Merge branch 'dev' into feature-remote-ocr-2

.github/workflows/ci.yml

@@ -17,11 +17,52 @@ env:
   DEFAULT_PYTHON_VERSION: "3.11"
   NLTK_DATA: "/usr/share/nltk_data"
 jobs:
+  detect-duplicate:
+    name: Detect Duplicate Run
+    runs-on: ubuntu-24.04
+    outputs:
+      should_run: ${{ steps.check.outputs.should_run }}
+    steps:
+      - name: Check if workflow should run
+        id: check
+        uses: actions/github-script@v7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            if (context.eventName !== 'push') {
+              core.info('Not a push event; running workflow.');
+              core.setOutput('should_run', 'true');
+              return;
+            }
+
+            const ref = context.ref || '';
+            if (!ref.startsWith('refs/heads/')) {
+              core.info('Push is not to a branch; running workflow.');
+              core.setOutput('should_run', 'true');
+              return;
+            }
+
+            const branch = ref.substring('refs/heads/'.length);
+            const { owner, repo } = context.repo;
+            const prs = await github.paginate(github.rest.pulls.list, {
+              owner,
+              repo,
+              state: 'open',
+              head: `${owner}:${branch}`,
+              per_page: 100,
+            });
+
+            if (prs.length === 0) {
+              core.info(`No open PR found for ${branch}; running workflow.`);
+              core.setOutput('should_run', 'true');
+            } else {
+              core.info(`Found ${prs.length} open PR(s) for ${branch}; skipping duplicate push run.`);
+              core.setOutput('should_run', 'false');
+            }
   pre-commit:
-    # We want to run on external PRs, but not on our own internal PRs as they'll be run
+    needs:
-    # by the push to the branch. Without this if check, checks are duplicated since
+      - detect-duplicate
-    # internal PRs match both the push and pull_request events.
+    if: needs.detect-duplicate.outputs.should_run == 'true'
-    if: github.event_name == 'push' || github.event.pull_request.head.repo.full_name != github.repository
     name: Linting Checks
     runs-on: ubuntu-24.04
     steps:

src/documents/management/commands/document_fuzzy_match.py

@@ -92,6 +92,9 @@ class Command(MultiProcessMixin, ProgressBarMixin, BaseCommand):
                 # doc to doc is obviously not useful
                 if first_doc.pk == second_doc.pk:
                     continue
+                # Skip empty documents (e.g. password-protected)
+                if first_doc.content.strip() == "" or second_doc.content.strip() == "":
+                    continue
                 # Skip matching which have already been matched together
                 # doc 1 to doc 2 is the same as doc 2 to doc 1
                 doc_1_to_doc_2 = (first_doc.pk, second_doc.pk)

src/documents/tests/test_management_fuzzy.py

@@ -206,3 +206,29 @@ class TestFuzzyMatchCommand(TestCase):
         self.assertEqual(Document.objects.count(), 2)
         self.assertIsNotNone(Document.objects.get(pk=1))
         self.assertIsNotNone(Document.objects.get(pk=2))
+
+    def test_empty_content(self):
+        """
+        GIVEN:
+            - 2 documents exist, content is empty (pw-protected)
+        WHEN:
+            - Command is called
+        THEN:
+            - No matches are found
+        """
+        Document.objects.create(
+            checksum="BEEFCAFE",
+            title="A",
+            content="",
+            mime_type="application/pdf",
+            filename="test.pdf",
+        )
+        Document.objects.create(
+            checksum="DEADBEAF",
+            title="A",
+            content="",
+            mime_type="application/pdf",
+            filename="other_test.pdf",
+        )
+        stdout, _ = self.call_command()
+        self.assertIn("No matches found", stdout)