bump unstructured-inference (#3711)

This PR bumps `unstructured-inference` to `0.8.0`, which introduces vectorized data structure for layout elements and text regions. This PR also cleans up a few places in CI that has repeated definition of env variables or missing installation of testing dependencies in cache. A few document ingest results are changed: - two places for `biomed-api` (actually processed locally on runner) are due to very small changes in numerical results of the bounding box areas: one results in a duplicated page number/header and another results in a deduplication of a word of a sentence that starts in a new line. (yes, two cases goes in opposite directions) - the layout parser paper now outputs the code lines with page number inside the code box as list items --------- Co-authored-by: ryannikolaidis <[email protected]> Co-authored-by: badGarnet <[email protected]> Co-authored-by: christinestraub <[email protected]>
Unstructured-IO · Oct 21, 2024 · a11ad22 · a11ad22
1 parent e764bc5
commit a11ad22
Show file tree

Hide file tree

Showing 23 changed files with 184 additions and 109 deletions.
diff --git a/.github/actions/base-cache/action.yml b/.github/actions/base-cache/action.yml
@@ -30,14 +30,17 @@ runs:
       shell: bash
       run: |
         python${{ inputs.python-version }} -m pip install --upgrade virtualenv
-        python${{ inputs.python-version }} -m venv .venv
+        if [ ! -d ".venv" ]; then
+          python${{ inputs.python-version }} -m venv .venv
+        fi
         source .venv/bin/activate
         [ ! -d "$NLTK_DATA" ] && mkdir "$NLTK_DATA"
         if [ "${{ inputs.python-version == '3.12' }}" == "true" ]; then
           python -m ensurepip --upgrade
           python -m pip install --upgrade setuptools
         fi
         make install-ci
+        make install-nltk-models
     - name: Save Cache
       if: steps.virtualenv-cache-restore.outputs.cache-hit != 'true'
       id: virtualenv-cache-save

diff --git a/.github/actions/base-ingest-cache/action.yml b/.github/actions/base-ingest-cache/action.yml
@@ -18,7 +18,7 @@ runs:
         path: |
           .venv
           nltk_data
-        key: unstructured-ingest-${{ runner.os }}-${{ inputs.python-version }}-${{ hashFiles('requirements/ingest/*.txt') }}-${{ hashFiles('requirements/*.txt') }}
+        key: unstructured-ingest-${{ runner.os }}-${{ inputs.python-version }}-${{ hashFiles('requirements/ingest/*.txt', 'requirements/*.txt') }}
         lookup-only: ${{ inputs.check-only }}
     - name: Set up Python ${{ inputs.python-version }}
       if: steps.ingest-virtualenv-cache-restore.outputs.cache-hit != 'true'
@@ -39,6 +39,8 @@ runs:
           python -m pip install --upgrade setuptools
         fi
         make install-ci
+        make install-nltk-models
+        make install-all-docs
         make install-ingest
     - name: Save Ingest Cache
       if: steps.ingest-virtualenv-cache-restore.outputs.cache-hit != 'true'
@@ -48,5 +50,5 @@ runs:
         path: |
           .venv
           nltk_data
-        key: unstructured-ingest-${{ runner.os }}-${{ inputs.python-version }}-${{ hashFiles('requirements/ingest/*.txt') }}-${{ hashFiles('requirements/*.txt') }}
+        key: unstructured-ingest-${{ runner.os }}-${{ inputs.python-version }}-${{ hashFiles('requirements/ingest/*.txt', 'requirements/*.txt') }}
 
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -12,14 +12,15 @@ permissions:
   id-token: write
   contents: read
 
+env:
+  NLTK_DATA: ${{ github.workspace }}/nltk_data
+
 jobs:
   setup:
     strategy:
       matrix:
         python-version: ["3.9","3.10","3.11", "3.12"]
     runs-on: ubuntu-latest
-    env:
-      NLTK_DATA: ${{ github.workspace }}/nltk_data
     steps:
     - uses: actions/checkout@v4
     - uses: ./.github/actions/base-cache
@@ -78,8 +79,6 @@ jobs:
     strategy:
       matrix:
         python-version: ["3.9","3.10","3.11"]
-    env:
-      NLTK_DATA: ${{ github.workspace }}/nltk_data
     runs-on: ubuntu-latest
     needs: [setup, changelog]
     steps:
@@ -185,8 +184,6 @@ jobs:
         python-version: ["3.10"]
         extra: ["csv", "docx", "odt", "markdown", "pypandoc", "pdf-image", "pptx", "xlsx"]
     runs-on: ubuntu-latest
-    env:
-      NLTK_DATA: ${{ github.workspace }}/nltk_data
     needs: [setup, lint, test_unit_no_extras]
     steps:
     - uses: actions/checkout@v4
@@ -220,15 +217,14 @@ jobs:
         sudo apt-get update
         sudo apt-get install -y tesseract-ocr tesseract-ocr-kor
         tesseract --version
+        make install-${{ matrix.extra }}
         make test-extra-${{ matrix.extra }} CI=true
 
   setup_ingest:
     strategy:
       matrix:
         python-version: [ "3.9","3.10" ]
     runs-on: ubuntu-latest
-    env:
-      NLTK_DATA: ${{ github.workspace }}/nltk_data
     needs: [setup]
     steps:
       - uses: actions/checkout@v4
@@ -307,7 +303,6 @@ jobs:
         MXBAI_API_KEY: ${{secrets.MXBAI_API_KEY}}
         OCR_AGENT: "unstructured.partition.utils.ocr_models.tesseract_ocr.OCRAgentTesseract"
         CI: "true"
-        NLTK_DATA: ${{ github.workspace }}/nltk_data
         PYTHON: python${{ matrix.python-version }}
       run: |
         source .venv/bin/activate
@@ -320,6 +315,8 @@ jobs:
         sudo apt-get install -y tesseract-ocr-kor
         sudo apt-get install diffstat
         tesseract --version
+        make install-all-docs
+        make install-ingest
         ./test_unstructured_ingest/test-ingest-src.sh
 
 
@@ -329,8 +326,6 @@ jobs:
         # NOTE(yuming): Unstructured API only use Python 3.10
         python-version: ["3.10"]
     runs-on: ubuntu-latest
-    env:
-      NLTK_DATA: ${{ github.workspace }}/nltk_data
     needs: [setup, lint]
     steps:
     - uses: actions/checkout@v4

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,7 +1,8 @@
-## 0.16.1-dev5
+## 0.16.1-dev6
 
 ### Enhancements
 
+* **Bump `unstructured-inference` to 0.7.39** and upgrade other dependencies
 * **Round coordinates** Round coordinates when computing bounding box overlaps in `pdfminer_processing.py` to nearest machine precision. This can help reduce underterministic behavior from machine precision that affects which bounding boxes to combine.
 
 ### Features

diff --git a/requirements/base.txt b/requirements/base.txt
@@ -4,7 +4,7 @@
 #
 #    pip-compile ./base.in
 #
-anyio==4.6.0
+anyio==4.6.2.post1
     # via httpx
 backoff==2.2.1
     # via -r ./base.in
@@ -20,15 +20,15 @@ cffi==1.17.1
     # via cryptography
 chardet==5.2.0
     # via -r ./base.in
-charset-normalizer==3.3.2
+charset-normalizer==3.4.0
     # via
     #   requests
     #   unstructured-client
 click==8.1.7
     # via
     #   nltk
     #   python-oxmsg
-cryptography==43.0.1
+cryptography==43.0.3
     # via unstructured-client
 dataclasses-json==0.6.7
     # via
@@ -62,7 +62,7 @@ langdetect==1.0.9
     # via -r ./base.in
 lxml==5.3.0
     # via -r ./base.in
-marshmallow==3.22.0
+marshmallow==3.23.0
     # via
     #   dataclasses-json
     #   unstructured-client
@@ -84,7 +84,7 @@ packaging==24.1
     # via
     #   marshmallow
     #   unstructured-client
-psutil==6.0.0
+psutil==6.1.0
     # via -r ./base.in
 pycparser==2.22
     # via cffi

diff --git a/requirements/dev.txt b/requirements/dev.txt
@@ -4,7 +4,7 @@
 #
 #    pip-compile ./dev.in
 #
-build==1.2.2
+build==1.2.2.post1
     # via pip-tools
 cfgv==3.4.0
     # via pre-commit
@@ -13,7 +13,7 @@ click==8.1.7
     #   -c ./base.txt
     #   -c ./test.txt
     #   pip-tools
-distlib==0.3.8
+distlib==0.3.9
     # via virtualenv
 filelock==3.16.1
     # via virtualenv
@@ -36,7 +36,7 @@ platformdirs==4.3.6
     # via
     #   -c ./test.txt
     #   virtualenv
-pre-commit==3.8.0
+pre-commit==4.0.1
     # via -r ./dev.in
 pyproject-hooks==1.2.0
     # via
@@ -51,7 +51,7 @@ tomli==2.0.2
     #   -c ./test.txt
     #   build
     #   pip-tools
-virtualenv==20.26.6
+virtualenv==20.27.0
     # via pre-commit
 wheel==0.44.0
     # via pip-tools

diff --git a/requirements/extra-epub.txt b/requirements/extra-epub.txt
@@ -4,5 +4,5 @@
 #
 #    pip-compile ./extra-epub.in
 #
-pypandoc==1.13
+pypandoc==1.14
     # via -r ./extra-epub.in
diff --git a/requirements/extra-odt.txt b/requirements/extra-odt.txt
@@ -8,7 +8,7 @@ lxml==5.3.0
     # via
     #   -c ./base.txt
     #   python-docx
-pypandoc==1.13
+pypandoc==1.14
     # via -r ./extra-odt.in
 python-docx==1.1.2
     # via -r ./extra-odt.in

diff --git a/requirements/extra-paddleocr.txt b/requirements/extra-paddleocr.txt
@@ -4,7 +4,7 @@
 #
 #    pip-compile ./extra-paddleocr.in
 #
-anyio==4.6.0
+anyio==4.6.2.post1
     # via
     #   -c ./base.txt
     #   httpx
@@ -16,7 +16,7 @@ certifi==2024.8.30
     #   httpcore
     #   httpx
     #   requests
-charset-normalizer==3.3.2
+charset-normalizer==3.4.0
     # via
     #   -c ./base.txt
     #   requests
@@ -52,7 +52,7 @@ idna==3.10
     #   anyio
     #   httpx
     #   requests
-imageio==2.35.1
+imageio==2.36.0
     # via
     #   imgaug
     #   scikit-image
@@ -104,7 +104,7 @@ paddlepaddle==3.0.0b1
     # via -r ./extra-paddleocr.in
 pdf2image==1.17.0
     # via unstructured-paddleocr
-pillow==10.4.0
+pillow==11.0.0
     # via
     #   imageio
     #   imgaug
@@ -117,9 +117,9 @@ protobuf==4.25.5
     # via
     #   -c ././deps/constraints.txt
     #   paddlepaddle
-pyclipper==1.3.0.post5
+pyclipper==1.3.0.post6
     # via unstructured-paddleocr
-pyparsing==3.1.4
+pyparsing==3.2.0
     # via matplotlib
 python-dateutil==2.9.0.post0
     # via

diff --git a/requirements/extra-pandoc.txt b/requirements/extra-pandoc.txt
@@ -4,5 +4,5 @@
 #
 #    pip-compile ./extra-pandoc.in
 #
-pypandoc==1.13
+pypandoc==1.14
     # via -r ./extra-pandoc.in
diff --git a/requirements/extra-pdf-image.in b/requirements/extra-pdf-image.in
@@ -11,5 +11,5 @@ google-cloud-vision
 effdet
 # Do not move to constraints.in, otherwise unstructured-inference will not be upgraded
 # when unstructured library is.
-unstructured-inference==0.7.36
+unstructured-inference==0.8.0
 unstructured.pytesseract>=0.3.12