Merge pull request #14 from KapiWow/1-windows-support

feat: windows support

.github/workflows/release_python.yml

@@ -91,48 +91,47 @@ jobs:
             cd bindings/extractous-python
             pytest -s
 
-#
-#  windows:
-#    runs-on: ${{ matrix.platform.runner }}
-#    strategy:
-#      matrix:
-#        platform:
-#          - runner: windows-latest
-#            target: x64
-#    steps:
-#      - uses: actions/checkout@v4
-#      - uses: actions/setup-python@v5
-#        with:
-#          python-version: '3.8'
-#          architecture: ${{ matrix.platform.target }}
-#      - uses: graalvm/[email protected]
-#        with:
-#          java-version: '22'
-#          distribution: 'graalvm-community'
-#          set-java-home: 'true'
-#      - name: Build wheels
-#        uses: PyO3/maturin-action@v1
-#        with:
-#          target: ${{ matrix.platform.target }}
-#          working-directory: "bindings/extractous-python"
-#          args: --release --out dist --find-interpreter
-#          sccache: 'true'
-#      - name: Upload wheels
-#        uses: actions/upload-artifact@v4
-#        with:
-#          name: wheels-windows-${{ matrix.platform.target }}
-#          path: bindings/extractous-python/dist
-#      - name: pytest
-#        if: ${{ !startsWith(matrix.platform.target, 'aarch64') }}
-#        shell: bash
-#        run: |
-#          set -e
-#          python3 -m venv .venv
-#          source .venv/Scripts/activate
-#          pip install extractous --find-links bindings/extractous-python/dist --force-reinstall
-#          pip install pytest scikit-learn
-#          cd bindings/extractous-python
-#          pytest -s
+
+  windows:
+    runs-on: ${{ matrix.platform.runner }}
+    strategy:
+      matrix:
+        platform:
+          - runner: windows-latest
+            target: x64
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: '3.8'
+          architecture: ${{ matrix.platform.target }}
+      - uses: graalvm/[email protected]
+        with:
+          java-version: '22'
+          distribution: 'graalvm-community'
+          set-java-home: 'true'
+      - name: Build wheels
+        uses: PyO3/maturin-action@v1
+        with:
+          target: ${{ matrix.platform.target }}
+          working-directory: "bindings/extractous-python"
+          args: --release --out dist --find-interpreter
+          sccache: 'true'
+      - name: Upload wheels
+        uses: actions/upload-artifact@v4
+        with:
+          name: wheels-windows-${{ matrix.platform.target }}
+          path: bindings/extractous-python/dist
+      - name: pytest
+        if: ${{ !startsWith(matrix.platform.target, 'aarch64') }}
+        shell: cmd
+        run: |
+          python -m venv .venv
+          .venv\Scripts\activate.bat
+          pip install extractous --find-links bindings/extractous-python/dist --force-reinstall
+          pip install pytest scikit-learn
+          cd bindings\extractous-python
+          pytest -s
 
   macos:
     runs-on: ${{ matrix.platform.runner }}
@@ -205,7 +204,7 @@ jobs:
   publish-to-testpypi:
     name: Publish to TestPyPI
     # publish to TestPyPI on any push
-    needs: [linux, macos, sdist]
+    needs: [linux, macos, windows, sdist]
     runs-on: ubuntu-latest
     environment:
       name: testpypi
@@ -230,7 +229,7 @@ jobs:
     name: Publish to PyPI
     # only publish to PyPI on tag pushes
     if: "startsWith(github.ref, 'refs/tags/')"
-    needs: [linux, macos, sdist]
+    needs: [linux, macos, windows, sdist]
     runs-on: ubuntu-latest
     environment:
       name: pypi
@@ -247,4 +246,4 @@ jobs:
         uses: pypa/gh-action-pypi-publish@release/v1
         with:
           skip-existing: true
-          packages-dir: bindings/extractous-python/dist
+          packages-dir: bindings/extractous-python/dist

bindings/extractous-python/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "extractous"
-version = '0.1.4'
+version = '0.1.5'
 classifiers = [
   "Programming Language :: Rust",
   "Programming Language :: Python :: Implementation :: CPython",
@@ -65,6 +65,5 @@ rustc-args = ["-C", "link-arg=-Wl,-rpath,$ORIGIN"]
 include = [ 
   {path = "**/*.so", format = ["wheel"]},
   {path = "**/*.dylib", format = ["wheel"]},
-  {path = "**/*.dll", format = ["wheel"]},
-  {path = "**/*.lib", format = ["wheel"]}
+  {path = "**/*.dll", format = ["wheel"]}
 ]

bindings/extractous-python/python/extractous/__init__.py

@@ -1,3 +1,12 @@
+import os
+from platform import system
+
+# On Windows there is no equivalent way of setting RPATH
+# This adds the current directory to PATH so that the graalvm libs will be found 
+if system() == "Windows":
+    libpath = os.path.dirname(__file__)
+    os.environ["PATH"] = libpath + os.pathsep + os.environ["PATH"]
+
 from ._extractous import *
 
 __doc__ = _extractous.__doc__

bindings/extractous-python/tests/test_extract_file_to_string.py

@@ -24,7 +24,7 @@ def test_extract_file_to_string(file_name, target_dist):
     expected_result_filepath = f"../../test_files/expected_result/{file_name}.txt"
     extractor = Extractor()
     result = extractor.extract_file_to_string(original_filepath)
-    with open(expected_result_filepath, "r") as file:
+    with open(expected_result_filepath, "r",  encoding="utf8") as file:
         expected = file.read()
 
     assert cosine_similarity(result, expected) > target_dist, \

extractous-core/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "extractous"
-version = "0.1.4"
+version = "0.1.5"
 edition = "2021"
 
 description = """

extractous-core/build.rs

@@ -1,7 +1,6 @@
 use std::env;
 use std::fs;
 use std::io;
-use std::io::Write;
 use std::path::{Path, PathBuf};
 use std::process::Command;
 
@@ -34,8 +33,11 @@ fn main() {
     // Try to find already built libs
     match find_already_built_libs(&out_dir) {
         Some(libs_dir) => {
-            // If the libs are already built, copy them to the output directory
-            copy_build_artifacts(&libs_dir, vec![&libs_out_dir], false);
+            // ignore if libs_dir/.. is the same as out_dir
+            if out_dir.join("libs") != libs_dir {
+                // If the libs are already built, copy them to the output directory
+                copy_build_artifacts(&libs_dir, vec![&libs_out_dir], false);
+            }
         }
         None => {
             // Launch the gradle build
@@ -52,7 +54,12 @@ fn main() {
     println!("cargo:rustc-link-search={}", libs_out_dir.display());
 
     // Tell cargo to tell rustc to link the `tika_native` shared library.
-    println!("cargo:rustc-link-lib=dylib=tika_native");
+    let lib_tika_name = if cfg!(target_os = "windows") {
+        "libtika_native"
+    } else {
+        "tika_native"
+    };
+    println!("cargo:rustc-link-lib=dylib={}", lib_tika_name);
 }
 
 /// Searches for directories two levels up from `out_dir` and checks if any of them
@@ -121,6 +128,7 @@ fn gradle_build(
     // Launch the gradle build
     Command::new(gradlew)
         .current_dir(&tika_native_dir)
+        .arg("--no-daemon")
         .arg("nativeCompile")
         .env("JAVA_HOME", graalvm_home)
         .status()
@@ -290,7 +298,7 @@ pub fn install_graalvm_ce(install_dir: &PathBuf) -> PathBuf {
             // in case of a download error
             let mut buffer: Vec<u8> = vec![];
             io::copy(&mut response.bytes()
-                .expect(&format!("Failed to download GraalVM JDK from {}", base_url))
+                .unwrap_or_else(|_| panic!("Failed to download GraalVM JDK from {}", base_url))
                 .as_ref(), &mut buffer
             ).unwrap();
             //let mut out = fs::File::create(&archive_path).unwrap();
@@ -335,4 +343,4 @@ pub fn install_graalvm_ce(install_dir: &PathBuf) -> PathBuf {
     }
 
     install_dir.join(main_dir)
-}
+}