fix: merge main, address code review feedback for security fix PR #816

- Merge origin/main into branch (resolve conflict in integrations_dashboard.html)
- Add defensive JSON parsing with try/except for integration.config
- Wrap tester() call in try/except to prevent 500 errors from bad config
- Add i18n key integrations.connection_test_failed_fallback in en.json
- Reference i18n key in template JS fallback message
- Update SECURITY_AUDIT.md: add fix date (2026-03-23), update doc date
- Remove accidental revert.sh file
- Fix missing MagicMock/patch imports in test file
- Add tests for invalid JSON config and tester exception error paths

Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Agent-Logs-Url: https://github.com/christianlouis/DocuElevate/sessions/daebb70e-059a-4601-8864-88eef49f99cf

tests/test_upload_email.py

@@ -121,6 +121,95 @@ class TestExtractMetadataFromFile:
 
         assert result == {}
 
+    def test_extract_metadata_from_pdf(self, tmp_path):
+        """Test extracting metadata from a PDF file using pypdf when JSON is missing."""
+        import pypdf
+
+        file_path = tmp_path / "test.pdf"
+
+        # Create a test PDF with metadata
+        writer = pypdf.PdfWriter()
+        writer.add_blank_page(width=100, height=100)
+        writer.add_metadata(
+            {
+                "/Title": "Test Title",
+                "/Author": "Test Author",
+                "/Subject": "Test Document",
+                "/Keywords": "test, metadata, pypdf",
+            }
+        )
+        with open(file_path, "wb") as f:
+            writer.write(f)
+
+        result = extract_metadata_from_file(str(file_path))
+
+        # Keys are mapped to application-specific names
+        assert result.get("filename") == "Test Title"
+        assert result.get("absender") == "Test Author"
+        assert result.get("document_type") == "Test Document"
+        assert result.get("tags") == "test, metadata, pypdf"
+
+    def test_extracts_embedded_metadata_from_pdf(self, tmp_path):
+        """Test that embedded PDF metadata is mapped to application-specific keys."""
+        import pypdf
+
+        file_path = tmp_path / "mapped.pdf"
+
+        writer = pypdf.PdfWriter()
+        writer.add_blank_page(width=100, height=100)
+        writer.add_metadata(
+            {
+                "/Title": "Invoice 2024",
+                "/Author": "Acme Corp",
+                "/Subject": "invoice",
+                "/Keywords": "finance, billing",
+            }
+        )
+        with open(file_path, "wb") as f:
+            writer.write(f)
+
+        result = extract_metadata_from_file(str(file_path))
+
+        # Verify the PDF-to-app key mapping
+        assert result["filename"] == "Invoice 2024"
+        assert result["absender"] == "Acme Corp"
+        assert result["document_type"] == "invoice"
+        assert result["tags"] == "finance, billing"
+
+    def test_pdf_metadata_does_not_overwrite_json(self, tmp_path):
+        """Test that JSON metadata takes precedence over embedded PDF metadata."""
+        import pypdf
+
+        file_path = tmp_path / "dual.pdf"
+
+        # Create a PDF with embedded metadata
+        writer = pypdf.PdfWriter()
+        writer.add_blank_page(width=100, height=100)
+        writer.add_metadata(
+            {
+                "/Title": "PDF Title",
+                "/Author": "PDF Author",
+                "/Subject": "PDF Subject",
+                "/Keywords": "pdf, keywords",
+            }
+        )
+        with open(file_path, "wb") as f:
+            writer.write(f)
+
+        # Create a companion JSON file that sets some overlapping fields
+        json_metadata = {"filename": "JSON Filename", "absender": "JSON Author"}
+        json_path = tmp_path / "dual.json"
+        json_path.write_text(json.dumps(json_metadata))
+
+        result = extract_metadata_from_file(str(file_path))
+
+        # JSON values must not be overwritten by PDF metadata
+        assert result["filename"] == "JSON Filename"
+        assert result["absender"] == "JSON Author"
+        # Fields missing from JSON are filled from PDF metadata
+        assert result["document_type"] == "PDF Subject"
+        assert result["tags"] == "pdf, keywords"
+
 
 @pytest.mark.unit
 class TestAttachLogo: