fix(ocr): use 1-based page numbers to match frontend PDF viewer

The PDF viewer uses 1-based currentPage (starting at 1) but the OCR
engines produced 0-based pageNumber from enumerate(). Annotations
created by OCR were assigned to page 0, which doesn't exist in the
viewer. Change enumerate() to start=1 in both engines and the
streaming endpoint.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

ocr-service/engines/kraken.py

@@ -88,7 +88,7 @@ def extract_blocks(images: list, language: str = "de") -> list[dict]:
     """
     all_blocks = []
 
-    for page_idx, image in enumerate(images):
+    for page_idx, image in enumerate(images, start=1):
         all_blocks.extend(extract_page_blocks(image, page_idx, language))
 
     return all_blocks

ocr-service/engines/surya.py

@@ -90,7 +90,7 @@ def extract_blocks(images: list, language: str = "de") -> list[dict]:
     """
     all_blocks = []
 
-    for page_idx, image in enumerate(images):
+    for page_idx, image in enumerate(images, start=1):
         all_blocks.extend(extract_page_blocks(image, page_idx, language))
         del image
 

ocr-service/main.py

@@ -108,7 +108,7 @@ async def run_ocr_stream(request: OcrRequest):
         total_blocks = 0
         skipped_pages = 0
 
-        for page_idx, image in enumerate(images):
+        for page_idx, image in enumerate(images, start=1):
             try:
                 engine = kraken_engine if use_kraken else surya_engine
                 blocks = await asyncio.to_thread(

ocr-service/test_engines.py

@@ -61,8 +61,8 @@ def test_surya_extract_blocks_delegates_to_extract_page_blocks():
         blocks = surya.extract_blocks([image1, image2])
 
     assert len(blocks) == 2
-    assert blocks[0]["pageNumber"] == 0
-    assert blocks[1]["pageNumber"] == 1
+    assert blocks[0]["pageNumber"] == 1
+    assert blocks[1]["pageNumber"] == 2
 
 
 # ─── Kraken extract_page_blocks ──────────────────────────────────────────────
@@ -128,5 +128,5 @@ def test_kraken_extract_blocks_delegates_to_extract_page_blocks():
     blocks = _run_kraken_with_mocks(kraken.extract_blocks, [image1, image2])
 
     assert len(blocks) == 2
-    assert blocks[0]["pageNumber"] == 0
-    assert blocks[1]["pageNumber"] == 1
+    assert blocks[0]["pageNumber"] == 1
+    assert blocks[1]["pageNumber"] == 2

ocr-service/test_stream.py

@@ -68,9 +68,9 @@ async def test_ocr_stream_emits_page_event_per_page_with_blocks(mock_images):
         lines = [json.loads(line) for line in response.text.strip().split("\n") if line.strip()]
         page_events = [l for l in lines if l["type"] == "page"]
         assert len(page_events) == 3
-        assert page_events[0]["pageNumber"] == 0
-        assert page_events[1]["pageNumber"] == 1
-        assert page_events[2]["pageNumber"] == 2
+        assert page_events[0]["pageNumber"] == 1
+        assert page_events[1]["pageNumber"] == 2
+        assert page_events[2]["pageNumber"] == 3
         assert len(page_events[0]["blocks"]) == 1
 
 
@@ -109,8 +109,8 @@ async def test_ocr_stream_emits_error_event_on_page_failure_and_continues(mock_i
     def page_blocks(image, page_idx, language="de"):
         nonlocal call_count
         call_count += 1
-        if page_idx == 1:
-            raise RuntimeError("Engine crashed on page 1")
+        if page_idx == 2:
+            raise RuntimeError("Engine crashed on page 2")
         return [_make_block(page_idx)]
 
     with patch("main._download_and_convert_pdf", new_callable=AsyncMock, return_value=mock_images), \
@@ -128,7 +128,7 @@ async def test_ocr_stream_emits_error_event_on_page_failure_and_continues(mock_i
         types = [l["type"] for l in lines]
         assert "error" in types
         error_event = [l for l in lines if l["type"] == "error"][0]
-        assert error_event["pageNumber"] == 1
+        assert error_event["pageNumber"] == 2
         # Error message must be generic, not the raw traceback
         assert "Engine crashed" not in error_event["message"]
 
@@ -148,7 +148,7 @@ async def test_old_ocr_endpoint_still_returns_flat_list(mock_images):
     with patch("main._download_and_convert_pdf", new_callable=AsyncMock, return_value=mock_images), \
          patch("main._models_ready", True), \
          patch("main.surya_engine") as mock_surya:
-        mock_surya.extract_blocks.return_value = [_make_block(0), _make_block(1)]
+        mock_surya.extract_blocks.return_value = [_make_block(1), _make_block(2)]
 
         async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
             response = await client.post("/ocr", json={
@@ -160,4 +160,4 @@ async def test_old_ocr_endpoint_still_returns_flat_list(mock_images):
         data = response.json()
         assert isinstance(data, list)
         assert len(data) == 2
-        assert data[0]["pageNumber"] == 0
+        assert data[0]["pageNumber"] == 1