chore: updated windows support

.github/workflows/ci.yaml

@@ -38,13 +38,11 @@ jobs:
 
       - name: Execute Pre-Commit
         run: uv run pre-commit run --show-diff-on-failure --color=always --all-files
-
   test:
     strategy:
       matrix:
-        os: [ubuntu-latest, windows-latest, macos-latest]
+        os: [ ubuntu-latest, windows-latest, macos-latest ]
     runs-on: ${{ matrix.os }}
-    timeout-minutes: 5
     steps:
       - name: Checkout
         uses: actions/checkout@v4
@@ -54,23 +52,33 @@ jobs:
         with:
           enable-cache: true
 
-      - name: Set up Python
+      - name: Install Python
         uses: actions/setup-python@v5
         with:
           python-version-file: "pyproject.toml"
 
-      - name: Install System Dependencies
+      - name: Install Dependencies
+        run: uv sync --all-extras --dev
+
+      - name: Install System Dependencies (Windows)
+        if: runner.os == 'Windows'
+        run: |
+          choco install -y tesseract pandoc
+          echo "C:\Program Files\Tesseract-OCR" >> $env:PATH
+          echo "C:\Program Files\Pandoc" >> $env:PATH
+
+      - name: Install System Dependencies (Linux)
         if: runner.os == 'Linux'
         run: sudo apt-get install -y tesseract-ocr pandoc
-      - name: Install System Dependencies
-        if: runner.os == 'Windows'
-        run: choco install -y tesseract pandoc
-      - name: Install System Dependencies
+
+      - name: Install System Dependencies (macOS)
         if: runner.os == 'macOS'
         run: brew install tesseract pandoc
 
-      - name: Install Dependencies
-        run: uv sync --all-extras --dev
+      - name: Verify Installed Dependencies (Windows)
+        run: |
+          tesseract --version
+          pandoc --version
 
-      - name: Test
+      - name: Run Tests
         run: uv run pytest tests

kreuzberg/_extractors.py

@@ -38,7 +38,8 @@ async def convert_pdf_to_images(file_path: Path) -> list[Image]:
         A list of Pillow Images.
     """
     try:
-        pdf = await run_sync(pypdfium2.PdfDocument, str(file_path))
+        resolved_path = str(await AsyncPath(file_path).resolve())
+        pdf = await run_sync(pypdfium2.PdfDocument, resolved_path)
         return [page.render(scale=2.0).to_pil() for page in pdf]
     except pypdfium2.PdfiumError as e:
         raise ParsingError(
@@ -73,7 +74,8 @@ async def extract_pdf_with_pdfium2(file_path: Path) -> str:
         The extracted text.
     """
     try:
-        document = await run_sync(pypdfium2.PdfDocument, file_path)
+        resolved_path = str(await AsyncPath(file_path).resolve())
+        document = await run_sync(pypdfium2.PdfDocument, resolved_path)
         text = "\n".join(page.get_textpage().get_text_bounded() for page in document)
         return normalize_spaces(text)
     except pypdfium2.PdfiumError as e:
@@ -122,7 +124,8 @@ async def extract_file_with_pandoc(file_path: Path | str, mime_type: str) -> str
     Returns:
         The extracted text.
     """
-    result = await process_file(file_path, mime_type=mime_type)
+    resolved_path = str(await AsyncPath(file_path).resolve())
+    result = await process_file(resolved_path, mime_type=mime_type)
     return normalize_spaces(result.content)
 
 
@@ -215,7 +218,7 @@ async def extract_xlsx_file(file_path_or_contents: Path | bytes) -> str:
                 xlsx_file.flush()
                 xlsx_path = xlsx_file.name
             else:
-                xlsx_path = str(file_path_or_contents)
+                xlsx_path = str(await AsyncPath(file_path_or_contents).resolve())
 
             await run_sync(Xlsx2csv(xlsx_path).convert, csv_file.name)
             result = await process_file(csv_file.name, mime_type="text/csv")

kreuzberg/_pandoc.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import json
+import platform
 import subprocess
 from asyncio import gather
 from dataclasses import dataclass
@@ -305,7 +306,8 @@ async def _validate_pandoc_version() -> None:
         if version_ref["checked"]:
             return
 
-        result = await run_sync(subprocess.run, ["pandoc", "--version"], capture_output=True)
+        command = ["pandoc.exe" if platform.system() == "Windows" else "pandoc", "--version"]
+        result = await run_sync(subprocess.run, command, capture_output=True)
         version = result.stdout.decode().split("\n")[0].split()[1]
         if not version.startswith("3."):
             raise MissingDependencyError("Pandoc version 3 or above is required.")
@@ -322,7 +324,7 @@ async def _handle_extract_metadata(input_file: str | PathLike[str], *, mime_type
     with NamedTemporaryFile(suffix=".json") as metadata_file:
         try:
             command = [
-                "pandoc",
+                "pandoc.exe" if platform.system() == "Windows" else "pandoc",
                 str(input_file),
                 f"--from={pandoc_type}",
                 "--to=json",
@@ -357,7 +359,7 @@ async def _handle_extract_file(
 
     with NamedTemporaryFile(suffix=".md") as output_file:
         command = [
-            "pandoc",
+            "pandoc.exe" if platform.system() == "Windows" else "pandoc",
             str(input_file),
             f"--from={pandoc_type}",
             "--to=markdown",

kreuzberg/_tesseract.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import platform
 import re
 import subprocess
 from asyncio import gather
@@ -186,7 +187,8 @@ async def validate_tesseract_version() -> None:
         if version_ref["checked"]:
             return
 
-        result = await run_sync(subprocess.run, ["tesseract", "--version"], capture_output=True)
+        command = ["tesseract.exe" if platform.system() == "Windows" else "tesseract", "--version"]
+        result = await run_sync(subprocess.run, command, capture_output=True)
         version_match = re.search(r"tesseract\s+(\d+)", result.stdout.decode())
         if not version_match or int(version_match.group(1)) < 5:
             raise MissingDependencyError("Tesseract version 5 or above is required.")
@@ -218,7 +220,7 @@ async def process_file(
         output_file_name = output_file.name.replace(".txt", "")
         try:
             command = [
-                "tesseract",
+                "tesseract.exe" if platform.system() == "Windows" else "tesseract",
                 str(input_file),
                 output_file_name,
                 "-l",