From 79edb9455810dcfad65fb5bea05a690443ec2995 Mon Sep 17 00:00:00 2001
From: Marcel <marcel@familienarchiv>
Date: Thu, 21 May 2026 16:05:36 +0200
Subject: [PATCH] feat(ocr): increment ocr_pages_total per successful page in
 stream

Bumps the counter inside both the standard and guided /ocr/stream
generators after a page yields its blocks, before the per-page json line is
emitted. Also moves the ocr_jobs_total increment for /ocr/stream right after
engine selection so the counter still fires when a page later errors out.

Refs #652 (AC3a)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 ocr-service/main.py         |  5 +++++
 ocr-service/test_metrics.py | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 37 insertions(+)

diff --git a/ocr-service/main.py b/ocr-service/main.py
index 83a092ee..08990e24 100644
--- a/ocr-service/main.py
+++ b/ocr-service/main.py
@@ -156,6 +156,9 @@ async def run_ocr_stream(request: OcrRequest):
         )
 
     engine = kraken_engine if use_kraken else surya_engine
+    engine_name = "kraken" if use_kraken else "surya"
+
+    metrics.ocr_jobs_total.labels(engine=engine_name, script_type=script_type).inc()
 
     if request.regions:
         # Guided mode: recognize only the user-drawn annotation regions
@@ -206,6 +209,7 @@ async def run_ocr_stream(request: OcrRequest):
                         })
 
                     total_blocks += len(blocks)
+                    metrics.ocr_pages_total.labels(engine=engine_name).inc()
                     yield json.dumps({
                         "type": "page",
                         "pageNumber": page_idx,
@@ -260,6 +264,7 @@ async def run_ocr_stream(request: OcrRequest):
                         block["text"] = correct_text(block["text"])
 
                 total_blocks += len(blocks)
+                metrics.ocr_pages_total.labels(engine=engine_name).inc()
                 yield json.dumps({
                     "type": "page",
                     "pageNumber": page_idx,
diff --git a/ocr-service/test_metrics.py b/ocr-service/test_metrics.py
index f034a31c..f91bd0af 100644
--- a/ocr-service/test_metrics.py
+++ b/ocr-service/test_metrics.py
@@ -155,3 +155,35 @@ async def test_ocr_jobs_total_incremented_with_surya_engine_label_for_typewriter
         engine="surya", script_type="TYPEWRITER"
     )._value.get()
     assert value == 1.0
+
+
+@pytest.mark.asyncio
+async def test_ocr_pages_total_incremented_once_per_page_in_stream(fresh_metrics):
+    """The /ocr/stream generator increments ocr_pages_total per successful page."""
+    mock_images = [Image.new("RGB", (100, 100)) for _ in range(3)]
+    mock_blocks = [{"pageNumber": 1, "x": 0.0, "y": 0.0, "width": 1.0, "height": 1.0,
+                    "polygon": None, "text": "hi", "words": []}]
+
+    with patch("main.kraken_engine.load_models"), \
+         patch("main.load_spell_checker"), \
+         patch("main._download_and_convert_pdf", new_callable=AsyncMock, return_value=mock_images), \
+         patch("main.preprocess_page", side_effect=lambda img: img), \
+         patch("main.surya_engine.extract_page_blocks", return_value=mock_blocks):
+        async with AsyncClient(transport=ASGITransport(app=app), base_url="http://test") as client:
+            import main as main_module
+            main_module._models_ready = True
+            try:
+                async with client.stream("POST", "/ocr/stream", json={
+                    "pdfUrl": "http://minio/doc.pdf",
+                    "scriptType": "TYPEWRITER",
+                    "language": "de",
+                }) as response:
+                    assert response.status_code == 200
+                    # Drain the stream so all per-page increments fire.
+                    async for _ in response.aiter_lines():
+                        pass
+            finally:
+                main_module._models_ready = False
+
+    value = fresh_metrics.ocr_pages_total.labels(engine="surya")._value.get()
+    assert value == 3.0