feat: add bulk download, cloud OCR, and basic OCR quality filter

Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>

tests/test_bulk_operations.py

@@ -2,6 +2,8 @@
 Tests for bulk file operations (delete and reprocess).
 """
 
+import io
+import zipfile
 from unittest.mock import MagicMock, patch
 
 import pytest
@@ -293,3 +295,239 @@ class TestStatusFilter:
         response = client.get("/files?status=failed")
         assert response.status_code == 200
         assert "failed.pdf" in response.text
+
+
+@pytest.mark.integration
+@pytest.mark.requires_db
+class TestBulkDownload:
+    """Tests for POST /api/files/bulk-download endpoint."""
+
+    def test_bulk_download_no_files_found(self, client: TestClient, db_session):
+        """Test bulk download with non-existent IDs."""
+        response = client.post("/api/files/bulk-download", json=[99999, 99998])
+        assert response.status_code == 404
+
+    def test_bulk_download_files_not_on_disk(self, client: TestClient, db_session):
+        """Test bulk download when files are not found on disk."""
+        file_record = FileRecord(
+            filehash="hash_dl1",
+            original_filename="nodisk.pdf",
+            local_filename="/nonexistent/nodisk.pdf",
+            file_size=1024,
+            mime_type="application/pdf",
+        )
+        db_session.add(file_record)
+        db_session.commit()
+
+        response = client.post("/api/files/bulk-download", json=[file_record.id])
+        assert response.status_code == 404
+        assert "None of the selected files" in response.json()["detail"]
+
+    def test_bulk_download_success(self, client: TestClient, db_session, tmp_path):
+        """Test successful bulk download returns a ZIP archive."""
+        # Create a real file on disk
+        pdf_file = tmp_path / "sample.pdf"
+        pdf_file.write_bytes(b"PDF content")
+
+        file_record = FileRecord(
+            filehash="hash_dl2",
+            original_filename="sample.pdf",
+            local_filename=str(pdf_file),
+            processed_file_path=str(pdf_file),
+            file_size=len(b"PDF content"),
+            mime_type="application/pdf",
+        )
+        db_session.add(file_record)
+        db_session.commit()
+
+        response = client.post("/api/files/bulk-download", json=[file_record.id])
+        assert response.status_code == 200
+        assert response.headers["content-type"] == "application/zip"
+        assert "attachment" in response.headers["content-disposition"]
+        assert ".zip" in response.headers["content-disposition"]
+
+    def test_bulk_download_multiple_files(self, client: TestClient, db_session, tmp_path):
+        """Test bulk download with multiple files produces a valid ZIP."""
+        ids = []
+        for i in range(3):
+            f = tmp_path / f"file{i}.pdf"
+            f.write_bytes(f"content {i}".encode())
+            rec = FileRecord(
+                filehash=f"hash_multi_{i}",
+                original_filename=f"file{i}.pdf",
+                local_filename=str(f),
+                processed_file_path=str(f),
+                file_size=len(f"content {i}".encode()),
+                mime_type="application/pdf",
+            )
+            db_session.add(rec)
+            db_session.commit()
+            ids.append(rec.id)
+
+        response = client.post("/api/files/bulk-download", json=ids)
+        assert response.status_code == 200
+
+        zip_data = io.BytesIO(response.content)
+        with zipfile.ZipFile(zip_data) as zf:
+            names = zf.namelist()
+        assert len(names) == 3
+
+    def test_bulk_download_duplicate_filenames(self, client: TestClient, db_session, tmp_path):
+        """Test bulk download disambiguates duplicate filenames."""
+        ids = []
+        for i in range(2):
+            f = tmp_path / f"dup_{i}.pdf"
+            f.write_bytes(b"data")
+            rec = FileRecord(
+                filehash=f"hash_dup_{i}",
+                original_filename="dup.pdf",  # same name
+                local_filename=str(f),
+                processed_file_path=str(f),
+                file_size=4,
+                mime_type="application/pdf",
+            )
+            db_session.add(rec)
+            db_session.commit()
+            ids.append(rec.id)
+
+        response = client.post("/api/files/bulk-download", json=ids)
+        assert response.status_code == 200
+        zip_data = io.BytesIO(response.content)
+        with zipfile.ZipFile(zip_data) as zf:
+            names = zf.namelist()
+        # Names must be unique
+        assert len(names) == len(set(names))
+
+
+@pytest.mark.integration
+@pytest.mark.requires_db
+class TestBulkReprocessCloudOcr:
+    """Tests for POST /api/files/bulk-reprocess-cloud-ocr endpoint."""
+
+    @patch("app.api.files.process_document")
+    def test_bulk_reprocess_cloud_ocr_success(self, mock_delay, client: TestClient, db_session, tmp_path):
+        """Test bulk Cloud OCR reprocessing queues tasks."""
+        mock_task = MagicMock()
+        mock_task.id = "task-cloud-ocr-1"
+        mock_delay.delay.return_value = mock_task
+
+        pdf_file = tmp_path / "ocr_test.pdf"
+        pdf_file.write_bytes(b"PDF data")
+
+        file_record = FileRecord(
+            filehash="hash_ocr1",
+            original_filename="ocr_test.pdf",
+            local_filename=str(pdf_file),
+            file_size=8,
+            mime_type="application/pdf",
+        )
+        db_session.add(file_record)
+        db_session.commit()
+
+        response = client.post("/api/files/bulk-reprocess-cloud-ocr", json=[file_record.id])
+        assert response.status_code == 200
+        data = response.json()
+        assert data["status"] == "success"
+        assert len(data["processed_files"]) == 1
+        assert data["errors"] == []
+
+        # Ensure force_cloud_ocr=True was passed
+        call_kwargs = mock_delay.delay.call_args.kwargs
+        assert call_kwargs.get("force_cloud_ocr") is True
+
+    def test_bulk_reprocess_cloud_ocr_no_file_on_disk(self, client: TestClient, db_session):
+        """Test Cloud OCR bulk reprocess skips files not on disk."""
+        file_record = FileRecord(
+            filehash="hash_ocr2",
+            original_filename="missing.pdf",
+            local_filename="/nonexistent/missing.pdf",
+            file_size=1024,
+            mime_type="application/pdf",
+        )
+        db_session.add(file_record)
+        db_session.commit()
+
+        response = client.post("/api/files/bulk-reprocess-cloud-ocr", json=[file_record.id])
+        assert response.status_code == 200
+        data = response.json()
+        assert data["status"] == "error"
+        assert len(data["errors"]) == 1
+
+    def test_bulk_reprocess_cloud_ocr_no_files_found(self, client: TestClient, db_session):
+        """Test Cloud OCR bulk reprocess with non-existent IDs."""
+        response = client.post("/api/files/bulk-reprocess-cloud-ocr", json=[99999])
+        assert response.status_code == 404
+
+
+@pytest.mark.integration
+@pytest.mark.requires_db
+class TestOcrQualityFilter:
+    """Tests for the ocr_quality filter on the /files view."""
+
+    def test_ocr_quality_no_ocr_filter(self, client: TestClient, db_session):
+        """Files without OCR text appear when filtering no_ocr."""
+        rec_no_ocr = FileRecord(
+            filehash="hash_noocr1",
+            original_filename="no_ocr.pdf",
+            local_filename="/tmp/no_ocr.pdf",
+            file_size=1024,
+            mime_type="application/pdf",
+            ocr_text=None,
+        )
+        rec_has_ocr = FileRecord(
+            filehash="hash_hasocr1",
+            original_filename="has_ocr.pdf",
+            local_filename="/tmp/has_ocr.pdf",
+            file_size=1024,
+            mime_type="application/pdf",
+            ocr_text="Some extracted text",
+        )
+        db_session.add_all([rec_no_ocr, rec_has_ocr])
+        db_session.commit()
+
+        response = client.get("/files?ocr_quality=no_ocr")
+        assert response.status_code == 200
+        assert "no_ocr.pdf" in response.text
+        assert "has_ocr.pdf" not in response.text
+
+    def test_ocr_quality_has_ocr_filter(self, client: TestClient, db_session):
+        """Files with OCR text appear when filtering has_ocr."""
+        rec_no_ocr = FileRecord(
+            filehash="hash_noocr2",
+            original_filename="no_ocr2.pdf",
+            local_filename="/tmp/no_ocr2.pdf",
+            file_size=1024,
+            mime_type="application/pdf",
+            ocr_text=None,
+        )
+        rec_has_ocr = FileRecord(
+            filehash="hash_hasocr2",
+            original_filename="has_ocr2.pdf",
+            local_filename="/tmp/has_ocr2.pdf",
+            file_size=1024,
+            mime_type="application/pdf",
+            ocr_text="Meaningful extracted text",
+        )
+        db_session.add_all([rec_no_ocr, rec_has_ocr])
+        db_session.commit()
+
+        response = client.get("/files?ocr_quality=has_ocr")
+        assert response.status_code == 200
+        assert "has_ocr2.pdf" in response.text
+        assert "no_ocr2.pdf" not in response.text
+
+    def test_ocr_quality_no_filter(self, client: TestClient, db_session):
+        """All files appear when no ocr_quality filter is applied."""
+        rec = FileRecord(
+            filehash="hash_all1",
+            original_filename="all_files.pdf",
+            local_filename="/tmp/all_files.pdf",
+            file_size=1024,
+            mime_type="application/pdf",
+        )
+        db_session.add(rec)
+        db_session.commit()
+
+        response = client.get("/files")
+        assert response.status_code == 200
+        assert "all_files.pdf" in response.text