From 23289e028f9ba0a063f10646bb0ee543cb8dab2c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Marek=20Po=C5=82om?= <marek.polom@deepsense.ai>
Date: Tue, 29 Apr 2025 13:36:58 +0200
Subject: [PATCH 1/3] Run local ingest test instead of wikipedia

---
 scripts/docker-smoke-test.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/docker-smoke-test.sh b/scripts/docker-smoke-test.sh
index 1d0950e923..0e66e05ae4 100755
--- a/scripts/docker-smoke-test.sh
+++ b/scripts/docker-smoke-test.sh
@@ -41,7 +41,7 @@ await_container
 docker cp test_unstructured_ingest $CONTAINER_NAME:/app
 docker cp requirements/ingest $CONTAINER_NAME:/app/requirements/ingest
 docker exec -u root "$CONTAINER_NAME" /bin/bash -c "chown -R notebook-user:notebook-user /app/test_unstructured_ingest"
-docker exec "$CONTAINER_NAME" /bin/bash -c "/app/test_unstructured_ingest/src/wikipedia.sh"
+docker exec "$CONTAINER_NAME" /bin/bash -c "/app/test_unstructured_ingest/src/local.sh"
 
 result=$?
 exit $result

From 85b628083c0c7ba027865302c88fbd32c2f431e7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Marek=20Po=C5=82om?= <marek.polom@deepsense.ai>
Date: Tue, 29 Apr 2025 13:37:26 +0200
Subject: [PATCH 2/3] Test image build

---
 .github/workflows/docker-publish.yml | 91 ++++++++++++++--------------
 1 file changed, 47 insertions(+), 44 deletions(-)

diff --git a/.github/workflows/docker-publish.yml b/.github/workflows/docker-publish.yml
index 8705bb39b8..ebe097e711 100644
--- a/.github/workflows/docker-publish.yml
+++ b/.github/workflows/docker-publish.yml
@@ -2,8 +2,11 @@ name: Build And Push Docker Image
 
 on:
   push:
-    branches:
-      - main
+    branches: [ main ]
+  pull_request:
+    branches: [ main ]
+  merge_group:
+    branches: [ main ]
 
 env:
   DOCKER_REPOSITORY: quay.io/unstructured-io/unstructured
@@ -70,45 +73,45 @@ jobs:
             make docker-test CI=true TEST_FILE=test_unstructured/partition/test_text.py
         fi
         DOCKER_IMAGE=$DOCKER_BUILD_REPOSITORY:$ARCH-$SHORT_SHA make docker-smoke-test
-    - name: Push images
-      run: |
-        # write to the build repository to cache for the publish-images job
-        ARCH=$(cut -d "/" -f2 <<< ${{ matrix.docker-platform }})
-        docker push "$DOCKER_BUILD_REPOSITORY:$ARCH-$SHORT_SHA"
-  publish-images:
-    runs-on: ubuntu-latest-m
-    needs: [set-short-sha, build-images]
-    env:
-      SHORT_SHA: ${{ needs.set-short-sha.outputs.short_sha }}
-    steps:
-    - uses: docker/setup-buildx-action@v1
-    - name: Checkout code
-      uses: actions/checkout@v3
-    - name: Login to Quay.io
-      uses: docker/login-action@v1
-      with:
-        registry: quay.io
-        username: ${{ secrets.QUAY_IO_ROBOT_USERNAME }}
-        password: ${{ secrets.QUAY_IO_ROBOT_TOKEN }}
-    - name: Pull AMD image
-      run: |
-        docker pull $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA
-    - name: Pull ARM image
-      run: |
-        docker pull $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA
-    - name: Push latest build tags for AMD and ARM
-      run: |
-        # these are used to construct the final manifest but also cache-from in subsequent runs
-        docker tag $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64
-        docker push $DOCKER_BUILD_REPOSITORY:amd64
-        docker tag $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:arm64
-        docker push $DOCKER_BUILD_REPOSITORY:arm64
-    - name: Push multiarch manifest
-      run: |
-        docker manifest create ${DOCKER_REPOSITORY}:latest $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
-        docker manifest push $DOCKER_REPOSITORY:latest
-        docker manifest create ${DOCKER_REPOSITORY}:$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
-        docker manifest push $DOCKER_REPOSITORY:$SHORT_SHA
-        VERSION=$(grep -Po '(?<=__version__ = ")[^"]*' unstructured/__version__.py)
-        docker manifest create ${DOCKER_REPOSITORY}:$VERSION $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
-        docker manifest push $DOCKER_REPOSITORY:$VERSION
+    # - name: Push images
+    #   run: |
+    #     # write to the build repository to cache for the publish-images job
+    #     ARCH=$(cut -d "/" -f2 <<< ${{ matrix.docker-platform }})
+    #     docker push "$DOCKER_BUILD_REPOSITORY:$ARCH-$SHORT_SHA"
+  # publish-images:
+  #   runs-on: ubuntu-latest-m
+  #   needs: [set-short-sha, build-images]
+  #   env:
+  #     SHORT_SHA: ${{ needs.set-short-sha.outputs.short_sha }}
+  #   steps:
+  #   - uses: docker/setup-buildx-action@v1
+  #   - name: Checkout code
+  #     uses: actions/checkout@v3
+  #   - name: Login to Quay.io
+  #     uses: docker/login-action@v1
+  #     with:
+  #       registry: quay.io
+  #       username: ${{ secrets.QUAY_IO_ROBOT_USERNAME }}
+  #       password: ${{ secrets.QUAY_IO_ROBOT_TOKEN }}
+  #   - name: Pull AMD image
+  #     run: |
+  #       docker pull $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA
+  #   - name: Pull ARM image
+  #     run: |
+  #       docker pull $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA
+  #   - name: Push latest build tags for AMD and ARM
+  #     run: |
+  #       # these are used to construct the final manifest but also cache-from in subsequent runs
+  #       docker tag $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64
+  #       docker push $DOCKER_BUILD_REPOSITORY:amd64
+  #       docker tag $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:arm64
+  #       docker push $DOCKER_BUILD_REPOSITORY:arm64
+  #   - name: Push multiarch manifest
+  #     run: |
+  #       docker manifest create ${DOCKER_REPOSITORY}:latest $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
+  #       docker manifest push $DOCKER_REPOSITORY:latest
+  #       docker manifest create ${DOCKER_REPOSITORY}:$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
+  #       docker manifest push $DOCKER_REPOSITORY:$SHORT_SHA
+  #       VERSION=$(grep -Po '(?<=__version__ = ")[^"]*' unstructured/__version__.py)
+  #       docker manifest create ${DOCKER_REPOSITORY}:$VERSION $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
+  #       docker manifest push $DOCKER_REPOSITORY:$VERSION

From 67a54fd5d52a84a450d187aa82063c2bbecd1083 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Marek=20Po=C5=82om?= <marek.polom@deepsense.ai>
Date: Tue, 29 Apr 2025 14:41:36 +0200
Subject: [PATCH 3/3] Revert "Test image build"

This reverts commit 85b628083c0c7ba027865302c88fbd32c2f431e7.
---
 .github/workflows/docker-publish.yml | 91 ++++++++++++++--------------
 1 file changed, 44 insertions(+), 47 deletions(-)

diff --git a/.github/workflows/docker-publish.yml b/.github/workflows/docker-publish.yml
index ebe097e711..8705bb39b8 100644
--- a/.github/workflows/docker-publish.yml
+++ b/.github/workflows/docker-publish.yml
@@ -2,11 +2,8 @@ name: Build And Push Docker Image
 
 on:
   push:
-    branches: [ main ]
-  pull_request:
-    branches: [ main ]
-  merge_group:
-    branches: [ main ]
+    branches:
+      - main
 
 env:
   DOCKER_REPOSITORY: quay.io/unstructured-io/unstructured
@@ -73,45 +70,45 @@ jobs:
             make docker-test CI=true TEST_FILE=test_unstructured/partition/test_text.py
         fi
         DOCKER_IMAGE=$DOCKER_BUILD_REPOSITORY:$ARCH-$SHORT_SHA make docker-smoke-test
-    # - name: Push images
-    #   run: |
-    #     # write to the build repository to cache for the publish-images job
-    #     ARCH=$(cut -d "/" -f2 <<< ${{ matrix.docker-platform }})
-    #     docker push "$DOCKER_BUILD_REPOSITORY:$ARCH-$SHORT_SHA"
-  # publish-images:
-  #   runs-on: ubuntu-latest-m
-  #   needs: [set-short-sha, build-images]
-  #   env:
-  #     SHORT_SHA: ${{ needs.set-short-sha.outputs.short_sha }}
-  #   steps:
-  #   - uses: docker/setup-buildx-action@v1
-  #   - name: Checkout code
-  #     uses: actions/checkout@v3
-  #   - name: Login to Quay.io
-  #     uses: docker/login-action@v1
-  #     with:
-  #       registry: quay.io
-  #       username: ${{ secrets.QUAY_IO_ROBOT_USERNAME }}
-  #       password: ${{ secrets.QUAY_IO_ROBOT_TOKEN }}
-  #   - name: Pull AMD image
-  #     run: |
-  #       docker pull $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA
-  #   - name: Pull ARM image
-  #     run: |
-  #       docker pull $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA
-  #   - name: Push latest build tags for AMD and ARM
-  #     run: |
-  #       # these are used to construct the final manifest but also cache-from in subsequent runs
-  #       docker tag $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64
-  #       docker push $DOCKER_BUILD_REPOSITORY:amd64
-  #       docker tag $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:arm64
-  #       docker push $DOCKER_BUILD_REPOSITORY:arm64
-  #   - name: Push multiarch manifest
-  #     run: |
-  #       docker manifest create ${DOCKER_REPOSITORY}:latest $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
-  #       docker manifest push $DOCKER_REPOSITORY:latest
-  #       docker manifest create ${DOCKER_REPOSITORY}:$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
-  #       docker manifest push $DOCKER_REPOSITORY:$SHORT_SHA
-  #       VERSION=$(grep -Po '(?<=__version__ = ")[^"]*' unstructured/__version__.py)
-  #       docker manifest create ${DOCKER_REPOSITORY}:$VERSION $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
-  #       docker manifest push $DOCKER_REPOSITORY:$VERSION
+    - name: Push images
+      run: |
+        # write to the build repository to cache for the publish-images job
+        ARCH=$(cut -d "/" -f2 <<< ${{ matrix.docker-platform }})
+        docker push "$DOCKER_BUILD_REPOSITORY:$ARCH-$SHORT_SHA"
+  publish-images:
+    runs-on: ubuntu-latest-m
+    needs: [set-short-sha, build-images]
+    env:
+      SHORT_SHA: ${{ needs.set-short-sha.outputs.short_sha }}
+    steps:
+    - uses: docker/setup-buildx-action@v1
+    - name: Checkout code
+      uses: actions/checkout@v3
+    - name: Login to Quay.io
+      uses: docker/login-action@v1
+      with:
+        registry: quay.io
+        username: ${{ secrets.QUAY_IO_ROBOT_USERNAME }}
+        password: ${{ secrets.QUAY_IO_ROBOT_TOKEN }}
+    - name: Pull AMD image
+      run: |
+        docker pull $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA
+    - name: Pull ARM image
+      run: |
+        docker pull $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA
+    - name: Push latest build tags for AMD and ARM
+      run: |
+        # these are used to construct the final manifest but also cache-from in subsequent runs
+        docker tag $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64
+        docker push $DOCKER_BUILD_REPOSITORY:amd64
+        docker tag $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:arm64
+        docker push $DOCKER_BUILD_REPOSITORY:arm64
+    - name: Push multiarch manifest
+      run: |
+        docker manifest create ${DOCKER_REPOSITORY}:latest $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
+        docker manifest push $DOCKER_REPOSITORY:latest
+        docker manifest create ${DOCKER_REPOSITORY}:$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
+        docker manifest push $DOCKER_REPOSITORY:$SHORT_SHA
+        VERSION=$(grep -Po '(?<=__version__ = ")[^"]*' unstructured/__version__.py)
+        docker manifest create ${DOCKER_REPOSITORY}:$VERSION $DOCKER_BUILD_REPOSITORY:amd64 $DOCKER_BUILD_REPOSITORY:arm64
+        docker manifest push $DOCKER_REPOSITORY:$VERSION