| [build-system] |
| requires = ["maturin==1.14.1"] |
| build-backend = "maturin" |
|
|
| [project] |
| name = "nedotokenizer" |
| version = "0.2.0" |
| description = "Byte-exact, morphology-aware tokenizer for Turkish and mixed Turkish-code text" |
| readme = "README.md" |
| requires-python = ">=3.11" |
| license = "Apache-2.0" |
| license-files = ["LICENSE", "THIRD_PARTY_LICENSES/ZEMBEREK_NOTICE.md"] |
| authors = [ |
| { name = "Nedim Mutlu Sezer" }, |
| { name = "Yağız Ekrem Dalar" }, |
| { name = "Feyzi Arda Salihoğlu" }, |
| { name = "Ömer Faruk Aksoy" }, |
| { name = "Ahmet Rıfat Öztürk" }, |
| ] |
| keywords = ["tokenizer", "turkish", "morphology", "nlp", "rust"] |
| classifiers = [ |
| "Development Status :: 3 - Alpha", |
| "Programming Language :: Python :: 3", |
| "Programming Language :: Rust", |
| "Operating System :: POSIX :: Linux", |
| "Topic :: Text Processing :: Linguistic", |
| ] |
|
|
| [project.urls] |
| Homepage = "https://github.com/ethosoftai/NedoTokenizer" |
| Repository = "https://github.com/ethosoftai/NedoTokenizer" |
| Issues = "https://github.com/ethosoftai/NedoTokenizer/issues" |
|
|
| [tool.maturin] |
| manifest-path = "crates/nedo-python/Cargo.toml" |
| python-source = "python" |
| module-name = "nedotokenizer._native" |
| bindings = "pyo3" |
| compatibility = "manylinux_2_17" |
| strip = true |
| include = [ |
| { path = "assets/native/MANIFEST.sha256", format = "sdist" }, |
| { path = "assets/native/model-compressed", format = "sdist" }, |
| { path = "assets/native/nedo-morph-v1.bin", format = "sdist" }, |
| { path = "assets/surface-vocab.bin", format = "wheel" }, |
| { path = "assets/surface-vocab.bin", format = "sdist" }, |
| ] |
|
|
