kreuzberg 3.1.1__tar.gz → 3.1.2__tar.gz

{kreuzberg-3.1.1 → kreuzberg-3.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kreuzberg
-Version: 3.1.1
+Version: 3.1.2
 Summary: A text extraction library supporting PDFs, images, office documents and more
 Author-email: Na'aman Hirschfeld <nhirschfed@gmail.com>
 License: MIT
@@ -27,12 +27,12 @@ License-File: LICENSE
 Requires-Dist: anyio>=4.9.0
 Requires-Dist: charset-normalizer>=3.4.1
 Requires-Dist: exceptiongroup>=1.2.2; python_version < "3.11"
-Requires-Dist: html-to-markdown>=1.2.1
+Requires-Dist: html-to-markdown>=1.3.0
 Requires-Dist: playa-pdf>=0.4.1
 Requires-Dist: pypdfium2==4.30.0
-Requires-Dist: python-calamine>=0.3.1
+Requires-Dist: python-calamine>=0.3.2
 Requires-Dist: python-pptx>=1.0.2
-Requires-Dist: typing-extensions>=4.12.2; python_version < "3.12"
+Requires-Dist: typing-extensions>=4.13.1; python_version < "3.12"
 Provides-Extra: all
 Requires-Dist: easyocr>=1.7.2; extra == "all"
 Requires-Dist: gmft>=0.4.1; extra == "all"
@@ -140,7 +140,7 @@ For comprehensive documentation, visit our [GitHub Pages](https://goldziher.gith
 
 Kreuzberg supports a wide range of document formats:
 
-- **Documents**: PDF, DOCX, DOC, RTF, TXT, EPUB, etc.
+- **Documents**: PDF, DOCX, RTF, TXT, EPUB, etc.
 - **Images**: JPG, PNG, TIFF, BMP, GIF, etc.
 - **Spreadsheets**: XLSX, XLS, CSV, etc.
 - **Presentations**: PPTX, PPT, etc.
@@ -162,11 +162,11 @@ This library is open to contribution. Feel free to open issues or submit PRs. It
 
 ### Local Development
 
-1. Clone the repo
-1. Install the system dependencies
-1. Install the full dependencies with `uv sync`
-1. Install the pre-commit hooks with: `pre-commit install && pre-commit install --hook-type commit-msg`
-1. Make your changes and submit a PR
+- Clone the repo
+- Install the system dependencies
+- Install the full dependencies with `uv sync`
+- Install the pre-commit hooks with: `pre-commit install && pre-commit install --hook-type commit-msg`
+- Make your changes and submit a PR
 
 ## License
 

{kreuzberg-3.1.1 → kreuzberg-3.1.2}/README.md

@@ -86,7 +86,7 @@ For comprehensive documentation, visit our [GitHub Pages](https://goldziher.gith
 
 Kreuzberg supports a wide range of document formats:
 
-- **Documents**: PDF, DOCX, DOC, RTF, TXT, EPUB, etc.
+- **Documents**: PDF, DOCX, RTF, TXT, EPUB, etc.
 - **Images**: JPG, PNG, TIFF, BMP, GIF, etc.
 - **Spreadsheets**: XLSX, XLS, CSV, etc.
 - **Presentations**: PPTX, PPT, etc.
@@ -108,11 +108,11 @@ This library is open to contribution. Feel free to open issues or submit PRs. It
 
 ### Local Development
 
-1. Clone the repo
-1. Install the system dependencies
-1. Install the full dependencies with `uv sync`
-1. Install the pre-commit hooks with: `pre-commit install && pre-commit install --hook-type commit-msg`
-1. Make your changes and submit a PR
+- Clone the repo
+- Install the system dependencies
+- Install the full dependencies with `uv sync`
+- Install the pre-commit hooks with: `pre-commit install && pre-commit install --hook-type commit-msg`
+- Make your changes and submit a PR
 
 ## License
 

{kreuzberg-3.1.1 → kreuzberg-3.1.2}/kreuzberg/_extractors/_pandoc.py

@@ -228,7 +228,7 @@ class PandocExtractor(Extractor):
             command = ["pandoc", "--version"]
             result = await run_process(command)
 
-            version_match = re.search(r"pandoc\s+v?(\d+)\.\d+\.\d+", result.stdout.decode())
+            version_match = re.search(r"pandoc\s+v?(\d+)\.\d+(\.\d+)?", result.stdout.decode())
             if not version_match or int(version_match.group(1)) < MINIMAL_SUPPORTED_PANDOC_VERSION:
                 raise MissingDependencyError(
                     "Pandoc version 2 or above is a required system dependency. Please install it on your system and make sure its available in $PATH."

{kreuzberg-3.1.1 → kreuzberg-3.1.2}/kreuzberg.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kreuzberg
-Version: 3.1.1
+Version: 3.1.2
 Summary: A text extraction library supporting PDFs, images, office documents and more
 Author-email: Na'aman Hirschfeld <nhirschfed@gmail.com>
 License: MIT
@@ -27,12 +27,12 @@ License-File: LICENSE
 Requires-Dist: anyio>=4.9.0
 Requires-Dist: charset-normalizer>=3.4.1
 Requires-Dist: exceptiongroup>=1.2.2; python_version < "3.11"
-Requires-Dist: html-to-markdown>=1.2.1
+Requires-Dist: html-to-markdown>=1.3.0
 Requires-Dist: playa-pdf>=0.4.1
 Requires-Dist: pypdfium2==4.30.0
-Requires-Dist: python-calamine>=0.3.1
+Requires-Dist: python-calamine>=0.3.2
 Requires-Dist: python-pptx>=1.0.2
-Requires-Dist: typing-extensions>=4.12.2; python_version < "3.12"
+Requires-Dist: typing-extensions>=4.13.1; python_version < "3.12"
 Provides-Extra: all
 Requires-Dist: easyocr>=1.7.2; extra == "all"
 Requires-Dist: gmft>=0.4.1; extra == "all"
@@ -140,7 +140,7 @@ For comprehensive documentation, visit our [GitHub Pages](https://goldziher.gith
 
 Kreuzberg supports a wide range of document formats:
 
-- **Documents**: PDF, DOCX, DOC, RTF, TXT, EPUB, etc.
+- **Documents**: PDF, DOCX, RTF, TXT, EPUB, etc.
 - **Images**: JPG, PNG, TIFF, BMP, GIF, etc.
 - **Spreadsheets**: XLSX, XLS, CSV, etc.
 - **Presentations**: PPTX, PPT, etc.
@@ -162,11 +162,11 @@ This library is open to contribution. Feel free to open issues or submit PRs. It
 
 ### Local Development
 
-1. Clone the repo
-1. Install the system dependencies
-1. Install the full dependencies with `uv sync`
-1. Install the pre-commit hooks with: `pre-commit install && pre-commit install --hook-type commit-msg`
-1. Make your changes and submit a PR
+- Clone the repo
+- Install the system dependencies
+- Install the full dependencies with `uv sync`
+- Install the pre-commit hooks with: `pre-commit install && pre-commit install --hook-type commit-msg`
+- Make your changes and submit a PR
 
 ## License
 

{kreuzberg-3.1.1 → kreuzberg-3.1.2}/kreuzberg.egg-info/requires.txt

@@ -1,16 +1,16 @@
 anyio>=4.9.0
 charset-normalizer>=3.4.1
-html-to-markdown>=1.2.1
+html-to-markdown>=1.3.0
 playa-pdf>=0.4.1
 pypdfium2==4.30.0
-python-calamine>=0.3.1
+python-calamine>=0.3.2
 python-pptx>=1.0.2
 
 [:python_version < "3.11"]
 exceptiongroup>=1.2.2
 
 [:python_version < "3.12"]
-typing-extensions>=4.12.2
+typing-extensions>=4.13.1
 
 [all]
 easyocr>=1.7.2

{kreuzberg-3.1.1 → kreuzberg-3.1.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "kreuzberg"
-version = "3.1.1"
+version = "3.1.2"
 description = "A text extraction library supporting PDFs, images, office documents and more"
 readme = "README.md"
 keywords = [
@@ -40,12 +40,12 @@ dependencies = [
   "anyio>=4.9.0",
   "charset-normalizer>=3.4.1",
   "exceptiongroup>=1.2.2; python_version<'3.11'",
-  "html-to-markdown>=1.2.1",
+  "html-to-markdown>=1.3.0",
   "playa-pdf>=0.4.1",
   "pypdfium2==4.30.0",                                # pinned due to bug in 4.30.1, until v5 is stable
-  "python-calamine>=0.3.1",
+  "python-calamine>=0.3.2",
   "python-pptx>=1.0.2",
-  "typing-extensions>=4.12.2; python_version<'3.12'",
+  "typing-extensions>=4.13.1; python_version<'3.12'",
 ]
 
 optional-dependencies.all = [
@@ -82,10 +82,10 @@ dev = [
   "mypy>=1.15.0",
   "pre-commit>=4.2.0",
   "pytest>=8.3.5",
-  "pytest-cov>=6.1.0",
+  "pytest-cov>=6.1.1",
   "pytest-mock>=3.14.0",
   "pytest-timeout>=2.3.1",
-  "ruff>=0.11.2",
+  "ruff>=0.11.4",
   "trio>=0.29.0",
   "uv-bump",
 ]