instance_id
stringlengths
10
57
file_changes
listlengths
1
15
repo
stringlengths
7
53
base_commit
stringlengths
40
40
problem_statement
stringlengths
11
52.5k
patch
stringlengths
251
7.06M
encode__httpx-2382
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_multipart.py:FileField.get_length", "httpx/_multipart.py:FileField.render_data", "httpx/_multipart.py:MultipartStream.iter_chunks_lengths", "httpx/_multipart.py:MultipartS...
encode/httpx
af56476a8c4b35a0b27e165a7fccae08b307e32c
Add support for streaming multipart/form-data (After initial discussion in https://github.com/encode/httpx/discussions/2227.) As far as I see, there's no way to stream `multipart/form-data` uploads when the `Content-Length` is unknown, since HTTPX tries to get byte size some way or another (last resort is to read it...
diff --git a/httpx/_multipart.py b/httpx/_multipart.py index 2c08776..1d46d96 100644 --- a/httpx/_multipart.py +++ b/httpx/_multipart.py @@ -135,19 +135,18 @@ class FileField: self.file = fileobj self.headers = headers - def get_length(self) -> int: + def get_length(self) -> typing.Optional[in...
encode__httpx-2400
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_multipart.py:FileField.__init__" ], "edited_modules": [ "httpx/_multipart.py:FileField" ] }, "file": "httpx/_multipart.py" }, { "changes": { "added...
encode/httpx
770d4f2254acd5f4c67eaf25634f70ec1640c443
The code in the document does not work Hi. The code in the [document](https://www.python-httpx.org/advanced/#:~:text=files%20%3D%20%7B%27upload%2Dfile%27%3A%20(None%2C%20%27text%20content%27%2C%20%27text/plain%27)%7D) does not work ``` import httpx files = {'upload-file': (None, 'text content', 'text/plain')} r =...
diff --git a/httpx/_multipart.py b/httpx/_multipart.py index 0329649..2c08776 100644 --- a/httpx/_multipart.py +++ b/httpx/_multipart.py @@ -122,8 +122,14 @@ class FileField: # requests does the opposite (it overwrites the header with the 3rd tuple element) headers["Content-Type"] = content_ty...
encode__httpx-2423
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_decoders.py:LineDecoder.__init__", "httpx/_decoders.py:LineDecoder.decode", "httpx/_decoders.py:LineDecoder.flush" ], "edited_modules": [ "httpx/_decoders.py:L...
encode/httpx
e486fbceea7a933baa4b52852681f9c6ac80ac96
LineDecoder is accidentally quadratic: iter_lines() seems to hang forever When calling `Response.iter_lines()`, things can seem to hang forever. The problem is that `LineDecoder` is quadratic in it's string copying behaviour. If a 31MB chunk with 18,768 lines is passed in to `LineDecoder()` then it takes 1m45s to pr...
diff --git a/httpx/_decoders.py b/httpx/_decoders.py index 2f3a447..500ce7f 100644 --- a/httpx/_decoders.py +++ b/httpx/_decoders.py @@ -259,66 +259,56 @@ class LineDecoder: """ Handles incrementally reading lines from text. - Uses universal line decoding, supporting any of `\n`, `\r`, or `\r\n` - as ...
encode__httpx-2481
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/_types.py" }, { "changes": { "added_entities": [ "httpx/_urls.py:URL.raw" ], "added_modules": null, "edited_enti...
encode/httpx
933551c51985375423ca60a880210aef34f1c620
Breaking drop of URL.raw in patch release 0.23.1 The patch release `0.23.1` included a breaking change in #2241 The justification seems to be that "we don't use it", but `url.raw` is a public property used by other libraries including `ddtrace` 0.x versions: https://github.com/DataDog/dd-trace-py/blob/v0.61.3/ddtra...
diff --git a/httpx/_types.py b/httpx/_types.py index f3c1f6e..6b610e1 100644 --- a/httpx/_types.py +++ b/httpx/_types.py @@ -16,6 +16,7 @@ from typing import ( Iterator, List, Mapping, + NamedTuple, Optional, Sequence, Tuple, @@ -31,6 +32,16 @@ if TYPE_CHECKING: # pragma: no cover P...
encode__httpx-2495
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_content.py:encode_content" ], "edited_modules": [ "httpx/_content.py:encode_content" ] }, "file": "httpx/_content.py" } ]
encode/httpx
563a1031f5bd37445d2fdbf622ba5cab926207c9
AsyncClient.request runtime error: "Attempted to send an sync request with an AsyncClient instance" Ran into this issue as well. It was surprising to end up with a `RuntimeError: Attempted to send an sync request with an AsyncClient instance` error, when my actual issue had nothing to do with sync/async (in my case sub...
diff --git a/httpx/_content.py b/httpx/_content.py index 3cbca7a..1c450b7 100644 --- a/httpx/_content.py +++ b/httpx/_content.py @@ -114,7 +114,11 @@ def encode_content( headers = {"Content-Length": str(content_length)} if body else {} return headers, ByteStream(body) - elif isinstance(content, I...
encode__httpx-2523
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_utils.py:primitive_value_to_str" ], "edited_modules": [ "httpx/_utils.py:primitive_value_to_str" ] }, "file": "httpx/_utils.py" } ]
encode/httpx
10a3b68a71d6cb16b170dec4efe2ec1ae197d65f
Support passing bytes keys/values in `params=...` ``` python >>> httpx.get('https://example.org', params={b'testparam': b'testvalue'}).url URL('https://example.org?b%27testparam%27=b%27testvalue%27') ``` I don't think this is intentional, as there is an isinstance check for bytes in the ``flatten_queryparams`` fu...
diff --git a/httpx/_utils.py b/httpx/_utils.py index 1f64dee..1e1570e 100644 --- a/httpx/_utils.py +++ b/httpx/_utils.py @@ -67,7 +67,11 @@ def primitive_value_to_str(value: "PrimitiveData") -> str: return "false" elif value is None: return "" - return str(value) + elif isinstance(value, (s...
encode__httpx-2659
[ { "changes": { "added_entities": [ "httpx/_utils.py:is_ipv4_hostname", "httpx/_utils.py:is_ipv6_hostname" ], "added_modules": [ "httpx/_utils.py:is_ipv4_hostname", "httpx/_utils.py:is_ipv6_hostname" ], "edited_entities": [ "httpx/_utils.py:ge...
encode/httpx
7d7c4f15b8784e4a550d974139acfa64193b32c2
The `get_environment_proxies` function in _utils.py does not support IPv4, IPv6 correctly Hi, I encountered error when my environment `no_proxy` includes IPv6 address like `::1`. It is wrongly transformed into `all://*::1` and causes urlparse error since the _urlparse.py parses the `:1` as port. ![image](https://use...
diff --git a/httpx/_utils.py b/httpx/_utils.py index c55d33a..2568fdc 100644 --- a/httpx/_utils.py +++ b/httpx/_utils.py @@ -1,5 +1,6 @@ import codecs import email.message +import ipaddress import mimetypes import os import re @@ -259,7 +260,16 @@ def get_environment_proxies() -> typing.Dict[str, typing.Optional[s...
encode__httpx-2671
[ { "changes": { "added_entities": [ "httpx/_urlparse.py:is_safe" ], "added_modules": [ "httpx/_urlparse.py:is_safe" ], "edited_entities": [ "httpx/_urlparse.py:quote" ], "edited_modules": [ "httpx/_urlparse.py:quote" ] }, "fi...
encode/httpx
15d09a3bbc20372cd87e48f17f7c9381c8220a0f
Percent sign not encoded when present in query params value Since httpx 0.24 the behavior regarding query params changed, and possibly contains a bug. Use case: passing a pre-signed storage url to some webservice which then downloads the file. The presigned url will contain some percent-escaped sequences. With...
diff --git a/httpx/_urlparse.py b/httpx/_urlparse.py index 0fbec35..6522d91 100644 --- a/httpx/_urlparse.py +++ b/httpx/_urlparse.py @@ -399,7 +399,7 @@ def normalize_path(path: str) -> str: def percent_encode(char: str) -> str: """ - Replace every character in a string with the percent-encoded representatio...
encode__httpx-2701
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_urlparse.py:urlparse" ], "edited_modules": [ "httpx/_urlparse.py:urlparse" ] }, "file": "httpx/_urlparse.py" } ]
encode/httpx
df5dbc05580a4c3225e70153729e2a306d67a472
Square brackets no longer work A GET request such as `/?arr[]=1&arr[]=2` is perfectly legal and it means that you are trying to send an array of values to the endpoint. Unluckily after 0.23.3 it stopped working because requests are sent as `/?arr%5B%5D=1&arr%5B%5D=2`
diff --git a/httpx/_urlparse.py b/httpx/_urlparse.py index 5ee6e58..69ff0b4 100644 --- a/httpx/_urlparse.py +++ b/httpx/_urlparse.py @@ -253,12 +253,19 @@ def urlparse(url: str = "", **kwargs: typing.Optional[str]) -> ParseResult: if has_authority: path = normalize_path(path) - parsed_path: str = quo...
encode__httpx-273
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py:Client.check_concurrency_backend" ], "edited_modules": [ "httpx/client.py:Client" ] }, "file": "httpx/client.py" }, { "changes": { "added_...
encode/httpx
a4b93b91c03239738f6c118570fdfa155e7fe11e
Backend test parametrization Now that almost all of HTTPX internals are relying on the concurrency backend, and before tackling the Trio backend (#120), it would be good to parametrize all tests with a `backend` fixture. For now, it would only point to `AsyncioBackend`, but we're going to need it for `TrioBackend` a...
diff --git a/httpx/client.py b/httpx/client.py index f7a3c5d..37e824f 100644 --- a/httpx/client.py +++ b/httpx/client.py @@ -633,12 +633,17 @@ class Client(BaseClient): # concurrency backends. # The sync client performs I/O on its own, so it doesn't need to support # arbitrary concurrency bac...
encode__httpx-2990
[ { "changes": { "added_entities": [ "httpx/_urlparse.py:percent_encoded" ], "added_modules": [ "httpx/_urlparse.py:percent_encoded" ], "edited_entities": [ "httpx/_urlparse.py:urlparse", "httpx/_urlparse.py:is_safe", "httpx/_urlparse.py:quote"...
encode/httpx
3b9060ee1121b48669e0b30045c9344065f4f2ae
ULR parser percent encoding corrupting query params It appears that the url parser is corrupting query params by incorrectly percent encoding them. I noticed there's [this comment](https://github.com/encode/httpx/blob/0.25.0/httpx/_urlparse.py#L263-L264) which indicates this percent encoding does not accurately foll...
diff --git a/httpx/_urlparse.py b/httpx/_urlparse.py index c44e351..07bbea9 100644 --- a/httpx/_urlparse.py +++ b/httpx/_urlparse.py @@ -260,10 +260,8 @@ def urlparse(url: str = "", **kwargs: typing.Optional[str]) -> ParseResult: # For 'path' we need to drop ? and # from the GEN_DELIMS set. parsed_path: str =...
encode__httpx-301
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py:BaseClient.__init__" ], "edited_modules": [ "httpx/client.py:BaseClient" ] }, "file": "httpx/client.py" }, { "changes": { "added_entities"...
encode/httpx
a46462764e89bbcf2ac9587e11296a171cef8306
Add support for SSLKEYLOGFILE for Python 3.8b4+ Python 3.8b4 (with OpenSSL 1.1.1) added support for the `SSLKEYLOGFILE` environment variable which will be very useful for using Wireshark to do analysis of HTTP/2 requests for example. This should be a small change within `SSLConfig._create_default_ssl_context()` see ...
diff --git a/.travis.yml b/.travis.yml index 07207f9..2a8da53 100644 --- a/.travis.yml +++ b/.travis.yml @@ -20,6 +20,7 @@ matrix: env: NOX_SESSION=test-3.7 - python: 3.8-dev env: NOX_SESSION=test-3.8 + dist: bionic # Required to get OpenSSL 1.1.1+ install: - pip install --upgrade nox diff...
encode__httpx-310
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/middleware.py:RedirectMiddleware.build_redirect_request", "httpx/middleware.py:RedirectMiddleware.redirect_headers" ], "edited_modules": [ "httpx/middleware.py:Redirect...
encode/httpx
b8c5e7a8528978e646c28a5837df5085ea3803fc
h11._util.LocalProtocolError: Too little data for declared Content-Length client POC: ```python import asyncio import logging import httpx logging.basicConfig(level=logging.DEBUG) async def test(): client = httpx.AsyncClient() url = 'http://127.0.0.1:8000/debug' resp = await client.post(url...
diff --git a/httpx/middleware.py b/httpx/middleware.py index 4ed750e..aa994db 100644 --- a/httpx/middleware.py +++ b/httpx/middleware.py @@ -88,7 +88,7 @@ class RedirectMiddleware(BaseMiddleware): ) -> AsyncRequest: method = self.redirect_method(request, response) url = self.redirect_url(request,...
encode__httpx-324
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/dispatch/http2.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/middle...
encode/httpx
6b97b2ed00a03d7528abcb45d59e111bebdf0c1a
Deletion of non-existent keys from `Headers` should raise KeyError. Prompted by #310 To replicate: ```python headers = httpx.Headers() del headers['DOES_NOT_EXIST'] # Should raise a KeyError, but doesn't ``` To resolve this issue we'll need to: * Ensure that deletion of non-existant keys raises a KeyEr...
diff --git a/httpx/dispatch/http2.py b/httpx/dispatch/http2.py index 786110c..fd40964 100644 --- a/httpx/dispatch/http2.py +++ b/httpx/dispatch/http2.py @@ -4,7 +4,7 @@ import typing import h2.connection import h2.events -from ..concurrency.base import BaseStream, ConcurrencyBackend, TimeoutFlag, BaseEvent +from .....
encode__httpx-3412
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__version__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_decoders...
encode/httpx
189fc4bcbe5f314128775dec66a616ac9a31ad48
Zstandard data is incomplete on HEAD requests Tried to do a HEAD request with a server answering with `zstd` encoding by default but since the content is empty `ZStandardDecoder` is raising `DecodingError("Zstandard data is incomplete")` when calling `flush()`. Tried to fix the problem by adding a check if `ret` is ...
diff --git a/CHANGELOG.md b/CHANGELOG.md index 3c16d66..4e2afe2 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,7 +4,7 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/). -## [Unreleased] +## 0.28.0 (...) The...
encode__httpx-377
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py:BaseClient.__init__", "httpx/client.py:BaseClient._get_response", "httpx/client.py:AsyncClient.get", "httpx/client.py:AsyncClient.options", "httpx/client....
encode/httpx
e62e5c3758b10a8830c751acbe588d7b572a9957
SSL handshake failed on verifying the certificate **im using charlesproxy for debugging each requests. Proxy ip and cert is from charlesproxy** My code: ``` import httpx client = httpx.Client(proxies={ "http": "http://192.168.1.78:8888", "https": "http://192.168.1.78:8888", }) client_cer = r'C:\...
diff --git a/httpx/client.py b/httpx/client.py index 30beeb7..73b3069 100644 --- a/httpx/client.py +++ b/httpx/client.py @@ -107,13 +107,6 @@ class BaseClient: else: self.base_url = URL(base_url) - if proxies is None and trust_env: - proxies = typing.cast(ProxiesTypes, get_envi...
encode__httpx-386
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/models.py:QueryParams.__init__" ], "edited_modules": [ "httpx/models.py:QueryParams" ] }, "file": "httpx/models.py" }, { "changes": { "added_entitie...
encode/httpx
31730e709597baaa7b2364fee041dfa985169789
Query params should support list-of-strings values. Hi, I recently tried to use `httpx` instead of `requests` in my simple app that calls few API endpoints and suddenly it did not work. The problem here is that `httpx` and `requests` process query parameters differently. Here's a simple script to see how both...
diff --git a/httpx/models.py b/httpx/models.py index f70fdf4..136aa41 100644 --- a/httpx/models.py +++ b/httpx/models.py @@ -32,6 +32,7 @@ from .exceptions import ( from .multipart import multipart_encode from .status_codes import StatusCode from .utils import ( + flatten_queryparams, guess_json_utf, is...
encode__httpx-502
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py:BaseClient._get_response" ], "edited_modules": [ "httpx/client.py:BaseClient" ] }, "file": "httpx/client.py" }, { "changes": { "added_enti...
encode/httpx
2fcf23bbfe41c5526cb35a65d32dd98dde786988
Logging levels I've been spending some time looking at logging levels with `uvicorn` lately, see eg. https://github.com/encode/uvicorn/pull/474 and would like to take a consistent approach across other projects in `encode`. We've currently got some very detailed logging, that's available, tho it's really at a "dig i...
diff --git a/docs/environment_variables.md b/docs/environment_variables.md index 74c2243..15723fb 100644 --- a/docs/environment_variables.md +++ b/docs/environment_variables.md @@ -16,9 +16,9 @@ and what function they serve: Valid values: `debug`, `trace` (case-insensitive) -If set to `trace`, then low-level detai...
encode__httpx-510
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/api.py:request" ], "edited_modules": [ "httpx/api.py:request" ] }, "file": "httpx/api.py" }, { "changes": { "added_entities": null, "added_mod...
encode/httpx
2fcf23bbfe41c5526cb35a65d32dd98dde786988
Hints on self-signed certificates In https://github.com/encode/httpx/issues/503, @gvbgduh mentioned the need to make HTTP/2 requests to a local server. Because of our current implementation (require TLS for HTTP/2, as browsers do), this typically requires to setup a local self-signed certificate for use in tests. I'...
diff --git a/docs/advanced.md b/docs/advanced.md index f0d402b..3e73f5b 100644 --- a/docs/advanced.md +++ b/docs/advanced.md @@ -362,3 +362,38 @@ is set to `None` or it cannot be inferred from it, HTTPX will default to ... } ``` + +## SSL certificates + +When making a request over HTTPS, HTTPX needs to verify the ...
encode__httpx-521
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__version__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/concurren...
encode/httpx
1a32cf036a825f6eb35395af5388a3b23180a82e
HTTPX claims Python 3.6+ support when asyncio backend TCPStream.start_tls explicitly does not support Python 3.6 As stated, HTTPX explicitly does not support Python 3.6 [here](https://github.com/encode/httpx/blob/master/httpx/concurrency/asyncio.py#L61) but declares Python 3.6 support [here](https://github.com/encode/h...
diff --git a/CHANGELOG.md b/CHANGELOG.md index 1cea654..34d47ec 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,17 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/). +## 0.7.7 (November 15, 2019) + +### Fi...
encode__httpx-524
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__version__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py...
encode/httpx
1a32cf036a825f6eb35395af5388a3b23180a82e
h11._util.RemoteProtocolError: can't handle event type ConnectionClosed when role=SERVER and state=SEND_RESPONSE I intermittently got this error when load testing uvicorn endpoint. This error comes from a proxy endpoint where I am also using `encode/http3` to perform HTTP client calls. ``` File "/project/venv/...
diff --git a/CHANGELOG.md b/CHANGELOG.md index 1cea654..3734b1f 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,23 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/). +## 0.7.8 (November 17, 2019) + +### Ad...
encode__httpx-534
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__version__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py...
encode/httpx
6045ee242fbe0308ce9d5faf654f5de68e1a3dd8
Use the existing runloop for sync calls in async handler Hi, this library looks awesome and I really want to use it; but I seem to have hit a blocker and can't seem to find a solution in the docs. TL;DR: Is there any way to provide our own runloop when using the `sync` calls as looking through the source the sync ca...
diff --git a/CHANGELOG.md b/CHANGELOG.md index 34d47ec..3734b1f 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,12 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/). +## 0.7.8 (November 17, 2019) + +### Ad...
encode__httpx-535
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__version__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py...
encode/httpx
6045ee242fbe0308ce9d5faf654f5de68e1a3dd8
Python 3.7: RuntimeError: read() called while another coroutine is already waiting for incoming data As we already discussed with @florimondmanca in https://github.com/encode/httpx/issues/382, I am reporting this issue here. I get the following error: `RuntimeError: read() called while another coroutine is already...
diff --git a/CHANGELOG.md b/CHANGELOG.md index 34d47ec..3734b1f 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,12 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/). +## 0.7.8 (November 17, 2019) + +### Ad...
encode__httpx-563
[ { "changes": { "added_entities": [ "httpx/client.py:Client.dispatcher_for_url" ], "added_modules": null, "edited_entities": [ "httpx/client.py:Client.send_single_request", "httpx/client.py:Client._dispatcher_for_request" ], "edited_modules": [ ...
encode/httpx
44ad295572f7569f6b02c9868ee22fda594c21a4
Timeout from ConnectionPool on number of concurrent requests larger than max_connections While investigating https://github.com/encode/httpx/issues/551 I noticed that when issuing a high number of concurrent requests to a server an `TimeoutError`. Running this example adapted from #551: ```python import asyncio ...
diff --git a/httpx/client.py b/httpx/client.py index 30e2181..4c594f2 100644 --- a/httpx/client.py +++ b/httpx/client.py @@ -535,7 +535,7 @@ class Client: Sends a single request, without handling any redirections. """ - dispatcher = self._dispatcher_for_request(request, self.proxies) + ...
encode__httpx-574
[ { "changes": { "added_entities": [ "httpx/models.py:Response.is_error" ], "added_modules": null, "edited_entities": [ "httpx/models.py:Response.raise_for_status" ], "edited_modules": [ "httpx/models.py:Response" ] }, "file": "httpx/models...
encode/httpx
9df76ccfe97d4df57a5dad832ef137d68282b432
Add equivalent to requests.response `ok` property Hey there, I'm wondering do we have some equivalent of `requests.response:ok` property in `httpx` library? If no, Is it planning to be added sometime in the future? thx!
diff --git a/docs/compatibility.md b/docs/compatibility.md index a95c12c..d652a0a 100644 --- a/docs/compatibility.md +++ b/docs/compatibility.md @@ -14,6 +14,7 @@ to the API in our own documentation. The following exceptions apply: but also provide lower-cased versions for API compatibility with `requests`. * `stream...
encode__httpx-593
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__init__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules":...
encode/httpx
c08ae7796f01bf0be6b3d43bfd4c4a0b169dca49
Fine-tuning timeouts from the default timeout config Currently, if we want to change only one of the three timeouts (connect, read, write) we need to build a full-blown `TimeoutConfig`. But there is no way to use the default timeout values. ```python >>> import httpx >>> httpx.config.DEFAULT_TIMEOUT_CONFIG Timeou...
diff --git a/README.md b/README.md index 808e77b..358e98a 100644 --- a/README.md +++ b/README.md @@ -24,12 +24,12 @@ or trio, and is able to support making large numbers of requests concurrently. **Note**: *HTTPX should still be considered in alpha. We'd love early users and feedback, but would strongly recommend pin...
encode__httpx-600
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__init__.py" }, { "changes": { "added_entities": [ "httpx/api.py:stream" ], "added_modules": [ "httpx/api.py:s...
encode/httpx
e56e120175f4938d45c91c66bdd9b7f9d415e9bb
Garbage collected responses should warn if still open. See https://github.com/encode/httpx/issues/393#issuecomment-538370554 Reviewing the streaming API Right now we issue streaming requests using... ```python response = await client.get(url, stream=True) ``` And have the following stream-specific methods... ...
diff --git a/CHANGELOG.md b/CHANGELOG.md index e6d5a84..a589d30 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,11 +4,34 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/). +## Master + +### Added + +- Added con...
encode__httpx-603
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__init__.py" }, { "changes": { "added_entities": [ "httpx/concurrency/asyncio.py:AsyncioBackend.fork" ], "added_module...
encode/httpx
bb6e52f3566a79978f590e5a677eb057a68ad3a9
Drop background manager (in favour of a two-function fork primitive). I think we're running against the grain with our current `read`, `write`, `background_manager` API on the backends. Switching to the same style as the `hip` project is using for its concurrency backends is likely a smart thing for us to do here......
diff --git a/httpx/__init__.py b/httpx/__init__.py index 686359a..b6cd6df 100644 --- a/httpx/__init__.py +++ b/httpx/__init__.py @@ -3,12 +3,7 @@ from .api import delete, get, head, options, patch, post, put, request, stream from .auth import BasicAuth, DigestAuth from .client import Client from .concurrency.asyncio...
encode__httpx-627
[ { "changes": { "added_entities": [ "httpx/concurrency/asyncio.py:AsyncioBackend.time" ], "added_modules": null, "edited_entities": null, "edited_modules": [ "httpx/concurrency/asyncio.py:AsyncioBackend" ] }, "file": "httpx/concurrency/asyncio.py" }, ...
encode/httpx
869714fbf5f2a1bfda8d6f906930f49032507d7a
Keep-Alive timeouts We should automatically close any Keep-Alive connections after a timeout period.
diff --git a/docs/compatibility.md b/docs/compatibility.md index 8f4a19b..506340c 100644 --- a/docs/compatibility.md +++ b/docs/compatibility.md @@ -4,7 +4,7 @@ HTTPX aims to be compatible with the `requests` API wherever possible. This documentation outlines places where the API differs... -## Request URLS +## Re...
encode__httpx-629
[ { "changes": { "added_entities": [ "httpx/concurrency/asyncio.py:AsyncioBackend.time" ], "added_modules": null, "edited_entities": null, "edited_modules": [ "httpx/concurrency/asyncio.py:AsyncioBackend" ] }, "file": "httpx/concurrency/asyncio.py" }, ...
encode/httpx
1d25bd58a86d725754c4fe1ae4fc4cd3c22b9b57
HTTP/2 download speeds / flow control settings Which having an initial look at HTTP/3, @jlaine dug out that we seem to have signiifcantly slow uploading right now. First thing to do would be to investigate and replicate, by eg. compare and contrast a simple upload from `httpx` vs `requests/urllib3` - does it replica...
diff --git a/docs/compatibility.md b/docs/compatibility.md index 8f4a19b..506340c 100644 --- a/docs/compatibility.md +++ b/docs/compatibility.md @@ -4,7 +4,7 @@ HTTPX aims to be compatible with the `requests` API wherever possible. This documentation outlines places where the API differs... -## Request URLS +## Re...
encode__httpx-649
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py:Client.merge_url" ], "edited_modules": [ "httpx/client.py:Client" ] }, "file": "httpx/client.py" }, { "changes": { "added_entities": null,...
encode/httpx
6c69e0936b0ee1849f0749f04113ec3e6c92c9ff
ssl.SSLError: [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:852) Hi guys by any chance any of you have seen this issue when using httpx + trio, if so how did you manage to fix it? This is happening when starting a request to an IP on port 80 that then redirects to 443. ``` Traceback (most recent call la...
diff --git a/httpx/client.py b/httpx/client.py index a6c22d8..1ceea14 100644 --- a/httpx/client.py +++ b/httpx/client.py @@ -338,7 +338,8 @@ class Client: """ url = self.base_url.join(relative_url=url) if url.scheme == "http" and hstspreload.in_hsts_preload(url.host): - url = url.c...
encode__httpx-653
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/models.py:URL.__init__" ], "edited_modules": [ "httpx/models.py:URL" ] }, "file": "httpx/models.py" } ]
encode/httpx
5ee512d803d1d6b49dc171e1114f0075618de78e
`params` overrides query string in url Hi, I noticed a difference between `httpx` and `requests` when processing query parameters. I provided the `params` for the url with query string, `requests` would **merge** `params` into query string, but `httpx` replaces the whole query string. Here is a simple script: ...
diff --git a/httpx/models.py b/httpx/models.py index ad3599b..fc938c3 100644 --- a/httpx/models.py +++ b/httpx/models.py @@ -90,9 +90,14 @@ class URL: if self.is_absolute_url: self._uri_reference = self._uri_reference.normalize() - # Add any query parameters. + # Add any query para...
encode__httpx-685
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/auth.py:DigestAuth.__call__" ], "edited_modules": [ "httpx/auth.py:DigestAuth" ] }, "file": "httpx/auth.py" } ]
encode/httpx
bc6163c55a75f2e655ff59301ce0a53fa12973ec
DigestAuth should raise a clear error if the request cannot replay. Our DigestAuth implementation cannot work with non-replayable requests. We ought to raise a nice clear error if `request.stream.is_replayable` is not True.
diff --git a/httpx/auth.py b/httpx/auth.py index e0ef50c..e412c57 100644 --- a/httpx/auth.py +++ b/httpx/auth.py @@ -6,7 +6,7 @@ import typing from base64 import b64encode from urllib.request import parse_http_list -from .exceptions import ProtocolError +from .exceptions import ProtocolError, RequestBodyUnavailable...
encode__httpx-697
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/content_streams.py:AsyncIteratorStream.__init__", "httpx/content_streams.py:AsyncIteratorStream.__aiter__" ], "edited_modules": [ "httpx/content_streams.py:AsyncIterato...
encode/httpx
35b7516674ced88073eace33b5d58c4c4d6bab65
AsnycIteratorStream should error if `__iter__` called twice. For defensive programming, the `AsyncIteratorStream` implementation should track `.stream_consumed` in `__iter__`, and raise a `RuntimeError` if it is called twice.
diff --git a/httpx/content_streams.py b/httpx/content_streams.py index 62d150b..73ef59e 100644 --- a/httpx/content_streams.py +++ b/httpx/content_streams.py @@ -7,6 +7,7 @@ from json import dumps as json_dumps from pathlib import Path from urllib.parse import urlencode +from .exceptions import StreamConsumed from ...
encode__httpx-758
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/decoders.py:DeflateDecoder.__init__", "httpx/decoders.py:DeflateDecoder.decode" ], "edited_modules": [ "httpx/decoders.py:DeflateDecoder" ] }, "file": "ht...
encode/httpx
bdfabe1e9a910e2fc63582199eb26386b69ea551
DecodingError with zlib-compressed responses Using the following minimal WSGI app: ```python import zlib def app(environ, start_response): start_response("200 OK", [("Content-Encoding", "deflate")]) return [zlib.compress(b"hello world")] ``` This works fine in a web browser. Requests is quite ha...
diff --git a/docs/quickstart.md b/docs/quickstart.md index 25c1cea..a147b04 100644 --- a/docs/quickstart.md +++ b/docs/quickstart.md @@ -255,7 +255,7 @@ httpx.exceptions.HttpError: 404 Not Found Any successful response codes will simply return `None` rather than raising an exception. -``` python +```python >>> r....
encode__httpx-763
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/decoders.py:DeflateDecoder.__init__", "httpx/decoders.py:DeflateDecoder.decode" ], "edited_modules": [ "httpx/decoders.py:DeflateDecoder" ] }, "file": "ht...
encode/httpx
956129fbf71495c97844dc7adf4b595a9da4cd18
urllib3.ProxyManager() instantiation is broken. python 3.7.5 httpx 0.11.0 urllib3 1.25.7 ``` $ ipython3 -c 'import httpx; r = httpx.get("https://www.google.com")' parse_url http://127.0.0.1:1234 <class 'httpx.models.URL'> --------------------------------------------------------------------------- TypeError ...
diff --git a/README.md b/README.md index cf6b9b3..27fd542 100644 --- a/README.md +++ b/README.md @@ -39,6 +39,18 @@ Let's get started... '<!doctype html>\n<html>\n<head>\n<title>Example Domain</title>...' ``` +Or, using the async API... + +_Use [IPython](https://ipython.readthedocs.io/en/stable/) or Python 3.8+ wit...
encode__httpx-774
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/client.py:BaseClient.redirect_url" ], "edited_modules": [ "httpx/client.py:BaseClient" ] }, "file": "httpx/client.py" }, { "changes": { "added_entit...
encode/httpx
5a63540e8ab35209849894819ac731c836fdcf27
Handle redirects with missing hostname in `Location:` header. While doing something like this: ``` import httpx def run_sync(url): with httpx.Client(verify=False) as client: response = client.get(url) print(response) run_sync('http://62.28.16.253') ``` I get a `httpx.exceptions.Inv...
diff --git a/httpx/client.py b/httpx/client.py index 590d5d0..4991478 100644 --- a/httpx/client.py +++ b/httpx/client.py @@ -330,6 +330,11 @@ class BaseClient: url = URL(location, allow_relative=True) + # Handle malformed 'Location' headers that are "absolute" form, have no host. + # See: htt...
encode__httpx-823
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "httpx/__version__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_client.p...
encode/httpx
50d337e807839c21e796fd8b01c67d8a672a9721
httpx.Client misses the prepare_request method
diff --git a/CHANGELOG.md b/CHANGELOG.md index 5d81336..748ac94 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,39 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/). +## 0.12.1 (March 19th, 2020) + +### Fi...
encode__httpx-861
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_content_streams.py:encode" ], "edited_modules": [ "httpx/_content_streams.py:encode" ] }, "file": "httpx/_content_streams.py" } ]
encode/httpx
43ec09c3cbb089edd2e83ccc717fb95beef0c189
Redundant `boundary` if files/data is empty If you try to send a post request with empty data and files — you will create a body with one `--boundary--` string, which can break the server. It's because httpx create MultipartStream for `data={}, files={}`: https://github.com/encode/httpx/blob/a82adcc933345c6b8cb1623...
diff --git a/httpx/_content_streams.py b/httpx/_content_streams.py index 5f3237e..afffed3 100644 --- a/httpx/_content_streams.py +++ b/httpx/_content_streams.py @@ -323,7 +323,7 @@ def encode( Handles encoding the given `data`, `files`, and `json`, returning a `ContentStream` implementation. """ - if ...
encode__httpx-883
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_content_streams.py:ByteStream.get_headers" ], "edited_modules": [ "httpx/_content_streams.py:ByteStream" ] }, "file": "httpx/_content_streams.py" } ]
encode/httpx
21f533774f87292c2381376eb40f3216a48eee06
Sync client fails on ASGI apps with unhelpful traceback I tried using httpx with with an ASGI app in the way that was implied by the docs. It throws an error that feels like a problem with httpx, as FastAPI is currently being used in production in other ways. If you confirm that this is a bug in httpx, I'm willing (a...
diff --git a/docs/api.md b/docs/api.md index 3889d8c..c63d4ec 100644 --- a/docs/api.md +++ b/docs/api.md @@ -38,6 +38,13 @@ :docstring: :members: headers cookies params request get head options post put patch delete build_request send close +## `AsyncClient` + +::: httpx.AsyncClient + :docstring: + :m...
encode__httpx-968
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_client.py:Client.init_transport", "httpx/_client.py:Client.init_proxy_transport", "httpx/_client.py:AsyncClient.init_transport", "httpx/_client.py:AsyncClient.init_proxy_t...
encode/httpx
9b6605c3d519142db7ebaf56c13094bf779c7dea
Deprecate ProxyLimits naming in favour of max_connections and max_keepalive Prompted by https://github.com/encode/httpx/pull/942#discussion_r424341967 The `PoolLimits` class uses `soft_limit` and `hard_limit` whereas in httpcore we use `max_connections` and `max_keepalive` which are clearer.
diff --git a/docs/advanced.md b/docs/advanced.md index 4d1f1d8..2c1f1a6 100644 --- a/docs/advanced.md +++ b/docs/advanced.md @@ -404,14 +404,14 @@ response = client.get('http://example.com/') You can control the connection pool size using the `pool_limits` keyword argument on the client. It takes instances of `httpx....
encode__httpx-995
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "httpx/_models.py:Request.prepare" ], "edited_modules": [ "httpx/_models.py:Request" ] }, "file": "httpx/_models.py" } ]
encode/httpx
66a45379594e5cf836e4743931119e76dc8f22c7
Content-Length header is missing in POST/DELETE/PUT/PATCH requests without body ### Checklist <!-- To help keep this issue tracker clean and focused, please make sure you tried *all* the following resources before submitting your question. --> - [x] I searched the [HTTPX documentation](https://www.python-httpx.or...
diff --git a/httpx/_models.py b/httpx/_models.py index 865fd9a..683dde1 100644 --- a/httpx/_models.py +++ b/httpx/_models.py @@ -616,6 +616,9 @@ class Request: auto_headers: typing.List[typing.Tuple[bytes, bytes]] = [] has_host = "host" in self.headers + has_content_length = ( + "c...
encode__starlette-1041
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/responses.py:StreamingResponse.__init__" ], "edited_modules": [ "starlette/responses.py:StreamingResponse" ] }, "file": "starlette/responses.py" } ]
encode/starlette
c300bdc5b88f6e3c4467f361c82e3445811cc70f
StreamingResponse doesn't support PEP-492 asynchronous iterators [PEP-492 asynchronous iterators](https://www.python.org/dev/peps/pep-0492/#asynchronous-iterators-and-async-for) have the following signature: ```python class AsyncIterable: def __aiter__(self): return self async def __anext__(sel...
diff --git a/starlette/responses.py b/starlette/responses.py index 24e3b54..d5f2734 100644 --- a/starlette/responses.py +++ b/starlette/responses.py @@ -1,6 +1,5 @@ import hashlib import http.cookies -import inspect import json import os import stat @@ -204,7 +203,7 @@ class StreamingResponse(Response): m...
encode__starlette-105
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/cors.py:CORSMiddleware.__init__", "starlette/middleware/cors.py:CORSMiddleware.__call__", "starlette/middleware/cors.py:CORSMiddleware.simple_response", "sta...
encode/starlette
cc09042c1ca5ccac78bf0ff689faa5dcd3e04f3a
Credentialed CORS standard requests should not respond with wildcard origins See https://developer.mozilla.org/en-US/docs/Web/HTTP/CORS#Credentialed_requests_and_wildcards If a standard request is made, that includes any cookie headers, then CORSMiddleware *ought* to strictly respond with the requested origin, rath...
diff --git a/starlette/middleware/cors.py b/starlette/middleware/cors.py index 7345531..8bc3380 100644 --- a/starlette/middleware/cors.py +++ b/starlette/middleware/cors.py @@ -32,6 +32,8 @@ class CORSMiddleware: simple_headers = {} if "*" in allow_origins: simple_headers["Access-Control-...
encode__starlette-109
[ { "changes": { "added_entities": [ "starlette/datastructures.py:URL.__repr__" ], "added_modules": null, "edited_entities": [ "starlette/datastructures.py:URL.__init__" ], "edited_modules": [ "starlette/datastructures.py:URL" ] }, "file": ...
encode/starlette
02aaa4bddfe126b1458184b1ee1e8604af5041c7
scope["server"] can be None From https://asgi.readthedocs.io/en/latest/specs/www.html#connection-scope > server: A two-item iterable of [host, port], where host is the listening address for this server as a unicode string, and port is the integer listening port. Optional, defaults to None. https://github.com/enco...
diff --git a/starlette/datastructures.py b/starlette/datastructures.py index 558c8a9..2705fd3 100644 --- a/starlette/datastructures.py +++ b/starlette/datastructures.py @@ -7,16 +7,20 @@ class URL: def __init__(self, url: str = "", scope: Scope = None) -> None: if scope is not None: assert no...
encode__starlette-1147
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/sessions.py:SessionMiddleware.__call__" ], "edited_modules": [ "starlette/middleware/sessions.py:SessionMiddleware" ] }, "file": "starlette/middlew...
encode/starlette
23e15789bf6b879b455f1249e47a43d4c752b3ed
Session cookie should use root path The session cookie currently uses '/'. It should really use the ASGI root path instead, in case the application is submounted.
diff --git a/starlette/middleware/sessions.py b/starlette/middleware/sessions.py index d1b1d5a..a13ec5c 100644 --- a/starlette/middleware/sessions.py +++ b/starlette/middleware/sessions.py @@ -49,14 +49,16 @@ class SessionMiddleware: async def send_wrapper(message: Message) -> None: if message["...
encode__starlette-1164
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/responses.py:RedirectResponse.__init__" ], "edited_modules": [ "starlette/responses.py:RedirectResponse" ] }, "file": "starlette/responses.py" } ]
encode/starlette
995d70c7c6de60b98d799601433f3963b97d28fc
RedirectResponse call to `quote_plus` replaces spaces & `%20`'s with `+` character ### Checklist <!-- Please make sure you check all these items before submitting your bug report. --> - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to f...
diff --git a/starlette/responses.py b/starlette/responses.py index ff122fb..d660cd9 100644 --- a/starlette/responses.py +++ b/starlette/responses.py @@ -7,7 +7,7 @@ import sys import typing from email.utils import formatdate from mimetypes import guess_type as mimetypes_guess_type -from urllib.parse import quote, qu...
encode__starlette-1220
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/staticfiles.py:StaticFiles.get_response", "starlette/staticfiles.py:StaticFiles.lookup_path" ], "edited_modules": [ "starlette/staticfiles.py:StaticFiles" ] ...
encode/starlette
28088573aebc70807ef2aa422000cddb93b57a70
`StaticFiles` causes Internal Server Error when user accesses existing files as directory ### Checklist <!-- Please make sure you check all these items before submitting your bug report. --> - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull reques...
diff --git a/starlette/staticfiles.py b/starlette/staticfiles.py index 33ea0b0..39a6972 100644 --- a/starlette/staticfiles.py +++ b/starlette/staticfiles.py @@ -7,12 +7,8 @@ from email.utils import parsedate import anyio from starlette.datastructures import URL, Headers -from starlette.responses import ( - FileR...
encode__starlette-1240
[ { "changes": { "added_entities": [ "starlette/datastructures.py:MutableHeaders.__ior__", "starlette/datastructures.py:MutableHeaders.__or__" ], "added_modules": null, "edited_entities": [ "starlette/datastructures.py:MutableHeaders.update" ], "edited_m...
encode/starlette
c91014dae84012cb59c1c22ae058aa9733212284
Unable to merge MutableHeaders with dict ### Checklist - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to fix it yet. ### Describe the bug Unable to merge `MutableHeaders` with `dict`. ### To reproduce ```python3 from starlett...
diff --git a/starlette/datastructures.py b/starlette/datastructures.py index 1a8b965..5986328 100644 --- a/starlette/datastructures.py +++ b/starlette/datastructures.py @@ -618,6 +618,19 @@ class MutableHeaders(Headers): for idx in reversed(pop_indexes): del self._list[idx] + def __ior__(self...
encode__starlette-134
[ { "changes": { "added_entities": [ "starlette/lifespan.py:LifespanContext.wait_shutdown" ], "added_modules": null, "edited_entities": [ "starlette/lifespan.py:LifespanHandler.add_event_handler", "starlette/lifespan.py:LifespanHandler.run_lifespan", "starle...
encode/starlette
49f76ab5e9ef0ce5d966485553ad6019a9d37da5
Support `shutdown` as a synonym for `cleanup` * Support either `cleanup` or `shutdown` as the ASGI lifespan message name. * Update uvicorn to move to shutdown - https://github.com/encode/uvicorn/issues/233 * Finally, after a small period of time, drop `cleanup` Easy PR for a contributor to jump on would be address...
diff --git a/docs/applications.md b/docs/applications.md index 2cce4ab..d4b44ca 100644 --- a/docs/applications.md +++ b/docs/applications.md @@ -56,7 +56,7 @@ There are two ways to add event handlers: * `@app.on_event(event_type)` - Add an event, decorator style * `app.add_event_handler(event_type, func)` - Add an ev...
encode__starlette-1349
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/formparsers.py:MultiPartParser.parse" ], "edited_modules": [ "starlette/formparsers.py:MultiPartParser" ] }, "file": "starlette/formparsers.py" } ]
encode/starlette
f12c92a21500d484b3d48f965bb605c1bbe193bc
Insufficient input validation of content-type 'multipart/form-data' ### Checklist - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to fix it yet. ### Describe the bug Not actually a bug, but insufficient input validation. ### To re...
diff --git a/starlette/formparsers.py b/starlette/formparsers.py index decaf0b..fd19492 100644 --- a/starlette/formparsers.py +++ b/starlette/formparsers.py @@ -159,7 +159,7 @@ class MultiPartParser: charset = params.get(b"charset", "utf-8") if type(charset) == bytes: charset = charset.de...
encode__starlette-1350
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/staticfiles.py:StaticFiles.__init__", "starlette/staticfiles.py:StaticFiles.get_directories" ], "edited_modules": [ "starlette/staticfiles.py:StaticFiles" ] ...
encode/starlette
1d1dcba1b811265b7e80e143b21ad316046ccdab
StaticFiles support for directories other than "statics" ### Checklist <!-- Please make sure you check all these items before submitting your feature request. --> - [/] There are no similar issues or pull requests for this yet. - [X - tried to get feedback but no replies] I discussed this idea on the [community ...
diff --git a/docs/staticfiles.md b/docs/staticfiles.md index d8786af..3591b4f 100644 --- a/docs/staticfiles.md +++ b/docs/staticfiles.md @@ -6,7 +6,7 @@ Starlette also includes a `StaticFiles` class for serving files in a given direc Signature: `StaticFiles(directory=None, packages=None, check_dir=True)` * `directo...
encode__starlette-1356
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/routing.py:WebSocketRoute.__init__" ], "edited_modules": [ "starlette/routing.py:WebSocketRoute" ] }, "file": "starlette/routing.py" } ]
encode/starlette
f53faba229e3fa2844bc3753e233d9c1f54cca52
WebSocketRoute does not work with functools.partial ### Checklist - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to fix it yet. ### Describe the bug Accessing a WebSocketRoute with async function wrapped in functools.partial raises Type...
diff --git a/starlette/routing.py b/starlette/routing.py index 3c11c1b..982980c 100644 --- a/starlette/routing.py +++ b/starlette/routing.py @@ -276,7 +276,10 @@ class WebSocketRoute(BaseRoute): self.endpoint = endpoint self.name = get_name(endpoint) if name is None else name - if inspect.isf...
encode__starlette-1377
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/staticfiles.py:StaticFiles.__init__", "starlette/staticfiles.py:StaticFiles.get_directories", "starlette/staticfiles.py:StaticFiles.get_path", "starlette/staticfiles.py...
encode/starlette
d81545c71a7988cfd57c613be02f4661449c0793
StaticFiles middleware doesn't follow symlinks ### Checklist - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to fix it yet. ### Describe the bug The StaticFiles middleware is checking the `os.realpath` of a file and returning a 404 f...
diff --git a/starlette/staticfiles.py b/starlette/staticfiles.py index d09630f..da10a39 100644 --- a/starlette/staticfiles.py +++ b/starlette/staticfiles.py @@ -3,6 +3,7 @@ import os import stat import typing from email.utils import parsedate +from pathlib import Path import anyio @@ -51,7 +52,7 @@ class Static...
encode__starlette-1395
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/responses.py:Response.init_headers" ], "edited_modules": [ "starlette/responses.py:Response" ] }, "file": "starlette/responses.py" } ]
encode/starlette
9d686a7125131fe026071d04e0d0e0a726e1afc5
Missing content-length header field in some responses ### Checklist <!-- Please make sure you check all these items before submitting your bug report. --> - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to fix it yet. ### Describe th...
diff --git a/starlette/responses.py b/starlette/responses.py index ffde4b9..da765cf 100644 --- a/starlette/responses.py +++ b/starlette/responses.py @@ -70,8 +70,8 @@ class Response: populate_content_length = b"content-length" not in keys populate_content_type = b"content-type" not in keys -...
encode__starlette-1397
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/responses.py:Response.init_headers" ], "edited_modules": [ "starlette/responses.py:Response" ] }, "file": "starlette/responses.py" } ]
encode/starlette
165592fb89ad26383b4a8122b4829bdfe4c64d60
Add EmptyResponse class HTTP responses with status codes 1xx, 204, 205, and 304 are special cases. 1xx, 204, 304 responses must not include a message body. The empty message body for 205 responses can be handled in a few ways, but the only one that would make sense for Starlette would be to add a Content-Length heade...
diff --git a/starlette/responses.py b/starlette/responses.py index da765cf..26d7305 100644 --- a/starlette/responses.py +++ b/starlette/responses.py @@ -71,7 +71,11 @@ class Response: populate_content_type = b"content-type" not in keys body = getattr(self, "body", None) - if body is not N...
encode__starlette-1417
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/websockets.py:WebSocketDisconnect.__init__", "starlette/websockets.py:WebSocket.close", "starlette/websockets.py:WebSocketClose.__init__", "starlette/websockets.py:WebS...
encode/starlette
9d282a937c968b6b3068730af2c8967aec80b736
[Feature] Reason for websocket.close I've recently started actively developing with websockets in starlette. When closing a websocket you can provide a code e.g. 1000, 1008 … to tell the client the basic reason. But the websocket protocol allows the developer to also give UTF-8-encoded data as a complementary descripti...
diff --git a/docs/websockets.md b/docs/websockets.md index 43406ac..1128bce 100644 --- a/docs/websockets.md +++ b/docs/websockets.md @@ -75,7 +75,7 @@ Use `websocket.receive_json(data, mode="binary")` to receive JSON over binary da ### Closing the connection -* `await websocket.close(code=1000)` +* `await websocke...
encode__starlette-145
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/formparsers.py:FormParser.on_field_start", "starlette/formparsers.py:FormParser.on_field_name", "starlette/formparsers.py:FormParser.on_field_data", "starlette/formpars...
encode/starlette
a32ea0a8e89567b24f3b8cd04847a1b01c9e98f0
Drop `StaticFile` app. We have `FileResponse` and `StaticFiles`. I think that including the `StaticFile` ASGI app complicates things unnecessarily, and that we should probably remove it. * Drop `StaticFile` app. * Put runtime checks that file exists, and file is a regular file in `FileResponse`.
diff --git a/docs/staticfiles.md b/docs/staticfiles.md index 03d9a58..f1b9d0a 100644 --- a/docs/staticfiles.md +++ b/docs/staticfiles.md @@ -1,20 +1,17 @@ -As well as the `FileResponse` class, Starlette also includes ASGI applications -for serving a specific file or directory: +Starlette also includes an `StaticFiles...
encode__starlette-1459
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/base.py:BaseHTTPMiddleware.__call__" ], "edited_modules": [ "starlette/middleware/base.py:BaseHTTPMiddleware" ] }, "file": "starlette/middleware/ba...
encode/starlette
d6269e2f26fd41aa7d08f72a896b45162df69115
Raising Exceptions in sub-applications routes ### Checklist - [X] The bug is reproducible against the latest release or `master`. - [X] There are no similar issues or pull requests to fix it yet. ### Describe the bug Let's start with this PR: #1262 It's about preventing raise `anyio.ExceptionGroup` in view...
diff --git a/starlette/middleware/base.py b/starlette/middleware/base.py index 423f407..bfb4a54 100644 --- a/starlette/middleware/base.py +++ b/starlette/middleware/base.py @@ -52,6 +52,9 @@ class BaseHTTPMiddleware: assert message["type"] == "http.response.body" yield ...
encode__starlette-1472
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/websockets.py:WebSocket.receive", "starlette/websockets.py:WebSocket.send", "starlette/websockets.py:WebSocket.receive_text", "starlette/websockets.py:WebSocket.receive...
encode/starlette
a6d3d8f0179cb1c2ac283bb5db779f6075854a66
Where assert statements are guarding against invalid ASGI messaging, use RuntimeError instead. ### Checklist - [X] There are no similar issues or pull requests for this yet. - [X] I discussed this idea on the [community chat](https://gitter.im/encode/community) and feedback is positive. ### Is your feature relat...
diff --git a/starlette/websockets.py b/starlette/websockets.py index da74060..03ed199 100644 --- a/starlette/websockets.py +++ b/starlette/websockets.py @@ -34,13 +34,21 @@ class WebSocket(HTTPConnection): if self.client_state == WebSocketState.CONNECTING: message = await self._receive() ...
encode__starlette-1553
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/routing.py:get_name" ], "edited_modules": [ "starlette/routing.py:get_name" ] }, "file": "starlette/routing.py" } ]
encode/starlette
487f05dfc641a0632be8bd405add6454cab4f6b0
Route naming introspection always return "method" for method endpoints Discussion was done at https://gitter.im/encode/community. Bug was confirmed by @Kludex. ## Description methods don't get detected on `is_function`, then we assume that `<object>.__class__.__name__ ` will give the right name (my_method on the e...
diff --git a/starlette/routing.py b/starlette/routing.py index 0388304..ea6ec21 100644 --- a/starlette/routing.py +++ b/starlette/routing.py @@ -84,7 +84,7 @@ def websocket_session(func: typing.Callable) -> ASGIApp: def get_name(endpoint: typing.Callable) -> str: - if inspect.isfunction(endpoint) or inspect.isc...
encode__starlette-1643
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/formparsers.py:MultiPartParser.parse" ], "edited_modules": [ "starlette/formparsers.py:MultiPartParser" ] }, "file": "starlette/formparsers.py" } ]
encode/starlette
82e07b3492802c43970314414dbd44d783f0e281
multipart/form-data subpart without name causes internal error ### Checklist <!-- Please make sure you check all these items before submitting your bug report. --> - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to fix it yet. ### De...
diff --git a/setup.cfg b/setup.cfg index 734fa81..8dad329 100644 --- a/setup.cfg +++ b/setup.cfg @@ -30,6 +30,7 @@ filterwarnings= error ignore: run_until_first_complete is deprecated and will be removed in a future version.:DeprecationWarning ignore: starlette\.middleware\.wsgi is deprecated and will be...
encode__starlette-1648
[ { "changes": { "added_entities": [ "starlette/schemas.py:BaseSchemaGenerator._remove_converter" ], "added_modules": null, "edited_entities": [ "starlette/schemas.py:BaseSchemaGenerator.get_endpoints" ], "edited_modules": [ "starlette/schemas.py:BaseSch...
encode/starlette
92c1f1e5503012f48d6bfbc3ed5af4f4bad23e28
Endpoint path parameters type specification compatibility with OAS When specifying a convertor type for a path parameter like so: ``` ... Route('/users/{user_id:int}', user, methods=["GET", "POST"]) ... ``` The OAS schema generated using `SchemaGenerator` interprets the whole portion within `{}` eg. `'user_id:...
diff --git a/starlette/schemas.py b/starlette/schemas.py index 6ca764f..55bf7b3 100644 --- a/starlette/schemas.py +++ b/starlette/schemas.py @@ -1,4 +1,5 @@ import inspect +import re import typing from starlette.requests import Request @@ -49,10 +50,11 @@ class BaseSchemaGenerator: for route in routes: ...
encode__starlette-1715
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/base.py:BaseHTTPMiddleware.__call__" ], "edited_modules": [ "starlette/middleware/base.py:BaseHTTPMiddleware" ] }, "file": "starlette/middleware/ba...
encode/starlette
70971eac55e5c8821d9e6af5fcf7bb995cc10570
Background tasks don't work with middleware that subclasses `BaseHTTPMiddleware` When using background tasks with middleware, requests are not processed until the background task has finished. 1. Use the example below 2. Make several requests in a row - works as expected 3. Uncomment `app.add_middleware(Transparen...
diff --git a/starlette/middleware/base.py b/starlette/middleware/base.py index 49a5e3e..586c987 100644 --- a/starlette/middleware/base.py +++ b/starlette/middleware/base.py @@ -4,12 +4,13 @@ import anyio from starlette.requests import Request from starlette.responses import Response, StreamingResponse -from starlet...
encode__starlette-173
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/routing.py:replace_params", "starlette/routing.py:Route.__init__", "starlette/routing.py:Route.url_path_for", "starlette/routing.py:WebSocketRoute.__init__", "s...
encode/starlette
ed970c86be89a497e8082429726ce94dedcc6c0e
[question] how to mount router into Starlette app? `app.mount('/', router)` does't work Example: ```python from starlette.applications import Starlette from starlette.middleware.cors import CORSMiddleware from starlette.requests import Request from starlette.routing import Router, Route import uvicorn from v...
diff --git a/starlette/routing.py b/starlette/routing.py index 9f29d99..a2b851f 100644 --- a/starlette/routing.py +++ b/starlette/routing.py @@ -8,7 +8,7 @@ from enum import Enum from starlette.datastructures import URL, URLPath from starlette.exceptions import HTTPException from starlette.requests import Request -f...
encode__starlette-195
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/staticfiles.py:StaticFiles.__init__" ], "edited_modules": [ "starlette/staticfiles.py:StaticFiles" ] }, "file": "starlette/staticfiles.py" } ]
encode/starlette
58888bb53fa54991479d8a9a4c32d6b41b334b2c
Check directory exists when instantiating `StaticFiles` The `StaticFiles` application should ensure that the directory exists at the point it is instantiated. (With an optional switch to turn this behavior off)
diff --git a/docs/staticfiles.md b/docs/staticfiles.md index 448cc3b..cc8db5e 100644 --- a/docs/staticfiles.md +++ b/docs/staticfiles.md @@ -1,7 +1,12 @@ -Starlette also includes a `StaticFiles` class for serving a specific directory: +Starlette also includes a `StaticFiles` class for serving files in a given directo...
encode__starlette-208
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/responses.py:Response.__init__", "starlette/responses.py:Response.init_headers" ], "edited_modules": [ "starlette/responses.py:Response" ] }, "file": ...
encode/starlette
3a16d660490b54daad5711749fa00fd7a0970b4e
the raw "Response" should'nt mandatory a content ... no ? I need to manage myself the etag/304 response ... to send a 304 if-none-match, I use : Response(status_code=304,content="",headers=rheaders) But the body content should not be written ... here the content is mandatory : None cause troubble, and emp...
diff --git a/starlette/responses.py b/starlette/responses.py index a6ebe8c..4390c29 100644 --- a/starlette/responses.py +++ b/starlette/responses.py @@ -31,13 +31,16 @@ class Response: def __init__( self, - content: typing.Any, + content: typing.Any = None, status_code: int = 200,...
encode__starlette-2711
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/staticfiles.py:StaticFiles.lookup_path" ], "edited_modules": [ "starlette/staticfiles.py:StaticFiles" ] }, "file": "starlette/staticfiles.py" } ]
encode/starlette
b68a142a356ede730083347f254e1eae8b5c803e
When the directory itself being a symlink does not work. My test code: ```python from fastapi import FastAPI import uvicorn from staticfiles import StaticFiles app = FastAPI() app.mount("/music", StaticFiles(directory="/home/hanxi/work/xiaomusic/music", follow_symlink=True), name="music") uvicorn.run( ...
diff --git a/starlette/staticfiles.py b/starlette/staticfiles.py index 7498c30..746e740 100644 --- a/starlette/staticfiles.py +++ b/starlette/staticfiles.py @@ -155,7 +155,7 @@ class StaticFiles: full_path = os.path.abspath(joined_path) else: full_path = os.path.realpath(j...
encode__starlette-272
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "starlette/datastructures.py" }, { "changes": { "added_entities": [ "starlette/requests.py:HTTPConnection.client" ], "added_m...
encode/starlette
6d04e284d9951b62ba90e6bf5a9a897afaba4a29
Get client IP address Hi, i started using **Starlette** recently and for my application i need to log the client's IP address. It's important to note that i use **Hypercorn** as my ASGI server and it works great. I started searching on Google how to get the IP address from **Starlette**, and i couldn't find it anywhe...
diff --git a/docs/requests.md b/docs/requests.md index e07d687..295fb85 100644 --- a/docs/requests.md +++ b/docs/requests.md @@ -61,6 +61,15 @@ Router path parameters are exposed as a dictionary interface. For example: `request.path_params['username']` +#### Client Address + +The client's remote address is exposed...
encode__starlette-2812
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/base.py:BaseHTTPMiddleware.__call__" ], "edited_modules": [ "starlette/middleware/base.py:BaseHTTPMiddleware" ] }, "file": "starlette/middleware/ba...
encode/starlette
f13d354e18141cea9041ffad603d98197d880a73
middleware causes exceptions to not be raised/handled silently (back again) Regression of #1976 #1977 #1609 #1940 This time I have noticed that changing `MyExc(Exception)` to `MyExc(BaseException)` means the error does get sent to stdout (if that helps) I tried to have a dig but I am not too sure where that catch e...
diff --git a/starlette/middleware/base.py b/starlette/middleware/base.py index f146984..2a59337 100644 --- a/starlette/middleware/base.py +++ b/starlette/middleware/base.py @@ -103,10 +103,9 @@ class BaseHTTPMiddleware: request = _CachedRequest(scope, receive) wrapped_receive = request.wrapped_receive...
encode__starlette-2815
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/formparsers.py:MultiPartParser.__init__" ], "edited_modules": [ "starlette/formparsers.py:MultiPartParser" ] }, "file": "starlette/formparsers.py" }, { ...
encode/starlette
d6ace924c4b8732a54c97c198ee48a2b582f84f5
MultiPartException: Part exceeded maximum size 1M It came from [here](https://github.com/encode/starlette/commit/fd038f3070c302bff17ef7d173dbb0b007617733#diff-59eaf7e36ef8b186735f99bca714455f02f04cee26e014dc89120417996d35b2R121). @ version [0.41.3](https://github.com/encode/starlette/releases/tag/0.41.3) Need to m...
diff --git a/docs/requests.md b/docs/requests.md index 9140bb9..368755f 100644 --- a/docs/requests.md +++ b/docs/requests.md @@ -113,12 +113,12 @@ state with `disconnected = await request.is_disconnected()`. Request files are normally sent as multipart form data (`multipart/form-data`). -Signature: `request.form(m...
encode__starlette-291
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/responses.py:Response.init_headers" ], "edited_modules": [ "starlette/responses.py:Response" ] }, "file": "starlette/responses.py" } ]
encode/starlette
9d20c9f6d6a8e659460e6781745d8a286dc42442
Content-Type and Content-Length headers not automatically populated if headers is given Hi! I was playing around with responses and stumbled upon what seems like a bug. I have already pushed a [fix](https://github.com/florimondmanca/starlette/commit/e6dc80d203b613f211003df63352a37550538e9a) on my fork and would be happ...
diff --git a/starlette/responses.py b/starlette/responses.py index 542e800..3b8d275 100644 --- a/starlette/responses.py +++ b/starlette/responses.py @@ -63,8 +63,8 @@ class Response: for k, v in headers.items() ] keys = [h[0] for h in raw_headers] - populate_content...
encode__starlette-320
[ { "changes": { "added_entities": [ "starlette/requests.py:Request.is_disconnected" ], "added_modules": null, "edited_entities": [ "starlette/requests.py:Request.__init__", "starlette/requests.py:Request.stream" ], "edited_modules": [ "starlette...
encode/starlette
af105b23d53efc41265793d4d74c2720038be529
Detect/handle closed client connections It appears like starlette does not cancel inner processing for when a client disconnects. With the following code it will keep on fetching from the remote, although the client has disconnected already, e.g. when cancelling a curl call. ```python @app.route(…) async def f...
diff --git a/docs/requests.md b/docs/requests.md index 295fb85..d0cbfe7 100644 --- a/docs/requests.md +++ b/docs/requests.md @@ -109,3 +109,7 @@ class App: If you access `.stream()` then the byte chunks are provided without storing the entire body to memory. Any subsequent calls to `.body()`, `.form()`, or `.json()` ...
encode__starlette-33
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/response.py:FileResponse.__call__" ], "edited_modules": [ "starlette/response.py:FileResponse" ] }, "file": "starlette/response.py" } ]
encode/starlette
590db3d6ea6c7f3b03013209822cda2c54dc38ae
Error serving static files larger than 4096 bytes Static files larger than 4096 bytes do not appear to be served correctly. Here's a test I just wrote that illustrates the problem: https://github.com/simonw/starlette/commit/e2d6665fa5c32e77a3fe22836b14620a7f5999bb Running that test gives me the following output: ...
diff --git a/starlette/response.py b/starlette/response.py index db5bac8..f189ef3 100644 --- a/starlette/response.py +++ b/starlette/response.py @@ -177,5 +177,9 @@ class FileResponse(Response): chunk = await file.read(self.chunk_size) more_body = len(chunk) == self.chunk_size ...
encode__starlette-409
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/cors.py:CORSMiddleware.__init__", "starlette/middleware/cors.py:CORSMiddleware.preflight_response" ], "edited_modules": [ "starlette/middleware/cors.py:C...
encode/starlette
9944bb7048458fed0c096a5f77baaa369935cec6
Config/Settings dump? What's the best way to dump all of the settings using starlette.config? I want to create an endpoint to make it easy to review the settings being used but I don't want the Secrets to be viewable, e.g. things like: JWT_SECRET = config("JWT_SECRET", cast=starlette.datastructures.Secret) We...
diff --git a/docs/config.md b/docs/config.md index baa9e38..7c5b2a6 100644 --- a/docs/config.md +++ b/docs/config.md @@ -9,13 +9,13 @@ that is not committed to source control. ```python from starlette.applications import Starlette from starlette.config import Config -from starlette.datastructures import CommaSeparat...
encode__starlette-488
[ { "changes": { "added_entities": [ "starlette/middleware/errors.py:ServerErrorMiddleware.genenrate_frame_html" ], "added_modules": null, "edited_entities": [ "starlette/middleware/errors.py:ServerErrorMiddleware.generate_frame_html", "starlette/middleware/errors.p...
encode/starlette
d23bfd0d8ff68d535d0283aa4099e5055da88bb9
Error in app startup event is not logged and does not prevent startup Related #431 #177 and mentioned here https://github.com/encode/uvicorn/issues/333#issuecomment-484885300 Scenario: I have a 'startup' event handler: ```python @app.on_event('startup') async def startup(): initialise() ``` and I am runn...
diff --git a/starlette/middleware/errors.py b/starlette/middleware/errors.py index 54f5fd2..b56d6ba 100644 --- a/starlette/middleware/errors.py +++ b/starlette/middleware/errors.py @@ -121,7 +121,7 @@ class ServerErrorMiddleware: # to optionally raise the error within the test case. raise exc ...
encode__starlette-557
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/base.py:BaseHTTPMiddleware.call_next" ], "edited_modules": [ "starlette/middleware/base.py:BaseHTTPMiddleware" ] }, "file": "starlette/middleware/b...
encode/starlette
03bfdbc674975a2149b034f32825001e15182f13
mypy errors when accessing Request.state From https://www.starlette.io/requests/: > Other state > > If you want to store additional information on the request you can do so using request.state. > > For example: > > request.state.time_started = time.time() This causes mypy errors since any accessed/set attribu...
diff --git a/starlette/middleware/base.py b/starlette/middleware/base.py index ea5afb2..67d35cc 100644 --- a/starlette/middleware/base.py +++ b/starlette/middleware/base.py @@ -29,7 +29,7 @@ class BaseHTTPMiddleware: loop = asyncio.get_event_loop() queue = asyncio.Queue() # type: asyncio.Queue - ...
encode__starlette-563
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/httpsredirect.py:HTTPSRedirectMiddleware.__call__" ], "edited_modules": [ "starlette/middleware/httpsredirect.py:HTTPSRedirectMiddleware" ] }, "fil...
encode/starlette
b2f9f4359b93a9f941779c575af764019e5183c7
Use "308 Permanent Redirect" for redirect slashes behavior. Hi, I stumbled upon a quirk in starlette that is not properly documented. It seems like all of my HTTP request to a route without trailing slash are being redirected to route with trailing slashes. Say I am hitting `http://hostname/mountpoint/api` it will be ...
diff --git a/starlette/middleware/httpsredirect.py b/starlette/middleware/httpsredirect.py index 13f3a70..7f646ed 100644 --- a/starlette/middleware/httpsredirect.py +++ b/starlette/middleware/httpsredirect.py @@ -13,7 +13,7 @@ class HTTPSRedirectMiddleware: redirect_scheme = {"http": "https", "ws": "wss"}[...
encode__starlette-699
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/datastructures.py:URL.__init__", "starlette/datastructures.py:URLPath.make_absolute_url" ], "edited_modules": [ "starlette/datastructures.py:URL", "starlett...
encode/starlette
1ea45ad54954e6af78a640acc4cd7cc665beb901
Starlette behind reverse proxy, url_for does not return externally valid urls Suppose I have my application reverse proxied, such that http://my.domain/foo (e.g. Traefik or Nginx) has the upstream http://localhost:8000 (the Starlette app). `url_for` will generate urls relative to the upstream url. How do I make it prod...
diff --git a/starlette/datastructures.py b/starlette/datastructures.py index 930c581..4080624 100644 --- a/starlette/datastructures.py +++ b/starlette/datastructures.py @@ -21,7 +21,7 @@ class URL: scheme = scope.get("scheme", "http") server = scope.get("server", None) path = scop...
encode__starlette-701
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/routing.py:Mount.url_path_for" ], "edited_modules": [ "starlette/routing.py:Mount" ] }, "file": "starlette/routing.py" } ]
encode/starlette
a92df1f61ede3b63ee9dcb315ce363c14242597b
Mounts without a name strips the path arg from calls to child routes in url_path_for I'm using an unnamed Mount which contains a mount to a StaticFiles instance and when using "url_for" it is unable to find the matching file. After a bit of digging it seems to be directly related to url_for_path in Mount clearing the p...
diff --git a/starlette/routing.py b/starlette/routing.py index 7a0d7ef..a1e4619 100644 --- a/starlette/routing.py +++ b/starlette/routing.py @@ -336,15 +336,18 @@ class Mount(BaseRoute): else: # 'name' matches "<mount_name>:<child_name>". remaining_name = name[len(self.nam...
encode__starlette-761
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/errors.py:ServerErrorMiddleware.__call__" ], "edited_modules": [ "starlette/middleware/errors.py:ServerErrorMiddleware" ] }, "file": "starlette/mid...
encode/starlette
6182d0a0bc7e5817197d2919b18d67f70e3a71d1
Background tasks exception handler Currently, if a background task faces an exception, it isn't handled and there's no way to handle it (rather than wrapping the task) My suggestion is to be able to add exception handler for background tasks, as there is for requests.
diff --git a/docs/exceptions.md b/docs/exceptions.md index bf460d2..9818a20 100644 --- a/docs/exceptions.md +++ b/docs/exceptions.md @@ -75,6 +75,24 @@ should bubble through the entire middleware stack as exceptions. Any error logging middleware should ensure that it re-raises the exception all the way up to the serv...
encode__starlette-8
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/decorators.py:asgi_application" ], "edited_modules": [ "starlette/decorators.py:asgi_application" ] }, "file": "starlette/decorators.py" }, { "changes...
encode/starlette
4c621d58c0ce2f111fee4745f944ea6417f1cb53
Request should present a scope-like interface The `Request` class should present a dict-like interface so that it can be used in the same way as `scope`. Should also allow it to be instantiated without a `receive` channel being set initially.
diff --git a/README.md b/README.md index f80bdb0..a8fb92f 100644 --- a/README.md +++ b/README.md @@ -152,7 +152,7 @@ the incoming request, rather than accessing the ASGI scope and receive channel d ### Request -Signature: `Request(scope, receive)` +Signature: `Request(scope, receive=None)` ```python class App:...
encode__starlette-801
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/middleware/cors.py:CORSMiddleware.is_allowed_origin" ], "edited_modules": [ "starlette/middleware/cors.py:CORSMiddleware" ] }, "file": "starlette/middleware/c...
encode/starlette
5c43dde0ec0917673bb280bcd7ab0c37b78061b7
Dangerous example regex for CORS Middleware? Looking at the docs for CORS Middlware here: https://www.starlette.io/middleware/#corsmiddleware , under the `allow_origin_regex` attribute, the example value is `https://.*\.example\.org`. However, based on the handler code for this at https://github.com/encode/starlette...
diff --git a/starlette/middleware/cors.py b/starlette/middleware/cors.py index ad2eeff..90ba180 100644 --- a/starlette/middleware/cors.py +++ b/starlette/middleware/cors.py @@ -87,7 +87,7 @@ class CORSMiddleware: if self.allow_all_origins: return True - if self.allow_origin_regex is not N...
encode__starlette-867
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "starlette/requests.py:HTTPConnection.cookies" ], "edited_modules": [ "starlette/requests.py:HTTPConnection" ] }, "file": "starlette/requests.py" } ]
encode/starlette
0cb2d909a2a6d21219c66f8eec75a95f4b6d1c53
[bug] Invalid cookie name leads to exception When handling a request with an invalid cookie name (does not conform to RFC2109) starlette raises an exception. i.e iam/cookiename This is because Starlette uses Python's stdlib cookie library, which is very strict. I do understand the strictness, but in real life scenari...
diff --git a/docs/requests.md b/docs/requests.md index f0ffc66..a72cb75 100644 --- a/docs/requests.md +++ b/docs/requests.md @@ -73,6 +73,8 @@ Cookies are exposed as a regular dictionary interface. For example: `request.cookies.get('mycookie')` +Cookies are ignored in case of an invalid cookie. (RFC2109) + #### B...
encode__starlette-92
[ { "changes": { "added_entities": [ "starlette/middleware/cors.py:CORSMiddleware.is_allowed_origin" ], "added_modules": null, "edited_entities": [ "starlette/middleware/cors.py:CORSMiddleware.__init__", "starlette/middleware/cors.py:CORSMiddleware.__call__", ...
encode/starlette
bdf99f1f6173dc4fbe9ea323bf3f86905d479ac1
Add `allow_origin_regex` to CORSMiddleware. It'd be helpful if `CORSMiddleware` supported an `allow_origin_regex`, so that users could do... ```python # Enforce a subdomain CORS policy app.add_middleware(CORSMiddleware, allow_origin_regex="(http|https)://*.example.com") ``` Or... ```python # Enforce an HTT...
diff --git a/starlette/middleware/cors.py b/starlette/middleware/cors.py index e299e74..99b0687 100644 --- a/starlette/middleware/cors.py +++ b/starlette/middleware/cors.py @@ -3,6 +3,7 @@ from starlette.responses import PlainTextResponse from starlette.types import ASGIApp, ASGIInstance, Scope import functools impo...
encode__typesystem-109
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "setup.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null ...
encode/typesystem
e806f2a4407615cd841f8938aafc656c9d7f4d0b
Password field What do you think about this kind of field? I was thinking about expanding String type as a first option: `typesystem.String(format='password')` or adding new kind of field `typesystem.Password` Both options will ensure to create "password" type of input with hidden value. The latter would ...
diff --git a/README.md b/README.md index 147ffa6..9c18e6b 100644 --- a/README.md +++ b/README.md @@ -39,16 +39,16 @@ Python 3.6+ $ pip3 install typesystem ``` -If you'd like you use the form rendering you'll also want to install `jinja2`. +If you'd like you use the form rendering using `jinja2`: ```shell -$ pip3...
encode__typesystem-111
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "typesystem/__init__.py" }, { "changes": { "added_entities": [ "typesystem/fields.py:Email.__init__" ], "added_modules": [ ...
encode/typesystem
851b60f3e08ae37b6548cc42e0e5407934f0a0d9
`Email` field adding new `Email` field: - Validates inputs to be valid emails - Displayed as `<input type="email">` in forms
diff --git a/typesystem/__init__.py b/typesystem/__init__.py index 165a090..1f72729 100644 --- a/typesystem/__init__.py +++ b/typesystem/__init__.py @@ -8,6 +8,7 @@ from typesystem.fields import ( Date, DateTime, Decimal, + Email, Field, Float, Integer, @@ -35,6 +36,7 @@ __all__ = [ ...
encode__typesystem-122
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "typesystem/json_schema.py:from_json_schema", "typesystem/json_schema.py:to_json_schema" ], "edited_modules": [ "typesystem/json_schema.py:from_json_schema", "typesyst...
encode/typesystem
2e9dc5671d72fde45a1684cc28434b56ee3eda06
References are broken for JSON schema (0.4.0) ### Checklist <!-- Please make sure you check all these items before submitting your bug report. --> - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull requests to fix it yet. ### Describe the bug Q...
diff --git a/docs/json_schema.md b/docs/json_schema.md index a66f287..7605a76 100644 --- a/docs/json_schema.md +++ b/docs/json_schema.md @@ -7,7 +7,7 @@ TypeSystem can convert Schema or Field instances to/from JSON Schema. All references should be of the style `{"$ref": "#/components/schemas/..."}`. Using h...
encode__typesystem-16
[ { "changes": { "added_entities": [ "typesystem/fields.py:Field.get_default_value" ], "added_modules": null, "edited_entities": [ "typesystem/fields.py:Object.validate_value" ], "edited_modules": [ "typesystem/fields.py:Object", "typesystem/fiel...
encode/typesystem
2fcc6daebd0193a02d135daee4564f0e219e95ed
Callable defaults Eg. ```python class Something(typesystem.Schema): created = typesystem.DateTime(default=datetime.datetime.now) ... ```
diff --git a/typesystem/fields.py b/typesystem/fields.py index 90716ef..266177c 100644 --- a/typesystem/fields.py +++ b/typesystem/fields.py @@ -63,6 +63,12 @@ class Field: def has_default(self) -> bool: return hasattr(self, "default") + def get_default_value(self) -> typing.Any: + default = g...
encode__typesystem-39
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "typesystem/forms.py:Form.render_field" ], "edited_modules": [ "typesystem/forms.py:Form" ] }, "file": "typesystem/forms.py" } ]
encode/typesystem
765ae6dfb810d853d5a91dcbc42bfc07750fdeee
Don't echo values back if `format="password"` Just needs a tweak in `Form.render_field` to set `value=""`, and a test case.
diff --git a/docs/schemas.md b/docs/schemas.md index 4064ca0..b75f524 100644 --- a/docs/schemas.md +++ b/docs/schemas.md @@ -105,7 +105,7 @@ print(dict(album)) Index lookup on schema instances returns serialized datatypes. ```python -print(type(album['release_date'])) +print(type(album.release_date)) # <class 'str...
encode__typesystem-44
[ { "changes": { "added_entities": [ "typesystem/fields.py:normalize_regex" ], "added_modules": [ "typesystem/fields.py:normalize_regex" ], "edited_entities": [ "typesystem/fields.py:String.__init__", "typesystem/fields.py:String.validate" ], ...
encode/typesystem
fc0800939c47c254f4fcc3f92dacd0b899cb8402
better error messages for regular expression validation string fields can have a `pattern` for validation, and since `re.search()` is used under the hood, the pattern can be either a regular expression as a string or a compiled regular expression (`re.compile(...)` is cached and can have flags). so far so good, but ...
diff --git a/docs/fields.md b/docs/fields.md index 0f694b0..5ce383b 100644 --- a/docs/fields.md +++ b/docs/fields.md @@ -58,7 +58,7 @@ For example: `username = typesystem.String(max_length=100)` * `trim_whitespace` - A boolean indicating if leading/trailing whitespace should be removed on validation. **Default: `True`...
encode__typesystem-65
[ { "changes": { "added_entities": [ "typesystem/base.py:Message.__hash__", "typesystem/base.py:BaseError.__hash__" ], "added_modules": null, "edited_entities": null, "edited_modules": [ "typesystem/base.py:Message", "typesystem/base.py:BaseError" ...
encode/typesystem
453c6fb05a43efe26ffeb476ca311a199a4fb704
ValidationError can't be shown with `traceback.print_exc()` in Python 3.6 Consider the following code: ```python import traceback import typesystem try: typesystem.Integer().validate("hello") except typesystem.ValidationError: traceback.print_exc() ``` Because of an issue with Python 3.6 discusse...
diff --git a/typesystem/base.py b/typesystem/base.py index bd15068..def115c 100644 --- a/typesystem/base.py +++ b/typesystem/base.py @@ -78,6 +78,10 @@ class Message: and self.end_position == other.end_position ) + def __hash__(self) -> int: + ident = (self.code, tuple(self.index)) + ...
encode__uvicorn-114
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "uvicorn/__init__.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "uvicorn/protocol...
encode/uvicorn
df2e97eaf6560dc4f7f3d321ab1f085553125ba0
RuntimeError in uvicorn for some requests `uvicorn==0.2.5` is throwing errors for some requests. ``` ERROR: Exception in ASGI application Traceback (most recent call last): File "/home/chillar/.virtualenvs/library/lib/python3.6/site-packages/uvicorn/protocols/http/httptools.py", line 196, in run_asgi resul...
diff --git a/uvicorn/__init__.py b/uvicorn/__init__.py index bfd395e..385fbc3 100644 --- a/uvicorn/__init__.py +++ b/uvicorn/__init__.py @@ -1,4 +1,4 @@ from uvicorn.main import main, run -__version__ = "0.2.5" +__version__ = "0.2.6" __all__ = ["main", "run"] diff --git a/uvicorn/protocols/http/h11.py b/uvicorn/pro...
encode__uvicorn-115
[ { "changes": { "added_entities": [ "uvicorn/protocols/http/h11.py:H11Protocol.handle_events", "uvicorn/protocols/http/h11.py:H11Protocol.on_response_complete" ], "added_modules": null, "edited_entities": [ "uvicorn/protocols/http/h11.py:H11Protocol.data_received",...
encode/uvicorn
870a8e24a6e9bda8be473e991b99ed7b172d5648
Pipelining support in `httptools` implementation We've temporarily dropped pipelining support from the `httptools` implementation, although everythings in place in order to support it. Need to: Keep track of queued request/response cycles if a new cycle starts before the existing one has finished. Callback to the pr...
diff --git a/README.md b/README.md index 07ab4ef..cb26c8d 100644 --- a/README.md +++ b/README.md @@ -48,7 +48,7 @@ class App(): }) await send({ 'type': 'http.response.body', - 'body': b'Hello, world!', + 'body': 'Hello, world!', }) ``` diff --git a/uvicor...
encode__uvicorn-1263
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "uvicorn/main.py" }, { "changes": { "added_entities": [ "uvicorn/protocols/http/httptools_impl.py:HttpToolsProtocol.on_message_begin" ...
encode/uvicorn
521b83ed0edd46eabba61a9ba4f4f3c0a58fce58
httptools.parser.errors.HttpParserInvalidURLError on fragmented first line of the HTTP request. ### Checklist <!-- Please make sure you check all these items before submitting your bug report. --> - [x] The bug is reproducible against the latest release and/or `master`. - [x] There are no similar issues or pull ...
diff --git a/docs/deployment.md b/docs/deployment.md index 60d1af7..2224f98 100644 --- a/docs/deployment.md +++ b/docs/deployment.md @@ -39,12 +39,15 @@ Options: of using the current working directory. --reload-include TEXT Set glob patterns to include while watching ...
encode__uvicorn-1267
[ { "changes": { "added_entities": [ "uvicorn/config.py:_normalize_dirs" ], "added_modules": [ "uvicorn/config.py:_normalize_dirs" ], "edited_entities": [ "uvicorn/config.py:resolve_reload_patterns", "uvicorn/config.py:Config.__init__" ], "...
encode/uvicorn
81802ceee2a2c62f0b10f78e2b6366b47bc76f72
Passing `reload_dirs` as a string instead of a list @Roang-zero1 @euri10 @Kludex this PR introduced a regression and deleted the test for it (`test_reload_dir_is_set()`). Previously you could pass `reload_dirs` a string instead of a list. Now: ```python Config('myapp:app', reload=True, reload_dirs='src') ``` prints...
diff --git a/uvicorn/config.py b/uvicorn/config.py index 683a830..59c94b4 100644 --- a/uvicorn/config.py +++ b/uvicorn/config.py @@ -180,7 +180,15 @@ def resolve_reload_patterns( directories = list(set(directories).difference(set(children))) - return (list(set(patterns)), directories) + return list(set(p...
encode__uvicorn-1329
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": null, "edited_modules": null }, "file": "uvicorn/main.py" }, { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "uvicorn/middleware/w...
encode/uvicorn
521b83ed0edd46eabba61a9ba4f4f3c0a58fce58
WSGI interface: reading request body is O(n^2) wrt payload size Hello! This is related to https://github.com/encode/uvicorn/issues/371 and potentially a duplicate/complementary information, feel free to close if it is deemed to be the case. I was fooling around with different WSGI and ASGI servers while hacking on ...
diff --git a/docs/deployment.md b/docs/deployment.md index 60d1af7..2224f98 100644 --- a/docs/deployment.md +++ b/docs/deployment.md @@ -39,12 +39,15 @@ Options: of using the current working directory. --reload-include TEXT Set glob patterns to include while watching ...
encode__uvicorn-1600
[ { "changes": { "added_entities": [ "uvicorn/server.py:Server._serve", "uvicorn/server.py:Server.capture_signals" ], "added_modules": null, "edited_entities": [ "uvicorn/server.py:Server.__init__", "uvicorn/server.py:Server.serve", "uvicorn/server.p...
encode/uvicorn
f73b8beeb1499ca5fcec3067cf89dad5326a0984
uvicorn eats SIGINTs, does not propagate exceptions The following snippet cannot be killed with a SIGINT (ctrl+c): ```python import asyncio from starlette.applications import Starlette from uvicorn import Config, Server async def web_ui(): await Server(Config(Starlette())).serve() async def task(): ...
diff --git a/uvicorn/server.py b/uvicorn/server.py index c7645f3..bfce1b1 100644 --- a/uvicorn/server.py +++ b/uvicorn/server.py @@ -1,6 +1,7 @@ from __future__ import annotations import asyncio +import contextlib import logging import os import platform @@ -11,7 +12,7 @@ import threading import time from emai...
encode__uvicorn-165
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "uvicorn/protocols/http/h11_impl.py:RequestResponseCycle.send" ], "edited_modules": [ "uvicorn/protocols/http/h11_impl.py:RequestResponseCycle" ] }, "file": "uvicorn/pro...
encode/uvicorn
5e2da780ae775dbe963066554d3bb230a41c7f9c
Drop response body on HEAD requests
diff --git a/uvicorn/protocols/http/h11_impl.py b/uvicorn/protocols/http/h11_impl.py index ab9cdaf..598b5e0 100644 --- a/uvicorn/protocols/http/h11_impl.py +++ b/uvicorn/protocols/http/h11_impl.py @@ -465,7 +465,10 @@ class RequestResponseCycle: more_body = message.get("more_body", False) # ...
encode__uvicorn-166
[ { "changes": { "added_entities": null, "added_modules": null, "edited_entities": [ "uvicorn/protocols/http/h11_impl.py:RequestResponseCycle.__init__", "uvicorn/protocols/http/h11_impl.py:RequestResponseCycle.send", "uvicorn/protocols/http/h11_impl.py:RequestResponseCycl...
encode/uvicorn
fa5185f348aa3be814f986abdec967d9de4a57f3
Support `Expect: 100-Continue` * Deal with `Expect: 100-Continue` headers gracefully, by only sending if the request body is read prior to sending the response headers.
diff --git a/uvicorn/protocols/http/h11_impl.py b/uvicorn/protocols/http/h11_impl.py index 598b5e0..23c021d 100644 --- a/uvicorn/protocols/http/h11_impl.py +++ b/uvicorn/protocols/http/h11_impl.py @@ -361,6 +361,7 @@ class RequestResponseCycle: # Connection state self.disconnected = False sel...