Migrate legacy setup.py to pyproject.toml (#108)

* Migrate legacy setup.py to pyproject.toml

* Prune setup.cfg

Moves the pytest config from setup.cfg to pyproject.toml

Removes some sections that didn't seem necessary or duplicate
metadata in pyproject.toml (resolves "Usage of dash-separated
'description-file' will not be supported in future" build warning)

Keeps only the flake8 config that doesn't support pyproject.toml: PyCQA/flake8#234

* Add just the detoxify package to the build

* Move sentencepiece to main deps

.github/workflows/ci-testing.yml

@@ -55,7 +55,7 @@ jobs:
 
     - name: Install dependencies
       run: |
-        pip install --requirement requirements.txt --upgrade --quiet --find-links https://download.pytorch.org/whl/cpu/torch_stable.html
+        pip install -e '.[dev]'
         pip install --requirement tests/requirements.txt --quiet
         python --version
         pip --version

README.md

@@ -176,12 +176,12 @@ python3 -m venv toxic-env
 source toxic-env/bin/activate
 
 # install project
-
 pip install -e detoxify
-cd detoxify
 
-# for training
-pip install -r requirements.txt
+# or for training
+pip install -e 'detoxify[dev]'
+
+cd detoxify
 
  ```
 

pyproject.toml

@@ -1,7 +1,38 @@
-[build-system]
-requires = [
-    "setuptools",
-    "wheel",
+[project]
+name = "detoxify"
+version = "0.5.2"
+description = "A python library for detecting toxic comments"
+readme = "README.md"
+authors = [
+    {name = "Unitary", email = "[email protected]"},
+]
+classifiers = [
+    "License :: OSI Approved :: Apache Software License",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+]
+requires-python = ">=3.9,<3.12"
+dependencies = [
+    "sentencepiece >= 0.1.94",
+    "torch < 2.2",
+    "transformers >= 3",
+]
+
+[tool.setuptools]
+packages = ["detoxify"]
+
+[project.urls]
+Homepage = "https://github.com/unitaryai/detoxify"
+
+[project.optional-dependencies]
+dev = [
+    "datasets >= 1.0.2",
+    "pandas >= 1.1.2",
+    "pytest",
+    "pytorch-lightning<2.0.0,>1.5.0",
+    "scikit-learn >= 0.23.2",
+    "tqdm",
+    "pre-commit",
 ]
 
 [tool.black]
@@ -16,3 +47,25 @@ known_first_party = [
 skip_glob = []
 profile = "black"
 line_length = 120
+
+[tool.pytest.ini_options]
+norecursedirs = [
+    ".git",
+    "dist",
+    "build",
+]
+addopts = [
+    "--strict-markers",
+    "--doctest-modules",
+    "--durations=0",
+    "--color=yes",
+]
+filterwarnings = [
+    "ignore:.*deprecated alias.*:DeprecationWarning:tensorboard.*tensorflow_stub*:",
+    "ignore:.*deprecated alias.*:DeprecationWarning:tensorboard.*tensor_util*:",
+    "ignore:.*deprecated alias.*:DeprecationWarning:pyarrow.*pandas_compat*:",
+]
+
+[build-system]
+requires = ["setuptools"]
+build-backend = "setuptools.build_meta"

setup.cfg

@@ -1,24 +1,3 @@
-[tool:pytest]
-norecursedirs =
-    .git
-    dist
-    build
-addopts =
-    --strict-markers
-    --doctest-modules
-    --durations=0
-    --color=yes
-# Silence numpy builtin type alias warnigns in third party libs
-# https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
-filterwarnings = ignore:.*deprecated alias.*:DeprecationWarning:tensorboard.*tensorflow_stub*:
-                 ignore:.*deprecated alias.*:DeprecationWarning:tensorboard.*tensor_util*:
-                 ignore:.*deprecated alias.*:DeprecationWarning:pyarrow.*pandas_compat*:
-
-[coverage:report]
-exclude_lines =
-    pragma: no-cover
-    pass
-
 [flake8]
 max-line-length = 120
 exclude = .tox,*.egg,build,temp
@@ -41,16 +20,3 @@ ignore =
     F841,
     # Invalid escape sequence 'x'
     W605
-
-# setup.cfg or tox.ini
-[check-manifest]
-ignore =
-    *.yml
-    .github
-    .github/*
-
-[metadata]
-license_file = LICENSE
-description-file = README.md
-# long_description = file:README.md
-# long_description_content_type = text/markdown