ci: simplify ci code

Unstructured-IO · Oct 19, 2024 · 2feeea5 · 2feeea5
1 parent 09c9079
commit 2feeea5
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 11 deletions.
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 ${{ inputs.python-version }}
       if: steps.ingest-virtualenv-cache-restore.outputs.cache-hit != 'true'
@@ -40,6 +40,7 @@ runs:
         fi
         make install-ci
         make install-nltk-models
+        make install-all-docs
         make install-ingest
     - name: Save Ingest Cache
       if: steps.ingest-virtualenv-cache-restore.outputs.cache-hit != 'true'
@@ -49,5 +50,5 @@ runs:
         path: |
           .venv
           nltk_data
-        key: unstructured-ingest-${{ 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
@@ -79,8 +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:
@@ -186,8 +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
@@ -229,8 +225,6 @@ jobs:
       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
@@ -309,7 +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
@@ -333,8 +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