feat: add windows support and refactor setup

.github/workflows/publish.yml

@@ -51,7 +51,7 @@ jobs:
     uses: ./.github/workflows/reusable-build.yml
     with:
       CIBW_SKIP: "pp* cp36-* cp37-*"
-      CIBW_BUILD: "cp*-macosx* cp*-manylinux*"
+      CIBW_BUILD: "cp*-macosx* cp*-manylinux* cp*-win*"
       VERSION: ${{ github.ref_name }}
     secrets: inherit
 

.github/workflows/reusable-build.yml

@@ -17,7 +17,7 @@ jobs:
   build-wheels:
     strategy:
       matrix:
-        os: [ubuntu-latest, macos-latest]
+        os: [ubuntu-latest, macos-latest, windows-latest]
       fail-fast: false
     name: Build wheels on ${{ matrix.os }}
     runs-on: ${{ matrix.os }}

.github/workflows/test.yml

@@ -41,7 +41,7 @@ jobs:
         with:
           python-version: '3.12'
           cache: 'pip' # caching pip dependencies
-      - name: Install dependencies
+      - name: Install lib from source and dependencies
         run: |
           python -m pip install -e .[test]
       - name: Run tests
@@ -52,15 +52,15 @@ jobs:
     uses: ./.github/workflows/reusable-build.yml
     with:
       CIBW_SKIP: "pp* cp36-* cp37-*"
-      CIBW_BUILD: "cp*-macosx* cp*-manylinux*"
+      CIBW_BUILD: "cp*-macosx* cp*-manylinux* cp*-win*"
     secrets: inherit
 
   full-tests-python:
     needs: [fast-tests-python, external-build-workflow]
     strategy:
       matrix:
         python-version: ['3.8', '3.9', '3.10', '3.11', '3.12']
-        os: [ubuntu-latest, macos-latest]
+        os: [ubuntu-latest, macos-latest, windows-latest]
       fail-fast: false
     name: Test wheel on ${{ matrix.os }} and Python ${{ matrix.python-version }}
     runs-on: ${{ matrix.os }}
@@ -72,16 +72,21 @@ jobs:
           path: dist
       - name: Show dist files
         run: ls -lah ./dist
+        shell: bash
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v4
         with:
           python-version: ${{ matrix.python-version }}
           cache: 'pip' # caching pip dependencies
-      - name: Install dependencies
+      - name: Remove sdist package to force install wheel later
+        run: |
+          rm -rf ./dist/*.tar.gz
+        shell: bash
+      - name: Install lib and dependencies
         run: |
           # force install package from local dist directory
+          pip uninstall -y codebleu || true
           # TODO: check the sdist package is not installed
-          rm -rf ./dist/*.tar.gz
           pip install --upgrade --no-deps --no-index --find-links=./dist codebleu
           # install dependencies for the package and tests
           pip install .[test]

.gitignore

@@ -4,6 +4,15 @@ codebleu/parser/*.so
 codebleu/parser/*.dylib
 codebleu/parser/*.dll
 
+/codebleu/parser/tree-sitter-c-sharp/
+/codebleu/parser/tree-sitter-go/
+/codebleu/parser/tree-sitter-java/
+/codebleu/parser/tree-sitter-javascript/
+/codebleu/parser/tree-sitter-php/
+/codebleu/parser/tree-sitter-python/
+/codebleu/parser/tree-sitter-ruby/
+/tree_sitter/
+codebleu/*.so
 
 
 # Byte-compiled / optimized / DLL files
@@ -166,10 +175,3 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
-/codebleu/parser/tree-sitter-c-sharp/
-/codebleu/parser/tree-sitter-go/
-/codebleu/parser/tree-sitter-java/
-/codebleu/parser/tree-sitter-javascript/
-/codebleu/parser/tree-sitter-php/
-/codebleu/parser/tree-sitter-python/
-/codebleu/parser/tree-sitter-ruby/

README.md

@@ -5,7 +5,7 @@
 [![PyPI version](https://badge.fury.io/py/codebleu.svg)](https://badge.fury.io/py/codebleu)
 
 
-This repository contains an unofficial `CodeBLEU` implementation that supports Linux and MacOS. It is available through `PyPI` and the `evaluate` library.
+This repository contains an unofficial `CodeBLEU` implementation that supports `Linux`, `MacOS` and `Windows`. It is available through `PyPI` and the `evaluate` library.
 
 The code is based on the original [CodeXGLUE/CodeBLEU](https://github.com/microsoft/CodeXGLUE/tree/main/Code-Code/code-to-code-trans/evaluator/CodeBLEU) and updated version by [XLCoST/CodeBLEU](https://github.com/reddy-lab-code-research/XLCoST/tree/main/code/translation/evaluator/CodeBLEU).  It has been refactored, tested, built for macOS, and multiple improvements have been made to enhance usability
 
@@ -28,7 +28,7 @@ The metric has shown higher correlation with human evaluation than `BLEU` and `a
 ## Installation
 
 As this library require `so` file compilation it is platform dependent.  
-Currently available for `Linux` (manylinux) and `MacOS` with Python 3.8+.
+Currently available for `Linux` (manylinux), `MacOS` and `Windows` with Python 3.8+.
 
 The metrics is available as [pip package](https://pypi.org/project/codebleu/) and can be installed as indicated above:
 ```bash

codebleu/codebleu.py

@@ -17,7 +17,7 @@ def calc_codebleu(
     weights: Tuple[float, float, float, float] = (0.25, 0.25, 0.25, 0.25),
     tokenizer: Optional[Callable] = None,
     keywords_dir: Path = PACKAGE_DIR / "keywords",
-    lang_so_file: Path = PACKAGE_DIR / "parser" / "my-languages.so",
+    lang_so_file: Path = PACKAGE_DIR / "my-languages.so",
 ) -> Dict[str, float]:
     """Calculate CodeBLEU score
 
@@ -69,10 +69,10 @@ def make_weights(reference_tokens, key_word_list):
     weighted_ngram_match_score = weighted_ngram_match.corpus_bleu(tokenized_refs_with_weights, tokenized_hyps)
 
     # calculate syntax match
-    syntax_match_score = syntax_match.corpus_syntax_match(references, hypothesis, lang, lang_so_file)
+    syntax_match_score = syntax_match.corpus_syntax_match(references, hypothesis, lang, str(lang_so_file))
 
     # calculate dataflow match
-    dataflow_match_score = dataflow_match.corpus_dataflow_match(references, hypothesis, lang, lang_so_file)
+    dataflow_match_score = dataflow_match.corpus_dataflow_match(references, hypothesis, lang, str(lang_so_file))
 
     alpha, beta, gamma, theta = weights
     code_bleu_score = (

pyproject.toml

@@ -1,17 +1,17 @@
 [build-system]
-requires = ["setuptools>=61.0.0", "wheel", "tree-sitter>=0.20.0,<1.0.0"]
+requires = ["setuptools>=61.0.0", "wheel", "tree-sitter>=0.20.0,<1.0.0", "requests>=2.0.0,<3.0.0"]
 build-backend = "setuptools.build_meta"
 
 
 [project]
 name = "codebleu"
-description = "Unofficial CodeBLEU implementation that supports Linux and MacOS available on PyPI."
+description = "Unofficial CodeBLEU implementation that supports Linux, MacOS and Windows available on PyPI."
 readme = "README.md"
 license = {text = "MIT License"}
 authors = [
   {name = "Konstantin Chernyshev", email = "[email protected]"},
 ]
-keywords = ["codebleu", "code", "bleu", "nlp", "natural language processing", "programming", "evaluate", "evaluation", "code generation", "matrics"]
+keywords = ["codebleu", "code", "bleu", "nlp", "natural language processing", "programming", "evaluate", "evaluation", "code generation", "metrics"]
 dynamic = ["version"]
 
 requires-python = ">=3.8"
@@ -77,7 +77,7 @@ warn_redundant_casts = true
 warn_unused_ignores = true
 warn_unreachable = true
 allow_untyped_decorators = true
-exclude = ["codebleu/parser/tree-sitter", "codebleu/parser/tree-sitter/python"]
+exclude = ["codebleu/parser/tree-sitter", "codebleu/parser/tree-sitter/python", "tree_sitter"]
 
 [tool.pytest.ini_options]
 minversion = "6.0"
@@ -86,7 +86,7 @@ python_files = "test_*.py"
 addopts = "--cov=codebleu/ --cov-report term-missing"
 
 [tool.coverage.run]
-omit = ["tests/*", "codebleu/parser/tree-sitter/*"]
+omit = ["tests/*", "codebleu/parser/tree-sitter/*", "tree_sitter"]
 
 
 [tool.isort]
@@ -95,7 +95,7 @@ src_paths = ["codebleu", "tests"]
 known_first_party = ["codebleu", "tests"]
 line_length = 120
 combine_as_imports = true
-skip = ["build", "dist", ".venv", ".eggs", ".mypy_cache", ".pytest_cache", ".git", ".tox", ".nox", "codebleu/parser"]
+skip = ["build", "dist", ".venv", ".eggs", ".mypy_cache", ".pytest_cache", ".git", ".tox", ".nox", "codebleu/parser", "tree_sitter"]
 
 [tool.black]
 line_length=120

setup.py

@@ -1,24 +1,87 @@
-import subprocess
+from __future__ import annotations
+
+import io
+import shutil
+import zipfile
 from pathlib import Path
 
+import requests
 from setuptools import setup
 from setuptools.dist import Distribution
 
+from tree_sitter import Language
+
 ROOT = Path(__file__).parent
 
 
-subprocess.run(
-    ["bash", "build.sh"],
-    cwd=ROOT / "codebleu" / "parser",
-    check=True,
+tree_sitter_languages = {
+    "go": "https://github.com/tree-sitter/tree-sitter-go/archive/refs/tags/v0.20.0.zip",
+    "javascript": "https://github.com/tree-sitter/tree-sitter-javascript/archive/refs/tags/v0.20.1.zip",
+    "python": "https://github.com/tree-sitter/tree-sitter-python/archive/refs/tags/v0.20.4.zip",
+    "ruby": "https://github.com/tree-sitter/tree-sitter-ruby/archive/refs/tags/v0.19.0.zip",
+    "php": "https://github.com/tree-sitter/tree-sitter-php/archive/refs/tags/v0.19.0.zip",
+    "java": "https://github.com/tree-sitter/tree-sitter-java/archive/refs/tags/v0.20.2.zip",
+    "c-sharp": "https://github.com/tree-sitter/tree-sitter-c-sharp/archive/refs/tags/v0.20.0.zip",
+    "c": "https://github.com/tree-sitter/tree-sitter-c/archive/refs/tags/v0.20.6.zip",
+    "cpp": "https://github.com/tree-sitter/tree-sitter-cpp/archive/refs/tags/v0.20.3.zip",
+}
+
+
+def download_tree_sitter_languages(languages: dict[str, str], languages_folder: Path) -> list[str]:
+    if languages_folder.exists():
+        shutil.rmtree(languages_folder)
+    languages_folder.mkdir(parents=True)
+
+    extracted_folders: list[str] = []
+    for lang, url in languages.items():
+        # Download the ZIP file
+        response = requests.get(url)
+        response.raise_for_status()
+
+        # Extract the ZIP file
+        with zipfile.ZipFile(io.BytesIO(response.content)) as zip_f:
+            zip_f.extractall(languages_folder)
+            extracted_folders.append(zip_f.namelist()[0])  # get the name of the extracted folder
+
+    return extracted_folders
+
+
+def build_tree_sitter_languages(languages: dict[str, str], languages_folder: Path, target_lib_file: Path) -> str:
+    extracted_folders = download_tree_sitter_languages(languages, languages_folder)
+
+    Language.build_library(
+        str(target_lib_file),
+        [str(languages_folder / lang_folder) for lang_folder in extracted_folders],
+    )
+
+    return str(target_lib_file)
+
+
+build_tree_sitter_languages(
+    tree_sitter_languages,
+    ROOT / "tree_sitter",
+    ROOT / "codebleu" / "my-languages.so",
 )
 
 
+# tree_sitter_extension = Extension(
+#     'codebleu.tree_sitter',
+#     sources=[],
+#     include_dirs=[],
+#     libraries=[],
+#     extra_objects=[
+#
+#     ],
+# )
+
+
 class PlatformSpecificDistribution(Distribution):
     """Distribution which always forces a binary package with platform name"""
 
     def has_ext_modules(self):
         return True
 
 
-setup(distclass=PlatformSpecificDistribution)
+setup(
+    distclass=PlatformSpecificDistribution,
+)

tests/test_codebleu.py

@@ -65,17 +65,17 @@ def test_error_when_input_length_mismatch() -> None:
             ["public static int Sign ( double d ) { return ( float ) ( ( d == 0 ) ? 0 : ( c < 0.0 ) ? - 1 : 1) ; }"],
             ["public static int Sign ( double d ) { return ( int ) ( ( d == 0 ) ? 0 : ( d < 0 ) ? - 1 : 1) ; }"],
             0.7846,
-            11/19,  # In example, it is 13/21, but with new version of tree-sitter it is 11/19
-            2/3,
+            11 / 19,  # In example, it is 13/21, but with new version of tree-sitter it is 11/19
+            2 / 3,
             0.7019,  # Should be 0.7238 if AST=13/21 in the paper, however at the moment tee-sitter AST is 11/19
         ),
         # https://arxiv.org/pdf/2009.10297.pdf "3.4 Two Examples" at the page 4
         (
             ["public static int Sign ( double d ) { return ( float ) ( ( d == 0 ) ? 0 : ( c < 0.0 ) ? - 1 : 1) ;"],
             ["public static int Sign ( double d ) { return ( int ) ( ( d == 0 ) ? 0 : ( d < 0 ) ? - 1 : 1) ; }"],
             0.7543,
-            11/19,  # In example, it is 13/21, but with new version of tree-sitter it is 11/19
-            2/3,
+            11 / 19,  # In example, it is 13/21, but with new version of tree-sitter it is 11/19
+            2 / 3,
             0.6873,  # Should be 0.6973 if AST=13/21 in the paper, however at the moment tee-sitter AST is 11/19
         ),
         # https://arxiv.org/pdf/2009.10297.pdf "3.4 Two Examples" at the page 4