple items. E.g., + 'Content-Disposition' fields. + + :param header_parts: + A sequence of (k, v) tuples or a :class:`dict` of (k, v) to format + as `k1="v1"; k2="v2"; ...`. + """ + iterable: typing.Iterable[tuple[str, _TYPE_FIELD_VALUE | None]] + + parts = [] + if isinstance(header_parts, dict): + iterable = header_parts.items() + else: + iterable = header_parts + + for name, value in iterable: + if value is not None: + parts.append(self._render_part(name, value)) + + return "; ".join(parts) + + def render_headers(self) -> str: + """ + Renders the headers for this request field. + """ + lines = [] + + sort_keys = ["Content-Disposition", "Content-Type", "Content-Location"] + for sort_key in sort_keys: + if self.headers.get(sort_key, False): + lines.append(f"{sort_key}: {self.headers[sort_key]}") + + for header_name, header_value in self.headers.items(): + if header_name not in sort_keys: + if header_value: + lines.append(f"{header_name}: {header_value}") + + lines.append("\r\n") + return "\r\n".join(lines) + + def make_multipart( + self, + content_disposition: str | None = None, + content_type: str | None = None, + content_location: str | None = None, + ) -> None: + """ + Makes this request field into a multipart request field. + + This method overrides "Content-Disposition", "Content-Type" and + "Content-Location" headers to the request parameter. + + :param content_disposition: + The 'Content-Disposition' of the request body. Defaults to 'form-data' + :param content_type: + The 'Content-Type' of the request body. + :param content_location: + The 'Content-Location' of the request body. + + """ + content_disposition = (content_disposition or "form-data") + "; ".join( + [ + "", + self._render_parts( + (("name", self._name), ("filename", self._filename)) + ), + ] + ) + + self.headers["Content-Disposition"] = content_disposition + self.headers["Content-Type"] = content_type + self.headers["Content-Location"] = content_location diff --git a/.venv/lib/python3.11/site-packages/urllib3/filepost.py b/.venv/lib/python3.11/site-packages/urllib3/filepost.py new file mode 100644 index 0000000000000000000000000000000000000000..14f70b05b4778f91137e4a9e7059d7514aa44d28 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/filepost.py @@ -0,0 +1,89 @@ +from __future__ import annotations + +import binascii +import codecs +import os +import typing +from io import BytesIO + +from .fields import _TYPE_FIELD_VALUE_TUPLE, RequestField + +writer = codecs.lookup("utf-8")[3] + +_TYPE_FIELDS_SEQUENCE = typing.Sequence[ + typing.Union[tuple[str, _TYPE_FIELD_VALUE_TUPLE], RequestField] +] +_TYPE_FIELDS = typing.Union[ + _TYPE_FIELDS_SEQUENCE, + typing.Mapping[str, _TYPE_FIELD_VALUE_TUPLE], +] + + +def choose_boundary() -> str: + """ + Our embarrassingly-simple replacement for mimetools.choose_boundary. + """ + return binascii.hexlify(os.urandom(16)).decode() + + +def iter_field_objects(fields: _TYPE_FIELDS) -> typing.Iterable[RequestField]: + """ + Iterate over fields. + + Supports list of (k, v) tuples and dicts, and lists of + :class:`~urllib3.fields.RequestField`. + + """ + iterable: typing.Iterable[RequestField | tuple[str, _TYPE_FIELD_VALUE_TUPLE]] + + if isinstance(fields, typing.Mapping): + iterable = fields.items() + else: + iterable = fields + + for field in iterable: + if isinstance(field, RequestField): + yield field + else: + yield RequestField.from_tuples(*field) + + +def encode_multipart_formdata( + fields: _TYPE_FIELDS, boundary: str | None = None +) -> tuple[bytes, str]: + """ + Encode a dictionary of ``fields`` using the multipart/form-data MIME format. + + :param fields: + Dictionary of fields or list of (key, :class:`~urllib3.fields.RequestField`). + Values are processed by :func:`urllib3.fields.RequestField.from_tuples`. + + :param boundary: + If not specified, then a random boundary will be generated using + :func:`urllib3.filepost.choose_boundary`. + """ + body = BytesIO() + if boundary is None: + boundary = choose_boundary() + + for field in iter_field_objects(fields): + body.write(f"--{boundary}\r\n".encode("latin-1")) + + writer(body).write(field.render_headers()) + data = field.data + + if isinstance(data, int): + data = str(data) # Backwards compatibility + + if isinstance(data, str): + writer(body).write(data) + else: + body.write(data) + + body.write(b"\r\n") + + body.write(f"--{boundary}--\r\n".encode("latin-1")) + + content_type = f"multipart/form-data; boundary={boundary}" + + return body.getvalue(), content_type diff --git a/.venv/lib/python3.11/site-packages/urllib3/py.typed b/.venv/lib/python3.11/site-packages/urllib3/py.typed new file mode 100644 index 0000000000000000000000000000000000000000..5f3ea3d919363f08ab03edbc85b6099bc4df5647 --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/py.typed @@ -0,0 +1,2 @@ +# Instruct type checkers to look for inline type annotations in this package. +# See PEP 561. diff --git a/.venv/lib/python3.11/site-packages/urllib3/response.py b/.venv/lib/python3.11/site-packages/urllib3/response.py new file mode 100644 index 0000000000000000000000000000000000000000..66c6a687ec06e2da1028b94a0a32b2fcd3580e2b --- /dev/null +++ b/.venv/lib/python3.11/site-packages/urllib3/response.py @@ -0,0 +1,1278 @@ +from __future__ import annotations + +import collections +import io +import json as _json +import logging +import re +import socket +import sys +import typing +import warnings +import zlib +from contextlib import contextmanager +from http.client import HTTPMessage as _HttplibHTTPMessage +from http.client import HTTPResponse as _HttplibHTTPResponse +from socket import timeout as SocketTimeout + +if typing.TYPE_CHECKING: + from ._base_connection import BaseHTTPConnection + +try: + try: + import brotlicffi as brotli # type: ignore[import-not-found] + except ImportError: + import brotli # type: ignore[import-not-found] +except ImportError: + brotli = None + +try: + import zstandard as zstd +except (AttributeError, ImportError, ValueError): # Defensive: + HAS_ZSTD = False +else: + # The package 'zstandard' added the 'eof' property starting + # in v0.18.0 which we require to ensure a complete and + # valid zstd stream was fed into the ZstdDecoder. + # See: https://github.com/urllib3/urllib3/pull/2624 + _zstd_version = tuple( + map(int, re.search(r"^([0-9]+)\.([0-9]+)", zstd.__version__).groups()) # type: ignore[union-attr] + ) + if _zstd_version < (0, 18): # Defensive: + HAS_ZSTD = False + else: + HAS_ZSTD = True + +from . import util +from ._base_connection import _TYPE_BODY +from ._collections import HTTPHeaderDict +from .connection import BaseSSLError, HTTPConnection, HTTPException +from .exceptions import ( + BodyNotHttplibCompatible, + DecodeError, + HTTPError, + IncompleteRead, + InvalidChunkLength, + InvalidHeader, + ProtocolError, + ReadTimeoutError, + ResponseNotChunked, + SSLError, +) +from .util.response import is_fp_closed, is_response_to_head +from .util.retry import Retry + +if typing.TYPE_CHECKING: + from .connectionpool import HTTPConnectionPool + +log = logging.getLogger(__name__) + + +class ContentDecoder: + def decompress(self, data: bytes) -> bytes: + raise NotImplementedError() + + def flush(self) -> bytes: + raise NotImplementedError() + + +class DeflateDecoder(ContentDecoder): + def __init__(self) -> None: + self._first_try = True + self._data = b"" + self._obj = zlib.decompressobj() + + def decompress(self, data: bytes) -> bytes: + if not data: + return data + + if not self._first_try: + return self._obj.decompress(data) + + self._data += data + try: + decompressed = self._obj.decompress(data) + if decompressed: + self._first_try = False + self._data = None # type: ignore[assignment] + return decompressed + except zlib.error: + self._first_try = False + self._obj = zlib.decompressobj(-zlib.MAX_WBITS) + try: + return self.decompress(self._data) + finally: + self._data = None # type: ignore[assignment] + + def flush(self) -> bytes: + return self._obj.flush() + + +class GzipDecoderState: + FIRST_MEMBER = 0 + OTHER_MEMBERS = 1 + SWALLOW_DATA = 2 + + +class GzipDecoder(ContentDecoder): + def __init__(self) -> None: + self._obj = zlib.decompressobj(16 + zlib.MAX_WBITS) + self._state = GzipDecoderState.FIRST_MEMBER + + def decompress(self, data: bytes) -> bytes: + ret = bytearray() + if self._state == GzipDecoderState.SWALLOW_DATA or not data: + return bytes(ret) + while True: + try: + ret += self._obj.decompress(data) + except zlib.error: + previous_state = self._state + # Ignore data after the first error + self._state = GzipDecoderState.SWALLOW_DATA + if previous_state == GzipDecoderState.OTHER_MEMBERS: + # Allow trailing garbage acceptable in other gzip clients + return bytes(ret) + raise + data = self._obj.unused_data + if not data: + return bytes(ret) + self._state = GzipDecoderState.OTHER_MEMBERS + self._obj = zlib.decompressobj(16 + zlib.MAX_WBITS) + + def flush(self) -> bytes: + return self._obj.flush() + + +if brotli is not None: + + class BrotliDecoder(ContentDecoder): + # Supports both 'brotlipy' and 'Brotli' packages + # since they share an import name. The top branches + # are for 'brotlipy' and bottom branches for 'Brotli' + def __init__(self) -> None: + self._obj = brotli.Decompressor() + if hasattr(self._obj, "decompress"): + setattr(self, "decompress", self._obj.decompress) + else: + setattr(self, "decompress", self._obj.process) + + def flush(self) -> bytes: + if hasattr(self._obj, "flush"): + return self._obj.flush() # type: ignore[no-any-return] + return b"" + + +if HAS_ZSTD: + + class ZstdDecoder(ContentDecoder): + def __init__(self) -> None: + self._obj = zstd.ZstdDecompressor().decompressobj() + + def decompress(self, data: bytes) -> bytes: + if not data: + return b"" + data_parts = [self._obj.decompress(data)] + while self._obj.eof and self._obj.unused_data: + unused_data = self._obj.unused_data + self._obj = zstd.ZstdDecompressor().decompressobj() + data_parts.append(self._obj.decompress(unused_data)) + return b"".join(data_parts) + + def flush(self) -> bytes: + ret = self._obj.flush() # note: this is a no-op + if not self._obj.eof: + raise DecodeError("Zstandard data is incomplete") + return ret + + +class MultiDecoder(ContentDecoder): + """ + From RFC7231: + If one or more encodings have been applied to a representation, the + sender that applied the encodings MUST generate a Content-Encoding + header field that lists the content codings in the order in which + they were applied. + """ + + def __init__(self, modes: str) -> None: + self._decoders = [_get_decoder(m.strip()) for m in modes.split(",")] + + def flush(self) -> bytes: + return self._decoders[0].flush() + + def decompress(self, data: bytes) -> bytes: + for d in reversed(self._decoders): + data = d.decompress(data) + return data + + +def _get_decoder(mode: str) -> ContentDecoder: + if "," in mode: + return MultiDecoder(mode) + + # According to RFC 9110 section 8.4.1.3, recipients should + # consider x-gzip equivalent to gzip + if mode in ("gzip", "x-gzip"): + return GzipDecoder() + + if brotli is not None and mode == "br": + return BrotliDecoder() + + if HAS_ZSTD and mode == "zstd": + return ZstdDecoder() + + return DeflateDecoder() + + +class BytesQueueBuffer: + """Memory-efficient bytes buffer + + To return decoded data in read() and still follow the BufferedIOBase API, we need a + buffer to always return the correct amount of bytes. + + This buffer should be filled using calls to put() + + Our maximum memory usage is determined by the sum of the size of: + + * self.buffer, which contains the full data + * the largest chunk that we will copy in get() + + The worst case scenario is a single chunk, in which case we'll make a full copy of + the data inside get(). + """ + + def __init__(self) -> None: + self.buffer: typing.Deque[bytes] = collections.deque() + self._size: int = 0 + + def __len__(self) -> int: + return self._size + + def put(self, data: bytes) -> None: + self.buffer.append(data) + self._size += len(data) + + def get(self, n: int) -> bytes: + if n == 0: + return b"" + elif not self.buffer: + raise RuntimeError("buffer is empty") + elif n < 0: + raise ValueError("n should be > 0") + + fetched = 0 + ret = io.BytesIO() + while fetched < n: + remaining = n - fetched + chunk = self.buffer.popleft() + chunk_length = len(chunk) + if remaining < chunk_length: + left_chunk, right_chunk = chunk[:remaining], chunk[remaining:] + ret.write(left_chunk) + self.buffer.appendleft(right_chunk) + self._size -= remaining + break + else: + ret.write(chunk) + self._size -= chunk_length + fetched += chunk_length + + if not self.buffer: + break + + return ret.getvalue() + + def get_all(self) -> bytes: + buffer = self.buffer + if not buffer: + assert self._size == 0 + return b"" + if len(buffer) == 1: + result = buffer.pop() + else: + ret = io.BytesIO() + ret.writelines(buffer.popleft() for _ in range(len(buffer))) + result = ret.getvalue() + self._size = 0 + return result + + +class BaseHTTPResponse(io.IOBase): + CONTENT_DECODERS = ["gzip", "x-gzip", "deflate"] + if brotli is not None: + CONTENT_DECODERS += ["br"] + if HAS_ZSTD: + CONTENT_DECODERS += ["zstd"] + REDIRECT_STATUSES = [301, 302, 303, 307, 308] + + DECODER_ERROR_CLASSES: tuple[type[Exception], ...] = (IOError, zlib.error) + if brotli is not None: + DECODER_ERROR_CLASSES += (brotli.error,) + + if HAS_ZSTD: + DECODER_ERROR_CLASSES += (zstd.ZstdError,) + + def __init__( + self, + *, + headers: typing.Mapping[str, str] | typing.Mapping[bytes, bytes] | None = None, + status: int, + version: int, + version_string: str, + reason: str | None, + decode_content: bool, + request_url: str | None, + retries: Retry | None = None, + ) -> None: + if isinstance(headers, HTTPHeaderDict): + self.headers = headers + else: + self.headers = HTTPHeaderDict(headers) # type: ignore[arg-type] + self.status = status + self.version = version + self.version_string = version_string + self.reason = reason + self.decode_content = decode_content + self._has_decoded_content = False + self._request_url: str | None = request_url + self.retries = retries + + self.chunked = False + tr_enc = self.headers.get("transfer-encoding", "").lower() + # Don't incur the penalty of creating a list and then discarding it + encodings = (enc.strip() for enc in tr_enc.split(",")) + if "chunked" in encodings: + self.chunked = True + + self._decoder: ContentDecoder | None = None + self.length_remaining: int | None + + def get_redirect_location(self) -> str | None | typing.Literal[False]: + """ + Should we redirect and where to? + + :returns: Truthy redirect location string if we got a redirect status + code and valid location. ``None`` if redirect status and no + location. ``False`` if not a redirect status code. + """ + if self.status in self.REDIRECT_STATUSES: + return self.headers.get("location") + return False + + @property + def data(self) -> bytes: + raise NotImplementedError() + + def json(self) -> typing.Any: + """ + Deserializes the body of the HTTP response as a Python object. + + The body of the HTTP response must be encoded using UTF-8, as per + `RFC 8529 Section 8.1 `_. + + To use a custom JSON decoder pass the result of :attr:`HTTPResponse.data` to + your custom decoder instead. + + If the body of the HTTP response is not decodable to UTF-8, a + `UnicodeDecodeError` will be raised. If the body of the HTTP response is not a + valid JSON document, a `json.JSONDecodeError` will be raised. + + Read more :ref:`here `. + + :returns: The body of the HTTP response as a Python object. + """ + data = self.data.decode("utf-8") + return _json.loads(data) + + @property + def url(self) -> str | None: + raise NotImplementedError() + + @url.setter + def url(self, url: str | None) -> None: + raise NotImplementedError() + + @property + def connection(self) -> BaseHTTPConnection | None: + raise NotImplementedError() + + @property + def retries(self) -> Retry | None: + return self._retries + + @retries.setter + def retries(self, retries: Retry | None) -> None: + # Override the request_url if retries has a redirect location. + if retries is not None and retries.history: + self.url = retries.history[-1].redirect_location + self._retries = retries + + def stream( + self, amt: int | None = 2**16, decode_content: bool | None = None + ) -> typing.Iterator[bytes]: + raise NotImplementedError() + + def read( + self, + amt: int | None = None, + decode_content: bool | None = None, + cache_content: bool = False, + ) -> bytes: + raise NotImplementedError() + + def read1( + self, + amt: int | None = None, + decode_content: bool | None = None, + ) -> bytes: + raise NotImplementedError() + + def read_chunked( + self, + amt: int | None = None, + decode_content: bool | None = None, + ) -> typing.Iterator[bytes]: + raise NotImplementedError() + + def release_conn(self) -> None: + raise NotImplementedError() + + def drain_conn(self) -> None: + raise NotImplementedError() + + def shutdown(self) -> None: + raise NotImplementedError() + + def close(self) -> None: + raise NotImplementedError() + + def _init_decoder(self) -> None: + """ + Set-up the _decoder attribute if necessary. + """ + # Note: content-encoding value should be case-insensitive, per RFC 7230 + # Section 3.2 + content_encoding = self.headers.get("content-encoding", "").lower() + if self._decoder is None: + if content_encoding in self.CONTENT_DECODERS: + self._decoder = _get_decoder(content_encoding) + elif "," in content_encoding: + encodings = [ + e.strip() + for e in content_encoding.split(",") + if e.strip() in self.CONTENT_DECODERS + ] + if encodings: + self._decoder = _get_decoder(content_encoding) + + def _decode( + self, data: bytes, decode_content: bool | None, flush_decoder: bool + ) -> bytes: + """ + Decode the data passed in and potentially flush the decoder. + """ + if not decode_content: + if self._has_decoded_content: + raise RuntimeError( + "Calling read(decode_content=False) is not supported after " + "read(decode_content=True) was called." + ) + return data + + try: + if self._decoder: + data = self._decoder.decompress(data) + self._has_decoded_content = True + except self.DECODER_ERROR_CLASSES as e: + content_encoding = self.headers.get("content-encoding", "").lower() + raise DecodeError( + "Received response with content-encoding: %s, but " + "failed to decode it." % content_encoding, + e, + ) from e + if flush_decoder: + data += self._flush_decoder() + + return data + + def _flush_decoder(self) -> bytes: + """ + Flushes the decoder. Should only be called if the decoder is actually + being used. + """ + if self._decoder: + return self._decoder.decompress(b"") + self._decoder.flush() + return b"" + + # Compatibility methods for `io` module + def readinto(self, b: bytearray) -> int: + temp = self.read(len(b)) + if len(temp) == 0: + return 0 + else: + b[: len(temp)] = temp + return len(temp) + + # Compatibility methods for http.client.HTTPResponse + def getheaders(self) -> HTTPHeaderDict: + warnings.warn( + "HTTPResponse.getheaders() is deprecated and will be removed " + "in urllib3 v2.1.0. Instead access HTTPResponse.headers directly.", + category=DeprecationWarning, + stacklevel=2, + ) + return self.headers + + def getheader(self, name: str, default: str | None = None) -> str | None: + warnings.warn( + "HTTPResponse.getheader() is deprecated and will be removed " + "in urllib3 v2.1.0. Instead use HTTPResponse.headers.get(name, default).", + category=DeprecationWarning, + stacklevel=2, + ) + return self.headers.get(name, default) + + # Compatibility method for http.cookiejar + def info(self) -> HTTPHeaderDict: + return self.headers + + def geturl(self) -> str | None: + return self.url + + +class HTTPResponse(BaseHTTPResponse): + """ + HTTP Response container. + + Backwards-compatible with :class:`http.client.HTTPResponse` but the response ``body`` is + loaded and decoded on-demand when the ``data`` property is accessed. This + class is also compatible with the Python standard library's :mod:`io` + module, and can hence be treated as a readable object in the context of that + framework. + + Extra parameters for behaviour not present in :class:`http.client.HTTPResponse`: + + :param preload_content: + If True, the response's body will be preloaded during construction. + + :param decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. + + :param original_response: + When this HTTPResponse wrapper is generated from an :class:`http.client.HTTPResponse` + object, it's convenient to include the original for debug purposes. It's + otherwise unused. + + :param retries: + The retries contains the last :class:`~urllib3.util.retry.Retry` that + was used during the request. + + :param enforce_content_length: + Enforce content length checking. Body returned by server must match + value of Content-Length header, if present. Otherwise, raise error. + """ + + def __init__( + self, + body: _TYPE_BODY = "", + headers: typing.Mapping[str, str] | typing.Mapping[bytes, bytes] | None = None, + status: int = 0, + version: int = 0, + version_string: str = "HTTP/?", + reason: str | None = None, + preload_content: bool = True, + decode_content: bool = True, + original_response: _HttplibHTTPResponse | None = None, + pool: HTTPConnectionPool | None = None, + connection: HTTPConnection | None = None, + msg: _HttplibHTTPMessage | None = None, + retries: Retry | None = None, + enforce_content_length: bool = True, + request_method: str | None = None, + request_url: str | None = None, + auto_close: bool = True, + sock_shutdown: typing.Callable[[int], None] | None = None, + ) -> None: + super().__init__( + headers=headers, + status=status, + version=version, + version_string=version_string, + reason=reason, + decode_content=decode_content, + request_url=request_url, + retries=retries, + ) + + self.enforce_content_length = enforce_content_length + self.auto_close = auto_close + + self._body = None + self._fp: _HttplibHTTPResponse | None = None + self._original_response = original_response + self._fp_bytes_read = 0 + self.msg = msg + + if body and isinstance(body, (str, bytes)): + self._body = body + + self._pool = pool + self._connection = connection + + if hasattr(body, "read"): + self._fp = body # type: ignore[assignment] + self._sock_shutdown = sock_shutdown + + # Are we using the chunked-style of transfer encoding? + self.chunk_left: int | None = None + + # Determine length of response + self.length_remaining = self._init_length(request_method) + + # Used to return the correct amount of bytes for partial read()s + self._decoded_buffer = BytesQueueBuffer() + + # If requested, preload the body. + if preload_content and not self._body: + self._body = self.read(decode_content=decode_content) + + def release_conn(self) -> None: + if not self._pool or not self._connection: + return None + + self._pool._put_conn(self._connection) + self._connection = None + + def drain_conn(self) -> None: + """ + Read and discard any remaining HTTP response data in the response connection. + + Unread data in the HTTPResponse connection blocks the connection from being released back to the pool. + """ + try: + self.read() + except (HTTPError, OSError, BaseSSLError, HTTPException): + pass + + @property + def data(self) -> bytes: + # For backwards-compat with earlier urllib3 0.4 and earlier. + if self._body: + return self._body # type: ignore[return-value] + + if self._fp: + return self.read(cache_content=True) + + return None # type: ignore[return-value] + + @property + def connection(self) -> HTTPConnection | None: + return self._connection + + def isclosed(self) -> bool: + return is_fp_closed(self._fp) + + def tell(self) -> int: + """ + Obtain the number of bytes pulled over the wire so far. May differ from + the amount of content returned by :meth:``urllib3.response.HTTPResponse.read`` + if bytes are encoded on the wire (e.g, compressed). + """ + return self._fp_bytes_read + + def _init_length(self, request_method: str | None) -> int | None: + """ + Set initial length value for Response content if available. + """ + length: int | None + content_length: str | None = self.headers.get("content-length") + + if content_length is not None: + if self.chunked: + # This Response will fail with an IncompleteRead if it can't be + # received as chunked. This method falls back to attempt reading + # the response before raising an exception. + log.warning( + "Received response with both Content-Length and " + "Transfer-Encoding set. This is expressly forbidden " + "by RFC 7230 sec 3.3.2. Ignoring Content-Length and " + "attempting to process response as Transfer-Encoding: " + "chunked." + ) + return None + + try: + # RFC 7230 section 3.3.2 specifies multiple content lengths can + # be sent in a single Content-Length header + # (e.g. Content-Length: 42, 42). This line ensures the values + # are all valid ints and that as long as the `set` length is 1, + # all values are the same. Otherwise, the header is invalid. + lengths = {int(val) for val in content_length.split(",")} + if len(lengths) > 1: + raise InvalidHeader( + "Content-Length contained multiple " + "unmatching values (%s)" % content_length + ) + length = lengths.pop() + except ValueError: + length = None + else: + if length < 0: + length = None + + else: # if content_length is None + length = None + + # Convert status to int for comparison + # In some cases, httplib returns a status of "_UNKNOWN" + try: + status = int(self.status) + except ValueError: + status = 0 + + # Check for responses that shouldn't include a body + if status in (204, 304) or 100 <= status < 200 or request_method == "HEAD": + length = 0 + + return length + + @contextmanager + def _error_catcher(self) -> typing.Generator[None]: + """ + Catch low-level python exceptions, instead re-raising urllib3 + variants, so that low-level exceptions are not leaked in the + high-level api. + + On exit, release the connection back to the pool. + """ + clean_exit = False + + try: + try: + yield + + except SocketTimeout as e: + # FIXME: Ideally we'd like to include the url in the ReadTimeoutError but + # there is yet no clean way to get at it from this context. + raise ReadTimeoutError(self._pool, None, "Read timed out.") from e # type: ignore[arg-type] + + except BaseSSLError as e: + # FIXME: Is there a better way to differentiate between SSLErrors? + if "read operation timed out" not in str(e): + # SSL errors related to framing/MAC get wrapped and reraised here + raise SSLError(e) from e + + raise ReadTimeoutError(self._pool, None, "Read timed out.") from e # type: ignore[arg-type] + + except IncompleteRead as e: + if ( + e.expected is not None + and e.partial is not None + and e.expected == -e.partial + ): + arg = "Response may not contain content." + else: + arg = f"Connection broken: {e!r}" + raise ProtocolError(arg, e) from e + + except (HTTPException, OSError) as e: + raise ProtocolError(f"Connection broken: {e!r}", e) from e + + # If no exception is thrown, we should avoid cleaning up + # unnecessarily. + clean_exit = True + finally: + # If we didn't terminate cleanly, we need to throw away our + # connection. + if not clean_exit: + # The response may not be closed but we're not going to use it + # anymore so close it now to ensure that the connection is + # released back to the pool. + if self._original_response: + self._original_response.close() + + # Closing the response may not actually be sufficient to close + # everything, so if we have a hold of the connection close that + # too. + if self._connection: + self._connection.close() + + # If we hold the original response but it's closed now, we should + # return the connection back to the pool. + if self._original_response and self._original_response.isclosed(): + self.release_conn() + + def _fp_read( + self, + amt: int | None = None, + *, + read1: bool = False, + ) -> bytes: + """ + Read a response with the thought that reading the number of bytes + larger than can fit in a 32-bit int at a time via SSL in some + known cases leads to an overflow error that has to be prevented + if `amt` or `self.length_remaining` indicate that a problem may + happen. + + The known cases: + * CPython < 3.9.7 because of a bug + https://github.com/urllib3/urllib3/issues/2513#issuecomment-1152559900. + * urllib3 injected with pyOpenSSL-backed SSL-support. + * CPython < 3.10 only when `amt` does not fit 32-bit int. + """ + assert self._fp + c_int_max = 2**31 - 1 + if ( + (amt and amt > c_int_max) + or ( + amt is None + and self.length_remaining + and self.length_remaining > c_int_max + ) + ) and (util.IS_PYOPENSSL or sys.version_info < (3, 10)): + if read1: + return self._fp.read1(c_int_max) + buffer = io.BytesIO() + # Besides `max_chunk_amt` being a maximum chunk size, it + # affects memory overhead of reading a response by this + # method in CPython. + # `c_int_max` equal to 2 GiB - 1 byte is the actual maximum + # chunk size that does not lead to an overflow error, but + # 256 MiB is a compromise. + max_chunk_amt = 2**28 + while amt is None or amt != 0: + if amt is not None: + chunk_amt = min(amt, max_chunk_amt) + amt -= chunk_amt + else: + chunk_amt = max_chunk_amt + data = self._fp.read(chunk_amt) + if not data: + break + buffer.write(data) + del data # to reduce peak memory usage by `max_chunk_amt`. + return buffer.getvalue() + elif read1: + return self._fp.read1(amt) if amt is not None else self._fp.read1() + else: + # StringIO doesn't like amt=None + return self._fp.read(amt) if amt is not None else self._fp.read() + + def _raw_read( + self, + amt: int | None = None, + *, + read1: bool = False, + ) -> bytes: + """ + Reads `amt` of bytes from the socket. + """ + if self._fp is None: + return None # type: ignore[return-value] + + fp_closed = getattr(self._fp, "closed", False) + + with self._error_catcher(): + data = self._fp_read(amt, read1=read1) if not fp_closed else b"" + if amt is not None and amt != 0 and not data: + # Platform-specific: Buggy versions of Python. + # Close the connection when no data is returned + # + # This is redundant to what httplib/http.client _should_ + # already do. However, versions of python released before + # December 15, 2012 (http://bugs.python.org/issue16298) do + # not properly close the connection in all cases. There is + # no harm in redundantly calling close. + self._fp.close() + if ( + self.enforce_content_length + and self.length_remaining is not None + and self.length_remaining != 0 + ): + # This is an edge case that httplib failed to cover due + # to concerns of backward compatibility. We're + # addressing it here to make sure IncompleteRead is + # raised during streaming, so all calls with incorrect + # Content-Length are caught. + raise IncompleteRead(self._fp_bytes_read, self.length_remaining) + elif read1 and ( + (amt != 0 and not data) or self.length_remaining == len(data) + ): + # All data has been read, but `self._fp.read1` in + # CPython 3.12 and older doesn't always close + # `http.client.HTTPResponse`, so we close it here. + # See https://github.com/python/cpython/issues/113199 + self._fp.close() + + if data: + self._fp_bytes_read += len(data) + if self.length_remaining is not None: + self.length_remaining -= len(data) + return data + + def read( + self, + amt: int | None = None, + decode_content: bool | None = None, + cache_content: bool = False, + ) -> bytes: + """ + Similar to :meth:`http.client.HTTPResponse.read`, but with two additional + parameters: ``decode_content`` and ``cache_content``. + + :param amt: + How much of the content to read. If specified, caching is skipped + because it doesn't make sense to cache partial content as the full + response. + + :param decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. + + :param cache_content: + If True, will save the returned data such that the same result is + returned despite of the state of the underlying file object. This + is useful if you want the ``.data`` property to continue working + after having ``.read()`` the file object. (Overridden if ``amt`` is + set.) + """ + self._init_decoder() + if decode_content is None: + decode_content = self.decode_content + + if amt and amt < 0: + # Negative numbers and `None` should be treated the same. + amt = None + elif amt is not None: + cache_content = False + + if len(self._decoded_buffer) >= amt: + return self._decoded_buffer.get(amt) + + data = self._raw_read(amt) + + flush_decoder = amt is None or (amt != 0 and not data) + + if not data and len(self._decoded_buffer) == 0: + return data + + if amt is None: + data = self._decode(data, decode_content, flush_decoder) + if cache_content: + self._body = data + else: + # do not waste memory on buffer when not decoding + if not decode_content: + if self._has_decoded_content: + raise RuntimeError( + "Calling read(decode_content=False) is not supported after " + "read(decode_content=True) was called." + ) + return data + + decoded_data = self._decode(data, decode_content, flush_decoder) + self._decoded_buffer.put(decoded_data) + + while len(self._decoded_buffer) < amt and data: + # TODO make sure to initially read enough data to get past the headers + # For example, the GZ file header takes 10 bytes, we don't want to read + # it one byte at a time + data = self._raw_read(amt) + decoded_data = self._decode(data, decode_content, flush_decoder) + self._decoded_buffer.put(decoded_data) + data = self._decoded_buffer.get(amt) + + return data + + def read1( + self, + amt: int | None = None, + decode_content: bool | None = None, + ) -> bytes: + """ + Similar to ``http.client.HTTPResponse.read1`` and documented + in :meth:`io.BufferedReader.read1`, but with an additional parameter: + ``decode_content``. + + :param amt: + How much of the content to read. + + :param decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. + """ + if decode_content is None: + decode_content = self.decode_content + if amt and amt < 0: + # Negative numbers and `None` should be treated the same. + amt = None + # try and respond without going to the network + if self._has_decoded_content: + if not decode_content: + raise RuntimeError( + "Calling read1(decode_content=False) is not supported after " + "read1(decode_content=True) was called." + ) + if len(self._decoded_buffer) > 0: + if amt is None: + return self._decoded_buffer.get_all() + return self._decoded_buffer.get(amt) + if amt == 0: + return b"" + + # FIXME, this method's type doesn't say returning None is possible + data = self._raw_read(amt, read1=True) + if not decode_content or data is None: + return data + + self._init_decoder() + while True: + flush_decoder = not data + decoded_data = self._decode(data, decode_content, flush_decoder) + self._decoded_buffer.put(decoded_data) + if decoded_data or flush_decoder: + break + data = self._raw_read(8192, read1=True) + + if amt is None: + return self._decoded_buffer.get_all() + return self._decoded_buffer.get(amt) + + def stream( + self, amt: int | None = 2**16, decode_content: bool | None = None + ) -> typing.Generator[bytes]: + """ + A generator wrapper for the read() method. A call will block until + ``amt`` bytes have been read from the connection or until the + connection is closed. + + :param amt: + How much of the content to read. The generator will return up to + much data per iteration, but may return less. This is particularly + likely when using compressed data. However, the empty string will + never be returned. + + :param decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. + """ + if self.chunked and self.supports_chunked_reads(): + yield from self.read_chunked(amt, decode_content=decode_content) + else: + while not is_fp_closed(self._fp) or len(self._decoded_buffer) > 0: + data = self.read(amt=amt, decode_content=decode_content) + + if data: + yield data + + # Overrides from io.IOBase + def readable(self) -> bool: + return True + + def shutdown(self) -> None: + if not self._sock_shutdown: + raise ValueError("Cannot shutdown socket as self._sock_shutdown is not set") + self._sock_shutdown(socket.SHUT_RD) + + def close(self) -> None: + self._sock_shutdown = None + + if not self.closed and self._fp: + self._fp.close() + + if self._connection: + self._connection.close() + + if not self.auto_close: + io.IOBase.close(self) + + @property + def closed(self) -> bool: + if not self.auto_close: + return io.IOBase.closed.__get__(self) # type: ignore[no-any-return] + elif self._fp is None: + return True + elif hasattr(self._fp, "isclosed"): + return self._fp.isclosed() + elif hasattr(self._fp, "closed"): + return self._fp.closed + else: + return True + + def fileno(self) -> int: + if self._fp is None: + raise OSError("HTTPResponse has no file to get a fileno from") + elif hasattr(self._fp, "fileno"): + return self._fp.fileno() + else: + raise OSError( + "The file-like object this HTTPResponse is wrapped " + "around has no file descriptor" + ) + + def flush(self) -> None: + if ( + self._fp is not None + and hasattr(self._fp, "flush") + and not getattr(self._fp, "closed", False) + ): + return self._fp.flush() + + def supports_chunked_reads(self) -> bool: + """ + Checks if the underlying file-like object looks like a + :class:`http.client.HTTPResponse` object. We do this by testing for + the fp attribute. If it is present we assume it returns raw chunks as + processed by read_chunked(). + """ + return hasattr(self._fp, "fp") + + def _update_chunk_length(self) -> None: + # First, we'll figure out length of a chunk and then + # we'll try to read it from socket. + if self.chunk_left is not None: + return None + line = self._fp.fp.readline() # type: ignore[union-attr] + line = line.split(b";", 1)[0] + try: + self.chunk_left = int(line, 16) + except ValueError: + self.close() + if line: + # Invalid chunked protocol response, abort. + raise InvalidChunkLength(self, line) from None + else: + # Truncated at start of next chunk + raise ProtocolError("Response ended prematurely") from None + + def _handle_chunk(self, amt: int | None) -> bytes: + returned_chunk = None + if amt is None: + chunk = self._fp._safe_read(self.chunk_left) # type: ignore[union-attr] + returned_chunk = chunk + self._fp._safe_read(2) # type: ignore[union-attr] # Toss the CRLF at the end of the chunk. + self.chunk_left = None + elif self.chunk_left is not None and amt < self.chunk_left: + value = self._fp._safe_read(amt) # type: ignore[union-attr] + self.chunk_left = self.chunk_left - amt + returned_chunk = value + elif amt == self.chunk_left: + value = self._fp._safe_read(amt) # type: ignore[union-attr] + self._fp._safe_read(2) # type: ignore[union-attr] # Toss the CRLF at the end of the chunk. + self.chunk_left = None + returned_chunk = value + else: # amt > self.chunk_left + returned_chunk = self._fp._safe_read(self.chunk_left) # type: ignore[union-attr] + self._fp._safe_read(2) # type: ignore[union-attr] # Toss the CRLF at the end of the chunk. + self.chunk_left = None + return returned_chunk # type: ignore[no-any-return] + + def read_chunked( + self, amt: int | None = None, decode_content: bool | None = None + ) -> typing.Generator[bytes]: + """ + Similar to :meth:`HTTPResponse.read`, but with an additional + parameter: ``decode_content``. + + :param amt: + How much of the content to read. If specified, caching is skipped + because it doesn't make sense to cache partial content as the full + response. + + :param decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. + """ + self._init_decoder() + # FIXME: Rewrite this method and make it a class with a better structured logic. + if not self.chunked: + raise ResponseNotChunked( + "Response is not chunked. " + "Header 'transfer-encoding: chunked' is missing." + ) + if not self.supports_chunked_reads(): + raise BodyNotHttplibCompatible( + "Body should be http.client.HTTPResponse like. " + "It should have have an fp attribute which returns raw chunks." + ) + + with self._error_catcher(): + # Don't bother reading the body of a HEAD request. + if self._original_response and is_response_to_head(self._original_response): + self._original_response.close() + return None + + # If a response is already read and closed + # then return immediately. + if self._fp.fp is None: # type: ignore[union-attr] + return None + + if amt and amt < 0: + # Negative numbers and `None` should be treated the same, + # but httplib handles only `None` correctly. + amt = None + + while True: + self._update_chunk_length() + if self.chunk_left == 0: + break + chunk = self._handle_chunk(amt) + decoded = self._decode( + chunk, decode_content=decode_content, flush_decoder=False + ) + if decoded: + yield decoded + + if decode_content: + # On CPython and PyPy, we should never need to flush the + # decoder. However, on Jython we *might* need to, so + # lets defensively do it anyway. + decoded = self._flush_decoder() + if decoded: # Platform-specific: Jython. + yield decoded + + # Chunk content ends with \r\n: discard it. + while self._fp is not None: + line = self._fp.fp.readline() + if not line: + # Some sites may not end with '\r\n'. + break + if line == b"\r\n": + break + + # We read everything; close the "file". + if self._original_response: + self._original_response.close() + + @property + def url(self) -> str | None: + """ + Returns the URL that was the source of this response. + If the request that generated this response redirected, this method + will return the final redirect location. + """ + return self._request_url + + @url.setter + def url(self, url: str) -> None: + self._request_url = url + + def __iter__(self) -> typing.Iterator[bytes]: + buffer: list[bytes] = [] + for chunk in self.stream(decode_content=True): + if b"\n" in chunk: + chunks = chunk.split(b"\n") + yield b"".join(buffer) + chunks[0] + b"\n" + for x in chunks[1:-1]: + yield x + b"\n" + if chunks[-1]: + buffer = [chunks[-1]] + else: + buffer = [] + else: + buffer.append(chunk) + if buffer: + yield b"".join(buffer)