chore: switch to uv (#135)

Signed-off-by: rmdg88 <rmdg88@gmail.com>

.github/actions/setup-poetry/action.yml

@@ -1,26 +0,0 @@
-name: 'Set up Poetry and install'
-description: 'Set up a specific version of Poetry and install dependencies using caching.'
-inputs:
-  python-version:
-    description: "Version range or exact version of Python or PyPy to use, using SemVer's version range syntax."
-    default: '3.11'
-runs:
-  using: 'composite'
-  steps:
-    - name: Install poetry
-      run: pipx install poetry==1.8.4
-      shell: bash
-    - uses: actions/setup-python@v5
-      id: py
-      with:
-        python-version: ${{ inputs.python-version }}
-        update-environment: false
-        cache: 'poetry'
-    - name: Setup poetry env with correct python
-      run: |
-        poetry env use ${{ steps.py.outputs.python-path }}
-        poetry run python --version
-      shell: bash
-    - name: Install only dependencies and not the package itself
-      run: poetry install --all-extras --no-root
-      shell: bash

.github/scripts/build_rhel.sh

@@ -3,8 +3,8 @@
 set -e  # trigger failure on error - do not remove!
 set -x  # display command on output
 
-# Build the Python package with Poetry
-poetry build -f sdist
+# Build the Python package with uv
+uv build --sdist
 
 sudo -E XDG_RUNTIME_DIR= podman build --progress=plain \
              --build-arg USE_SYSTEM_DEPS="$USE_SYSTEM_DEPS" \
@@ -35,7 +35,8 @@ sudo -E XDG_RUNTIME_DIR= podman build --progress=plain \
     # pre-install build requirements + wheel for "--no-build-isolation"
     # build docling-parse wheel in an isolated network namespace (unshare -rn)
     # install the wheel and its dependencies
-    RUN pip3.11 install poetry-core pybind11 wheel \
+    RUN pip3.11 install --upgrade pip \
+        && pip3.11 install --upgrade --ignore-installed "setuptools>=77.0.3" "wheel>=0.43.0,<1.0.0" "pybind11>=2.13.6" \
         && unshare -rn pip3.11 wheel \
             --no-deps --no-build-isolation -w /dist/ \
             /src/docling_parse*.tar.gz \
@@ -46,4 +47,4 @@ sudo -E XDG_RUNTIME_DIR= podman build --progress=plain \
 
     RUN pip3.11 install pytest \
         && pytest
-EOF
+EOF

.github/scripts/release.sh

@@ -10,11 +10,11 @@ fi
 CHGLOG_FILE="${CHGLOG_FILE:-CHANGELOG.md}"
 
 # update package version
-poetry version "${TARGET_VERSION}"
+uv version "${TARGET_VERSION}"
 
 # collect release notes
 REL_NOTES=$(mktemp)
-poetry run semantic-release changelog --unreleased >> "${REL_NOTES}"
+uv run semantic-release changelog --unreleased >> "${REL_NOTES}"
 
 # update changelog
 TMP_CHGLOG=$(mktemp)

.github/workflows/cd.yml

@@ -11,27 +11,28 @@ jobs:
   code-checks:
     uses: ./.github/workflows/checks.yml
   pre-release-check:
-    runs-on: ubuntu-22.04
+    runs-on: ubuntu-24.04
     outputs:
       TARGET_TAG_V: ${{ steps.version_check.outputs.TRGT_VERSION }}
     steps:
       - uses: actions/checkout@v4
         with:
           fetch-depth: 0  # for fetching tags, required for semantic-release
-      - uses: ./.github/actions/setup-poetry
+      - name: Install uv and set the python version
+        uses: ./.astral-sh/setup-uv@v5
       - name: Check version of potential release
         id: version_check
         run: |
-            TRGT_VERSION=$(poetry run semantic-release print-version)
+            TRGT_VERSION=$(uv run semantic-release print-version)
             echo "TRGT_VERSION=${TRGT_VERSION}" >> $GITHUB_OUTPUT
             echo "${TRGT_VERSION}"
       - name: Check notes of potential release
-        run: poetry run semantic-release changelog --unreleased
+        run: uv run semantic-release changelog --unreleased
   release:
     needs: [code-checks, pre-release-check]
     if: needs.pre-release-check.outputs.TARGET_TAG_V != ''
     environment: auto-release
-    runs-on: ubuntu-22.04
+    runs-on: ubuntu-24.04
     concurrency: release
     steps:
       - uses: actions/create-github-app-token@v1
@@ -43,7 +44,8 @@ jobs:
         with:
           token: ${{ steps.app-token.outputs.token }}
           fetch-depth: 0  # for fetching tags, required for semantic-release
-      - uses: ./.github/actions/setup-poetry
+      - name: Install uv and set the python version
+        uses: ./.astral-sh/setup-uv@v5
       - name: Run release script
         env:
           GH_TOKEN: ${{ steps.app-token.outputs.token }}

.github/workflows/checks.yml

@@ -6,24 +6,26 @@ env:
 
 jobs:
   run-checks:
-    runs-on: ubuntu-22.04
+    runs-on: ubuntu-24.04
     strategy:
       matrix:
         python-version: ['3.9', '3.10', '3.11', '3.12', '3.13']
     steps:
       - uses: actions/checkout@v4
-      - uses: ./.github/actions/setup-poetry
+      - name: Install uv and set the python version
+        uses: astral-sh/setup-uv@v5
         with:
           python-version: ${{ matrix.python-version }}
-      - name: Run styling check
-        run: poetry run pre-commit run --all-files
-      - name: Install with poetry
+      - name: Sync and install with uv
         run: |
-          poetry install --all-extras
+          uv sync --frozen --all-extras
           ls -l
           ls -l docling_parse
+      - name: Run styling check
+        run: |
+          uv run pre-commit run --all-files
       - name: Testing
         run: |
-          poetry run pytest -v tests
-      - name: Build with poetry
-        run: poetry build
+          uv run pytest -v tests
+      - name: Build with uv
+        run: uv build

.github/workflows/rhel.yml

@@ -1,12 +1,28 @@
 on:
   workflow_call:
 
+# jobs:
+#   run-checks:
+#     runs-on: ubuntu-latest
+#     steps:
+#       - uses: actions/checkout@v4
+#       - name: Install uv and python
+#         uses: astral-sh/setup-uv@v5
+#       - name: Install podman
+#         run: sudo apt-get update && sudo apt-get install -y podman
+#       - name: Run build in docker
+#         run: ./.github/scripts/build_rhel.sh
+#         shell: bash
+
 jobs:
   run-checks:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
     steps:
       - uses: actions/checkout@v4
-      - uses: ./.github/actions/setup-poetry
+      - name: Setup uv with python 3.11
+        uses: astral-sh/setup-uv@v5
+        with:
+          python-version: 3.11
       - name: Install podman
         run: sudo apt-get update && sudo apt-get install -y podman
       - name: Run build in docker

.github/workflows/wheels.yml

@@ -12,19 +12,18 @@ env:
 jobs:
   build_sdist:
     name: Build sdist artifacts
-    runs-on: ubuntu-22.04
+    runs-on: ubuntu-24.04
 
     steps:
       - name: Checkout
         uses: actions/checkout@v4
-      - name: Setup Python
-        uses: actions/setup-python@v5
-      - name: Install poetry
+      - name: Install uv and set the python version
+        uses: astral-sh/setup-uv@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Build sdist
         run: |
-          pipx install poetry
-      - name: Build sdit
-        run: |
-          poetry build -f sdist
+          uv build
           ls ./dist
       - name: Upload artifacts
         uses: actions/upload-artifact@v4
@@ -44,11 +43,11 @@ jobs:
         python-version: ["3.9", "3.10", "3.11", "3.12", "3.13"]
 
         os:
-          - name: "ubuntu-22.04"
+          - name: "ubuntu-24.04"
             platform: "linux"
             platform_id: "manylinux_x86_64"
 
-          - name: "ubuntu-22.04-arm"
+          - name: "ubuntu-24.04-arm"
             platform: "linux"
             platform_id: "manylinux_aarch64"
 
@@ -70,23 +69,11 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v4
 
-      - name: Setup Python
-        uses: actions/setup-python@v5
+      - name: Install uv and set the python version
+        uses: astral-sh/setup-uv@v5
         id: py
         with:
           python-version: ${{ matrix.python-version }}
-          update-environment: false
-
-      - name: Install Poetry
-        run: |
-          which python
-          python --version
-          which python3
-          python3 --version
-          echo "pythonpath: ${{ steps.py.outputs.python-path }}"
-          ${{ steps.py.outputs.python-path }} --version
-          pipx install poetry==1.8.4
-          poetry env use ${{ steps.py.outputs.python-path }}
 
       - name: Set up custom PATH and set py version to cpXYZ [windows]
         if: ${{matrix.os.platform_id == 'win_amd64'}}
@@ -124,32 +111,33 @@ jobs:
           BUILD_THREADS: "4"
           PYTORCH_MPS_HIGH_WATERMARK_RATIO: "0.0"
         run: |
-          PY_CACHE_TAG=$(poetry run python -c 'import sys;print(sys.implementation.cache_tag)')
+          PY_CACHE_TAG=$(uv run python -c 'import sys;print(sys.implementation.cache_tag)')
           echo "Building wheel ${CIBW_BUILD} ${{ env.CIBW_BUILD }}"
           echo "Building cp: ${{ env.python_cp_version }}"
           echo "Building cache_tag: ${PY_CACHE_TAG}"
           echo "Building platform_id: ${{ matrix.os.platform_id }}"
-          poetry run python --version
-          poetry run python --version | grep ${{ matrix.python-version }} 
-          poetry install --no-root --only=build
+          uv run python --version
+          uv run python --version | grep ${{ matrix.python-version }} 
           cat ./pyproject.toml
-          poetry run python -m cibuildwheel --output-dir wheelhouse
+          uv pip install --group build
+          rm -rf ./build || true
+          python -m cibuildwheel --output-dir ./wheelhouse
           echo "step 1"
           ls -l wheelhouse
-          poetry run wheel tags --remove --platform-tag macosx_${{ matrix.os.min_macos_version }}_0_x86_64 ./wheelhouse/*.whl
+          uv run wheel tags --remove --platform-tag macosx_${{ matrix.os.min_macos_version }}_0_x86_64 ./wheelhouse/*.whl
           rm -f ./wheelhouse/*arm64.whl
           echo "step 2"
           ls -l wheelhouse
-          poetry run delocate-wheel --require-archs x86_64 -v ./wheelhouse/*.whl
+          uv run delocate-wheel --require-archs x86_64 -v ./wheelhouse/*.whl
           echo "step 3"
           ls -l wheelhouse
           for file in ./wheelhouse/*.whl; do
             echo "Inspecting $file"
-            poetry run python -m zipfile --list "$file"
+            uv run python -m zipfile --list "$file"
             echo "Checking if .so is contained in the wheel"
-            poetry run python -m zipfile --list "$file" | grep \\.so
+            uv run python -m zipfile --list "$file" | grep \\.so
             echo "Checking if the correct python version is contained in the wheel"
-            poetry run python -m zipfile --list "$file" | grep ${PY_CACHE_TAG}
+            uv run python -m zipfile --list "$file" | grep ${PY_CACHE_TAG}
           done
           mkdir -p ./dist
           cp wheelhouse/*.whl ./dist/
@@ -175,32 +163,33 @@ jobs:
           PYTORCH_MPS_HIGH_WATERMARK_RATIO: "0.0"
           CUDA_VISIBLE_DEVICES: "cpu"
         run: |
-          PY_CACHE_TAG=$(poetry run python -c 'import sys;print(sys.implementation.cache_tag)')
+          PY_CACHE_TAG=$(uv run python -c 'import sys;print(sys.implementation.cache_tag)')
           echo "Building wheel ${CIBW_BUILD} ${{ env.CIBW_BUILD }}"
           echo "Building cp: ${{ env.python_cp_version }}"
           echo "Building cache_tag: ${PY_CACHE_TAG}"
           echo "Building platform_id: ${{ matrix.os.platform_id }}"
-          poetry run python --version
-          poetry run python --version | grep ${{ matrix.python-version }} 
-          poetry install --no-root --only=build
+          uv run python --version
+          uv run python --version | grep ${{ matrix.python-version }} 
           cat ./pyproject.toml
-          poetry run python -m cibuildwheel --output-dir wheelhouse
+          uv pip install --group build
+          rm -rf ./build || true
+          python -m cibuildwheel --output-dir ./wheelhouse
           echo "step 1"
           ls -l wheelhouse
-          poetry run wheel tags --remove --platform-tag macosx_${{ matrix.os.min_macos_version }}_0_arm64 ./wheelhouse/*.whl
+          uv run wheel tags --remove --platform-tag macosx_${{ matrix.os.min_macos_version }}_0_arm64 ./wheelhouse/*.whl
           rm -f ./wheelhouse/*x86_64.whl
           echo "step 2"
           ls -l wheelhouse
-          poetry run delocate-wheel --require-archs arm64 -v ./wheelhouse/*.whl
+          uv run delocate-wheel --require-archs arm64 -v ./wheelhouse/*.whl
           echo "step 3"
           ls -l wheelhouse
           for file in ./wheelhouse/*.whl; do
             echo "Inspecting $file"
-            poetry run python -m zipfile --list "$file"
+            uv run python -m zipfile --list "$file"
             echo "Checking if .so is contained in the wheel"
-            poetry run python -m zipfile --list "$file" | grep \\.so
+            uv run python -m zipfile --list "$file" | grep \\.so
             echo "Checking if the correct python version is contained in the wheel"
-            poetry run python -m zipfile --list "$file" | grep ${PY_CACHE_TAG}
+            uv run python -m zipfile --list "$file" | grep ${PY_CACHE_TAG}
           done
           mkdir -p ./dist
           cp wheelhouse/*.whl ./dist/
@@ -222,23 +211,24 @@ jobs:
           CIBW_BUILD_VERBOSITY: 3
           BUILD_THREADS: "8"
         run: |
-          PY_CACHE_TAG=$(poetry run python -c 'import sys;print(sys.implementation.cache_tag)')
+          PY_CACHE_TAG=$(uv run python -c 'import sys;print(sys.implementation.cache_tag)')
           echo "Building cp: ${{ env.python_cp_version }}"
           echo "Building cache_tag: ${PY_CACHE_TAG}"
           echo "Building platform_id: ${{ matrix.os.platform_id }}"
-          poetry run python --version
-          poetry run python --version | grep ${{ matrix.python-version }} 
-          poetry install --no-root --only=build
+          uv run python --version
+          uv run python --version | grep ${{ matrix.python-version }} 
           cat ./pyproject.toml
-          poetry run python -m cibuildwheel --output-dir ./wheelhouse
+          uv pip install --group build
+          rm -rf ./build || true
+          python -m cibuildwheel --output-dir ./wheelhouse
           ls -l ./wheelhouse
           for file in ./wheelhouse/*.whl; do
             echo "Inspecting $file"
-            poetry run python -m zipfile --list "$file"
+            uv run python -m zipfile --list "$file"
             echo "Checking if .so is contained in the wheel"
-            poetry run python -m zipfile --list "$file" | grep \\.so
+            uv run python -m zipfile --list "$file" | grep \\.so
             echo "Checking if the correct python version is contained in the wheel"
-            poetry run python -m zipfile --list "$file" | grep ${PY_CACHE_TAG}
+            uv run python -m zipfile --list "$file" | grep ${PY_CACHE_TAG}
           done
           mkdir -p ./dist
           cp wheelhouse/*.whl ./dist/
@@ -293,15 +283,20 @@ jobs:
           ASM_NASM: "C:/nasm/nasm.exe"
         shell: pwsh
         run: |
-          Remove-Item -Recurse -Force "C:\Strawberry\"
+          if (Test-Path "C:\Strawberry\") {
+              Remove-Item -Recurse -Force "C:\Strawberry\"
+          }
           $env:CMAKE_ARGS = "-DZLIB_LIBRARY=C:/windows-libs/external-libs/lib-mingw64/libz.a -DZLIB_INCLUDE_DIR=C:/windows-libs/external-libs/include -DJPEG_LIBRARY=C:/windows-libs/external-libs/lib-mingw64/libjpeg.a -DJPEG_INCLUDE_DIR=C:/windows-libs/external-libs/include"
-          poetry install --no-interaction --no-root --only=build
-          poetry run python -m cibuildwheel --output-dir .\wheelhouse
+          uv pip install --group build
+          if (Test-Path ".\build") {
+              Remove-Item -Recurse -Force ".\build"
+          }
+          uv run python -m cibuildwheel --output-dir .\wheelhouse
           Get-ChildItem -Path .\wheelhouse -Filter *.whl | Format-List
           Get-ChildItem -Path .\wheelhouse -Filter *.whl | ForEach-Object {
               $file = $_.FullName
               Write-Output "Inspecting $file"
-              poetry run python -m zipfile --list "$file"
+              uv run python -m zipfile --list "$file"
           }
           if (-not (Test-Path -Path .\dist)) {
               New-Item -Path .\dist -ItemType Directory
@@ -321,7 +316,7 @@ jobs:
     needs:
       - build_sdist
       - build_wheels
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
     environment:
       name: pypi
       url: https://pypi.org/p/docling-parse

.pre-commit-config.yaml

@@ -4,7 +4,7 @@ repos:
     hooks:
       - id: system
         name: Black
-        entry: poetry run black docling_parse tests
+        entry: uv run black docling_parse tests
         pass_filenames: false
         language: system
         files: '\.py$'
@@ -12,7 +12,7 @@ repos:
     hooks:
       - id: system
         name: isort
-        entry: poetry run isort docling_parse tests
+        entry: uv run isort docling_parse tests
         pass_filenames: false
         language: system
         files: '\.py$'
@@ -20,7 +20,7 @@ repos:
     hooks:
       - id: autoflake
         name: autoflake
-        entry: poetry run autoflake docling_parse tests
+        entry: uv run autoflake docling_parse tests
         pass_filenames: false
         language: system
         files: '\.py$'
@@ -28,7 +28,7 @@ repos:
     hooks:
       - id: mypy
         name: MyPy
-        entry: poetry run mypy docling_parse tests
+        entry: uv run mypy docling_parse tests
         pass_filenames: false
         language: system
         files: '\.py$'
@@ -36,14 +36,14 @@ repos:
 #    hooks:
 #      - id: pytest
 #        name: Pytest
-#        entry: poetry run pytest tests/
+#        entry: uv run pytest tests/
 #        pass_filenames: false
 #        language: system
 #        files: '\.py$'	
   - repo: local
     hooks:
       - id: system
-        name: Poetry check
-        entry: poetry check --lock
+        name: uv check
+        entry: uv lock --check
         pass_filenames: false
         language: system

CMakeLists.txt

@@ -1,4 +1,5 @@
-cmake_minimum_required(VERSION 3.12..3.26)
+# cmake_minimum_required(VERSION 3.12..3.26)
+cmake_minimum_required(VERSION 3.12)
 
 project(docling_parse VERSION 1.0.0 LANGUAGES CXX C)
 #set(CMAKE_VERBOSE_MAKEFILE off)

CONTRIBUTING.md

@@ -80,46 +80,33 @@ Please feel free to connect with us using the [discussion section](https://githu
 
 ## Developing
 
-### Usage of Poetry
+### Usage of uv
 
-We use Poetry to manage dependencies.
+We use uv to manage dependencies.
 
 
 #### Install
 
-To install, see the documentation here: https://python-poetry.org/docs/master/#installing-with-the-official-installer
+To install, see the documentation here: https://docs.astral.sh/uv/getting-started/installation/
 
-1. Install the Poetry globally in your machine
+1. Install uv standalone in your machine
     ```bash
-    curl -sSL https://install.python-poetry.org | python3 -
+    # On macOS and Linux.
+    curl -LsSf https://astral.sh/uv/install.sh | sh
     ```
-    The installation script will print the installation bin folder `POETRY_BIN` which you need in the next steps.
-
-2. Make sure Poetry is in your `$PATH`
-    - for `zsh`
-        ```sh
-        echo 'export PATH="POETRY_BIN:$PATH"' >> ~/.zshrc
-        ```
-    - for `bash`
-        ```sh
-        echo 'export PATH="POETRY_BIN:$PATH"' >> ~/.bashrc
-        ```
-
-3. The official guidelines linked above include useful details on the configuration of autocomplete for most shell environments, e.g. Bash and Zsh.
-
 
 #### Create a Virtual Environment and Install Dependencies
 
 To activate the Virtual Environment, run:
 
 ```bash
-poetry shell
+uv shell
 ```
 
-To spawn a shell with the Virtual Environment activated. If the Virtual Environment doesn't exist, Poetry will create one for you. Then, to install dependencies, run:
+To spawn a shell with the Virtual Environment activated. If the Virtual Environment doesn't exist, uv will create one for you. Then, to install dependencies, run:
 
 ```bash
-poetry install
+uv install
 ```
 
 **(Advanced) Use a Specific Python Version**
@@ -127,7 +114,7 @@ poetry install
 If for whatever reason you need to work in a specific (older) version of Python, run:
 
 ```bash
-poetry env use $(which python3.9)
+uv venv use $(which python3.9)
 ```
 
 This creates a Virtual Environment with Python 3.9. For other versions, replace `$(which python3.9)` by the path to the interpreter (e.g., `/usr/bin/python3.9`) or use `$(which pythonX.Y)`.
@@ -136,7 +123,7 @@ This creates a Virtual Environment with Python 3.9. For other versions, replace
 #### Add a new dependency
 
 ```bash
-poetry add NAME
+uv add NAME
 ```
 
 ## Coding style guidelines

MANIFEST.in

@@ -0,0 +1,14 @@
+include CMakeLists.txt
+include build.py
+include *.md
+include uv.lock
+include LICENSE
+
+recursive-include app *
+recursive-include cmake *
+recursive-include src *
+recursive-include tests *
+
+recursive-include docling_parse *.so *.pyd *.dll
+recursive-include docling_parse/pdf_resources *
+recursive-include docling_parse/pdf_resources_v2 *

README.md

@@ -2,7 +2,7 @@
 
 [![PyPI version](https://img.shields.io/pypi/v/docling-parse)](https://pypi.org/project/docling-parse/)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/docling-parse)](https://pypi.org/project/docling-parse/)
-[![Poetry](https://img.shields.io/endpoint?url=https://python-poetry.org/badge/v0.json)](https://python-poetry.org/)
+[![uv](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/uv/main/assets/badge/v0.json)](https://github.com/astral-sh/uv)
 [![Pybind11](https://img.shields.io/badge/build-pybind11-blue)](https://github.com/pybind/pybind11/)
 [![Platforms](https://img.shields.io/badge/platform-macos%20|%20linux%20|%20windows-blue)](https://github.com/docling-project/docling-parse/)
 [![License MIT](https://img.shields.io/github/license/docling-project/docling-parse)](https://opensource.org/licenses/MIT)
@@ -12,7 +12,7 @@ Simple package to extract text, paths and bitmap images with coordinates from pr
 To do the visualizations yourself, simply run (change `word` into `char` or `line`),
 
 ```sh
-poetry run python ./docling_parse/visualize.py -i <path-to-pdf-file> -c word --interactive
+uv run python ./docling_parse/visualize.py -i <path-to-pdf-file> -c word --interactive
 ```
 
 <table>
@@ -183,16 +183,16 @@ If you dont have an input file, then a template input file will be printed on th
 
 ### Python
 
-To build the package, simply run (make sure [poetry](https://python-poetry.org/) is [installed](https://python-poetry.org/docs/#installing-with-the-official-installer)),
+To build the package, simply run (make sure [uv](https://docs.astral.sh/uv/) is [installed](https://docs.astral.sh/uv/getting-started/installation)),
 
 ```
-poetry install
+uv sync
 ```
 
 To test the package, run:
 
 ```
-poetry run pytest ./tests -v -s
+uv run pytest ./tests -v -s
 ```
 
 

cmake/extlib_cxxopts.cmake

@@ -23,7 +23,8 @@ else()
     include(CMakeParseArguments)
 
     set(CXXOPTS_URL https://github.com/jarro2783/cxxopts.git)
-    set(CXXOPTS_TAG v3.2.0)
+    # set(CXXOPTS_TAG v3.2.0)
+    set(CXXOPTS_TAG v3.3.1)
 
     ExternalProject_Add(extlib_cxxopts
 

cmake/extlib_jpeg.cmake

@@ -17,8 +17,8 @@ else()
     include(CMakeParseArguments)
 
     set(JPEG_URL https://github.com/libjpeg-turbo/libjpeg-turbo.git)
-    set(JPEG_TAG 3.0.3)
-
+    # set(JPEG_TAG 3.0.3)
+    set(JPEG_TAG 3.1.1)
     ExternalProject_Add(extlib_jpeg
 
         PREFIX extlib_jpeg

cmake/extlib_json.cmake

@@ -16,8 +16,8 @@ else()
     include(CMakeParseArguments)
 
     set(JSON_URL https://github.com/nlohmann/json.git)
-    set(JSON_TAG v3.11.3)
-
+    # set(JSON_TAG v3.11.3)
+    set(JSON_TAG v3.12.0)
     ExternalProject_Add(extlib_json
         
         PREFIX extlib_json

cmake/extlib_pybind11.cmake

@@ -5,7 +5,8 @@ include(ExternalProject)
 include(CMakeParseArguments)
 
 set(PYBIND11_URL https://github.com/pybind/pybind11.git)
-set(PYBIND11_TAG v2.13.5)
+# set(PYBIND11_TAG v2.13.5)
+set(PYBIND11_TAG v2.13.6)
 
 ExternalProject_Add(extlib_pybind11
     PREFIX extlib_pybind11

cmake/extlib_qpdf_v11.cmake

@@ -18,8 +18,8 @@ else()
   include(CMakeParseArguments)
 
   set(QPDF_URL https://github.com/qpdf/qpdf.git)
-  set(QPDF_TAG v11.9.1 )
-  
+  #set(QPDF_TAG v11.9.1 )
+  set(QPDF_TAG v11.10.0 )
   set(QPDF_LIB ${EXTERNALS_PREFIX_PATH}/lib/libqpdf.a)
   set(JPEG_LIB ${EXTERNALS_PREFIX_PATH}/lib/libjpeg.a)
 

cmake/extlib_utf8.git.cmake

@@ -19,8 +19,8 @@ else()
     include(CMakeParseArguments)
 
     set(UTF8_URL https://github.com/nemtrif/utfcpp.git)
-    set(UTF8_TAG v4.0.5)
-
+    # set(UTF8_TAG v4.0.5)
+    set(UTF8_TAG v4.0.6)
     ExternalProject_Add(extlib_utf8
         PREFIX extlib_utf8
 

pyproject.toml

@@ -1,20 +1,10 @@
-[tool.poetry]
+[project]
 name = "docling-parse"
 version = "4.1.0"
 description = "Simple package to extract text with coordinates from programmatic PDFs"
-authors = ["Peter Staar <taa@zurich.ibm.com>"]
-maintainers = [
-    "Peter Staar <taa@zurich.ibm.com>",
-    "Christoph Auer <cau@zurich.ibm.com>",
-    "Michele Dolfi <dol@zurich.ibm.com>",
-    "Panos Vagenas <pva@zurich.ibm.com>",
-    "Maxim Lysak <mly@zurich.ibm.com>",
-]
-repository = "https://github.com/docling-project/docling-parse"
-homepage = "https://github.com/docling-project/docling-parse"
-keywords= ["docling", "pdf", "parser"]
+license = "MIT"
+keywords = ["docling", "pdf", "parser"]
 classifiers = [
-    "License :: OSI Approved :: MIT License",
     "Operating System :: MacOS :: MacOS X",
     "Operating System :: POSIX :: Linux",
     "Operating System :: Microsoft :: Windows",
@@ -24,59 +14,78 @@ classifiers = [
     "Programming Language :: C++",
     "Programming Language :: Python :: 3"
 ]
-license = "MIT"
 readme = "README.md"
-packages = [{include = "docling_parse"}]
-include = [
-    {path = "docling_parse/*.so", format = "wheel"},
-    {path = "docling_parse/*.pyd", format = "wheel"},
-    {path = "docling_parse/*.dll", format = "wheel"},
-    {path = "docling_parse/pdf_resources", format = ["sdist", "wheel"]},
-    {path = "docling_parse/pdf_resources", format = ["sdist", "wheel"]},
-    {path = "docling_parse/pdf_resources_v2", format = ["sdist", "wheel"]},
-    {path = "CMakeLists.txt", format = "sdist"},
-    {path = "build.py", format = "sdist"},
-    {path = "*.md", format = "sdist"},
-    {path = "poetry.lock", format = "sdist"},
-    {path = "app/*.cpp", format = "sdist"},
-    {path = "cmake/", format = "sdist"},
-    {path = "app/", format = "sdist"},
-    {path = "src/", format = "sdist"},
-    {path = "tests/", format = "sdist"},
+authors = [
+    {name = "Peter Staar", email = "taa@zurich.ibm.com"},
+    {name = "Christoph Auer", email = "cau@zurich.ibm.com"},
+    {name = "Michele Dolfi", email = "dol@zurich.ibm.com"},
+    {name = "Panos Vagenas", email = "pva@zurich.ibm.com"},
+    {name = "Maxim Lysak", email = "mly@zurich.ibm.com"},
 ]
-build = "build.py"
-
-[tool.poetry.dependencies]
-python = "^3.9"
-tabulate = ">=0.9.0,<1.0.0"
-pywin32 = { version = ">=305", markers = "sys_platform == 'win32'" }
-pillow = ">=10.0.0,<12.0.0"
-pydantic = "^2.0.0"
-docling-core = "^2.29.0"
-
-[tool.poetry.group.dev.dependencies]
-pytest = "^7.4.2"
-black = {extras = ["jupyter"], version = "^24.4.2"}
-python-semantic-release = "^7.32.2"
-pre-commit = "^3.7.1"
-isort = "^5.10.1"
-mypy = "^1.13.0"
-tqdm = "^4.67.0"
-boto = "^2.49.0"
-boto3 = "^1.35.67"
-autoflake = "^2.3.1"
-
-[tool.poetry.group.build.dependencies]
-cibuildwheel = "^2.19.2"
-wheel = "^0.43.0"
-delocate = "^0.11.0"
-
-[tool.poetry.group.test.dependencies]
-pytest = "^7.4.2"
+requires-python = ">=3.9"
+dependencies = [
+    "tabulate>=0.9.0,<1.0.0",
+    "pillow>=10.0.0,<12.0.0",
+    "pydantic>=2.0.0",
+    "docling-core>=2.29.0",
+    "pywin32>=305; sys_platform == 'win32'",
+]
+[project.urls]
+Homepage = "https://github.com/docling-project/docling-parse"
+Repository = "https://github.com/docling-project/docling-parse"
 
 [build-system]
-requires = ["poetry-core", "pybind11>=2.13.1"]
-build-backend = "poetry.core.masonry.api"
+requires = [
+    "setuptools>=77.0.3",
+    "pybind11>=2.13.6",
+    "cibuildwheel>=2.19.2,<3.0.0",
+    "wheel>=0.43.0,<1.0.0",
+    "delocate>=0.11.0,<1.0.0",
+    "cmake>=3.27.0,<4.0.0"
+]
+build-backend = "setuptools.build_meta"
+
+[dependency-groups]
+build = [
+    "setuptools>=77.0.3",
+    "pybind11>=2.13.6",
+    "cibuildwheel>=2.19.2,<3.0.0",
+    "wheel>=0.43.0,<1.0.0",
+    "delocate>=0.11.0,<1.0.0",
+    "cmake>=3.27.0,<4.0.0"
+]
+dev = [
+    "pytest>=7.4.2,<8.0.0",
+    "black[jupyter]>=24.4.2,<25.0.0",
+    "python-semantic-release>=7.32.2,<8.0.0",
+    "pre-commit>=3.7.1,<4.0.0",
+    "isort>=5.10.1,<6.0.0",
+    "mypy>=1.13.0,<2.0.0",
+    "tqdm>=4.67.0,<5.0.0",
+    "boto>=2.49.0,<3.0.0",
+    "boto3>=1.35.67,<2.0.0",
+    "autoflake>=2.3.1,<3.0.0"
+]
+[tool.uv]
+package = true
+default-groups = "all"
+
+[tool.setuptools]
+include-package-data = true
+
+[tool.setuptools.packages.find]
+where = [".", "docling_parse*"]
+include = ["docling_parse*"]
+
+[tool.setuptools.package-data]
+"docling_parse" = [
+    "*.so", "*.pyd", "*.dll",
+    "pdf_resources/*",
+    "pdf_resources_v2/*"
+]
+
+[tool.setuptools.exclude-package-data]
+"docling_parse" = ["*.pyc", "__pycache__"]
 
 [tool.black]
 line-length = 88
@@ -87,7 +96,7 @@ preview = true
 [tool.isort]
 profile = "black"
 line_length = 88
-py_version=310
+py_version = 310
 multi_line_output = 3
 include_trailing_comma = true
 
@@ -101,34 +110,22 @@ recursive = true
 
 [tool.mypy]
 pretty = true
-# strict = true
 no_implicit_optional = true
 namespace_packages = true
 show_error_codes = true
 python_version = "3.9"
-# plugins = ["pydantic.mypy"]
-
-#[mypy-docling_parse.*]
-#ignore_missing_imports = True
 
 [[tool.mypy.overrides]]
 module = [
     "tabulate.*",
     "botocore.*",
-    "boto3.*",
+    "boto3.*"
 ]
 ignore_missing_imports = true
 
-
 [tool.semantic_release]
-# for default values check:
-# https://github.com/python-semantic-release/python-semantic-release/blob/v7.32.2/semantic_release/defaults.cfg
-
 version_source = "tag_only"
 branch = "main"
-
-# configure types which should trigger minor and patch version bumps respectively
-# (note that they must be a subset of the configured allowed types):
 parser_angular_allowed_types = "build,chore,ci,docs,feat,fix,perf,style,refactor,test"
 parser_angular_minor_types = "feat"
 parser_angular_patch_types = "fix,perf"

setup.py

@@ -0,0 +1,28 @@
+from setuptools import setup, find_packages, Distribution
+from setuptools.command.build_py import build_py as _build_py
+import subprocess
+import sys
+
+class CustomBuildPy(_build_py):
+    def run(self):
+        subprocess.check_call([sys.executable, "build.py"])
+        super().run()
+
+class BinaryDistribution(Distribution):
+    def has_ext_modules(self):
+        return True
+
+setup(
+    packages=find_packages(include=["docling_parse", "docling_parse.*"]),
+    distclass=BinaryDistribution,
+    cmdclass={"build_py": CustomBuildPy},
+    zip_safe=False,
+    include_package_data=True,
+    package_data={
+        "docling_parse": [
+            "*.so", "*.pyd", "*.dll",
+            "pdf_resources/*",
+            "pdf_resources_v2/*",
+        ],
+    },
+)