full ingestion -> OCR -> extraction flow is now working correctly.

apps/svc_extract/main.py

@@ -64,28 +64,6 @@ Return a JSON object with the extracted fields and confidence scores.
 """
 
 
-# Create app and settings
-app, settings = create_app(
-    service_name="svc-extract",
-    title="Tax Agent Extraction Service",
-    description="LLM-based field extraction service",
-    settings_class=ExtractionSettings,
-)
-
-# Add middleware
-middleware_factory = create_trusted_proxy_middleware(settings.internal_cidrs)
-app.add_middleware(middleware_factory)
-
-# Global clients
-storage_client: StorageClient | None = None
-document_storage: DocumentStorage | None = None
-event_bus: EventBus | None = None
-confidence_calibrator: ConfidenceCalibrator | None = None
-tracer = get_tracer("svc-extract")
-metrics = get_metrics()
-
-
-@app.on_event("startup")
 async def startup_event() -> None:
     """Initialize service dependencies"""
     global storage_client, document_storage, event_bus, confidence_calibrator
@@ -116,7 +94,6 @@ async def startup_event() -> None:
     logger.info("Extraction service started successfully")
 
 
-@app.on_event("shutdown")
 async def shutdown_event() -> None:
     """Cleanup service dependencies"""
     global event_bus
@@ -129,6 +106,29 @@ async def shutdown_event() -> None:
     logger.info("Extraction service shutdown complete")
 
 
+# Create app and settings
+app, settings = create_app(
+    service_name="svc-extract",
+    title="Tax Agent Extraction Service",
+    description="LLM-based field extraction service",
+    settings_class=ExtractionSettings,
+    startup_hooks=[startup_event],
+    shutdown_hooks=[shutdown_event],
+)
+
+# Add middleware
+middleware_factory = create_trusted_proxy_middleware(settings.internal_cidrs)
+app.add_middleware(middleware_factory)
+
+# Global clients
+storage_client: StorageClient | None = None
+document_storage: DocumentStorage | None = None
+event_bus: EventBus | None = None
+confidence_calibrator: ConfidenceCalibrator | None = None
+tracer = get_tracer("svc-extract")
+metrics = get_metrics()
+
+
 @app.post("/extract/{doc_id}", response_model=ExtractionResponse)
 async def extract_fields(
     doc_id: str,
@@ -334,13 +334,14 @@ async def _extract_fields_async(
                 )
 
             # Update metrics
-            metrics.counter("extractions_completed_total").labels(
-                tenant_id=tenant_id, strategy=strategy
-            ).inc()
+            metrics.counter(
+                "extract_extractions_completed_total",
+                labelnames=["tenant_id", "strategy"],
+            ).labels(tenant_id=tenant_id, strategy=strategy).inc()
 
-            metrics.histogram("extraction_confidence").labels(
-                strategy=strategy
-            ).observe(calibrated_confidence)
+            metrics.histogram(
+                "extract_extraction_confidence", labelnames=["strategy"]
+            ).labels(strategy=strategy).observe(calibrated_confidence)
 
             # Publish completion event
             event_payload = EventPayload(
@@ -371,7 +372,10 @@ async def _extract_fields_async(
             logger.error("Field extraction failed", doc_id=doc_id, error=str(e))
 
             # Update error metrics
-            metrics.counter("extraction_errors_total").labels(
+            metrics.counter(
+                "extract_extraction_errors_total",
+                labelnames=["tenant_id", "strategy", "error_type"],
+            ).labels(
                 tenant_id=tenant_id, strategy=strategy, error_type=type(e).__name__
             ).inc()
 

apps/svc_ingestion/main.py

@@ -77,11 +77,20 @@ def init_dependencies(app_settings: IngestionSettings) -> None:
 
 
 # Create app and settings
+async def startup_event() -> None:
+    """Initialize service dependencies"""
+    if event_bus is None:
+        raise ValueError("Event bus not initialized")
+
+    await event_bus.start()
+
+
 app, _settings = create_app(
     service_name="svc-ingestion",
     title="Tax Agent Ingestion Service",
     description="Document upload and storage service",
     settings_class=IngestionSettings,
+    startup_hooks=[startup_event],
 )
 
 # Initialize dependencies immediately
@@ -158,6 +167,7 @@ async def upload_document(
             event_payload = EventPayload(
                 data={
                     "doc_id": doc_id,
+                    "tenant_id": tenant_id,
                     "filename": file.filename or "unknown",
                     "kind": kind.value,
                     "source": source,

apps/svc_ocr/Dockerfile

@@ -21,8 +21,10 @@ RUN apt-get update && apt-get install -y \
 WORKDIR /app
 
 # Copy service-specific requirements and install
+# Copy base requirements and service-specific requirements
+COPY libs/requirements-base.txt /tmp/libs-requirements.txt
 COPY apps/svc_ocr/requirements.txt /tmp/service-requirements.txt
-RUN pip install --no-cache-dir -r /tmp/service-requirements.txt
+RUN pip install --no-cache-dir -r /tmp/libs-requirements.txt -r /tmp/service-requirements.txt
 
 # Copy application code
 COPY libs/ ./libs/

apps/svc_ocr/main.py

@@ -118,7 +118,7 @@ async def init_dependencies(app_settings: OCRSettings) -> None:
             if attempt == max_retries:
                 raise HTTPException(
                     status_code=500, detail="Failed to connect to NATS after retries"
-                )
+                ) from e
             await asyncio.sleep(delay)
             delay *= 2  # exponential backoff
 
@@ -280,7 +280,7 @@ async def _handle_document_ingested(topic: str, payload: EventPayload) -> None:
         return
 
     # Auto-process PDF documents
-    if data.get("content_type") == "application/pdf":
+    if data.get("mime_type") == "application/pdf":
         logger.info("Auto-processing ingested document", doc_id=doc_id)
 
         try:
@@ -347,13 +347,13 @@ async def _process_document_async(
             await ds.store_ocr_result(tenant_id, doc_id, ocr_results)
 
             # Update metrics
-            metrics.counter("documents_processed_total").labels(
-                tenant_id=tenant_id, strategy=strategy
-            ).inc()
+            metrics.counter(
+                "ocr_documents_processed_total", labelnames=["tenant_id", "strategy"]
+            ).labels(tenant_id=tenant_id, strategy=strategy).inc()
 
-            metrics.histogram("processing_duration_seconds").labels(
-                strategy=strategy
-            ).observe(
+            metrics.histogram(
+                "ocr_processing_duration_seconds", labelnames=["strategy"]
+            ).labels(strategy=strategy).observe(
                 datetime.utcnow().timestamp()
                 - datetime.fromisoformat(
                     ocr_results["processed_at"].replace("Z", "")  # type: ignore
@@ -386,7 +386,10 @@ async def _process_document_async(
             logger.error("OCR processing failed", doc_id=doc_id, error=str(e))
 
             # Update error metrics
-            metrics.counter("processing_errors_total").labels(
+            metrics.counter(
+                "ocr_processing_errors_total",
+                labelnames=["tenant_id", "strategy", "error_type"],
+            ).labels(
                 tenant_id=tenant_id, strategy=strategy, error_type=type(e).__name__
             ).inc()