nv-ingest 2025.7.15.dev20250715__tar.gz → 2025.7.17.dev20250717__tar.gz

{nv_ingest-2025.7.15.dev20250715 → nv_ingest-2025.7.17.dev20250717}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nv-ingest
-Version: 2025.7.15.dev20250715
+Version: 2025.7.17.dev20250717
 Summary: Python module for multimodal document ingestion
 Author-email: Jeremy Dyer <jdyer@nvidia.com>
 License:                                  Apache License

{nv_ingest-2025.7.15.dev20250715 → nv_ingest-2025.7.17.dev20250717}/nv_ingest/api/v1/health.py

@@ -18,7 +18,7 @@ router = APIRouter()
 
 # List of ALL of the HTTP environment variable endpoints that should be checked
 READY_CHECK_ENV_VAR_MAP = {
-    "paddle": "PADDLE_HTTP_ENDPOINT",
+    "ocr": "OCR_HTTP_ENDPOINT",
     "yolox_graphic_elements": "YOLOX_GRAPHIC_ELEMENTS_HTTP_ENDPOINT",
     "yolox_page_elements": "YOLOX_HTTP_ENDPOINT",
     "yolox_table_structure": "YOLOX_TABLE_STRUCTURE_HTTP_ENDPOINT",

{nv_ingest-2025.7.15.dev20250715 → nv_ingest-2025.7.17.dev20250717}/nv_ingest/framework/orchestration/ray/examples/pipeline_test_harness.py

@@ -147,8 +147,9 @@ if __name__ == "__main__":
     os.environ["YOLOX_GRAPHIC_ELEMENTS_GRPC_ENDPOINT"] = "127.0.0.1:8004"
     os.environ["YOLOX_GRAPHIC_ELEMENTS_HTTP_ENDPOINT"] = "http://localhost:8003/v1/infer"
     os.environ["YOLOX_GRAPHIC_ELEMENTS_INFER_PROTOCOL"] = "http"
-    os.environ["PADDLE_GRPC_ENDPOINT"] = "localhost:8010"
-    os.environ["PADDLE_INFER_PROTOCOL"] = "grpc"
+    os.environ["OCR_GRPC_ENDPOINT"] = "localhost:8010"
+    os.environ["OCR_INFER_PROTOCOL"] = "grpc"
+    os.environ["OCR_MODEL_NAME"] = "paddle"
     os.environ["NEMORETRIEVER_PARSE_HTTP_ENDPOINT"] = "https://integrate.api.nvidia.com/v1/chat/completions"
     os.environ["VLM_CAPTION_ENDPOINT"] = "https://integrate.api.nvidia.com/v1/chat/completions"
     os.environ["VLM_CAPTION_MODEL_NAME"] = "nvidia/llama-3.1-nemotron-nano-vl-8b-v1"
@@ -172,7 +173,7 @@ if __name__ == "__main__":
     nemoretriever_parse_grpc, nemoretriever_parse_http, nemoretriever_parse_auth, nemoretriever_parse_protocol = (
         get_nim_service("nemoretriever_parse")
     )
-    paddle_grpc, paddle_http, paddle_auth, paddle_protocol = get_nim_service("paddle")
+    ocr_grpc, ocr_http, ocr_auth, ocr_protocol = get_nim_service("ocr")
 
     model_name = os.environ.get("NEMORETRIEVER_PARSE_MODEL_NAME", "nvidia/nemoretriever-parse")
     pdf_extractor_config = {
@@ -201,8 +202,8 @@ if __name__ == "__main__":
         "endpoint_config": {
             "yolox_endpoints": (yolox_graphic_elements_grpc, yolox_graphic_elements_http),
             "yolox_infer_protocol": yolox_graphic_elements_protocol,
-            "paddle_endpoints": (paddle_grpc, paddle_http),
-            "paddle_infer_protocol": paddle_protocol,
+            "ocr_endpoints": (ocr_grpc, ocr_http),
+            "ocr_infer_protocol": ocr_protocol,
             "auth_token": yolox_auth,
         }
     }
@@ -210,8 +211,8 @@ if __name__ == "__main__":
         "endpoint_config": {
             "yolox_endpoints": (yolox_table_structure_grpc, yolox_table_structure_http),
             "yolox_infer_protocol": yolox_table_structure_protocol,
-            "paddle_endpoints": (paddle_grpc, paddle_http),
-            "paddle_infer_protocol": paddle_protocol,
+            "ocr_endpoints": (ocr_grpc, ocr_http),
+            "ocr_infer_protocol": ocr_protocol,
             "auth_token": yolox_auth,
         }
     }

{nv_ingest-2025.7.15.dev20250715 → nv_ingest-2025.7.17.dev20250717}/nv_ingest/framework/orchestration/ray/stages/sources/message_broker_task_source.py

@@ -269,8 +269,11 @@ class MessageBrokerTaskSourceStage(RayActorSourceStage):
             self._logger.debug("Received message type: %s", type(job))
             if isinstance(job, BaseModel):
                 self._logger.debug("Message is a BaseModel with response_code: %s", job.response_code)
-                if job.response_code != 0:
-                    self._logger.debug("Message response_code != 0, returning None")
+                if job.response_code not in (0, 2):
+                    self._logger.debug("Message received with unhandled response_code, returning None")
+                    return None
+                if job.response_code == 2:
+                    self._logger.debug("Message response_code == 2, returning None")
                     return None
                 job = json.loads(job.response)
             self._logger.debug("Successfully fetched message with job_id: %s", job.get("job_id", "unknown"))

{nv_ingest-2025.7.15.dev20250715 → nv_ingest-2025.7.17.dev20250717}/nv_ingest/framework/orchestration/ray/util/pipeline/pipeline_runners.py

@@ -78,8 +78,9 @@ class PipelineCreationSchema(BaseModel):
     otel_exporter_otlp_endpoint: str = os.getenv("OTEL_EXPORTER_OTLP_ENDPOINT", "localhost:4317")
 
     # OCR settings
-    paddle_http_endpoint: str = os.getenv("PADDLE_HTTP_ENDPOINT", "https://ai.api.nvidia.com/v1/cv/baidu/paddleocr")
-    paddle_infer_protocol: str = os.getenv("PADDLE_INFER_PROTOCOL", "http")
+    ocr_http_endpoint: str = os.getenv("OCR_HTTP_ENDPOINT", "https://ai.api.nvidia.com/v1/cv/baidu/paddleocr")
+    ocr_infer_protocol: str = os.getenv("OCR_INFER_PROTOCOL", "http")
+    ocr_model_name: str = os.getenv("OCR_MODEL_NAME", "paddle")
 
     # Task queue settings
     REDIS_INGEST_TASK_QUEUE: str = "ingest_task_queue"

{nv_ingest-2025.7.15.dev20250715 → nv_ingest-2025.7.17.dev20250717}/nv_ingest/framework/orchestration/ray/util/pipeline/stage_builders.py

@@ -223,15 +223,15 @@ def add_table_extractor_stage(pipeline, default_cpu_count, stage_name="table_ext
     yolox_table_structure_grpc, yolox_table_structure_http, yolox_auth, yolox_table_structure_protocol = (
         get_nim_service("yolox_table_structure")
     )
-    paddle_grpc, paddle_http, paddle_auth, paddle_protocol = get_nim_service("paddle")
+    ocr_grpc, ocr_http, ocr_auth, ocr_protocol = get_nim_service("ocr")
 
     table_extractor_config = TableExtractorSchema(
         **{
             "endpoint_config": {
                 "yolox_endpoints": (yolox_table_structure_grpc, yolox_table_structure_http),
                 "yolox_infer_protocol": yolox_table_structure_protocol,
-                "paddle_endpoints": (paddle_grpc, paddle_http),
-                "paddle_infer_protocol": paddle_protocol,
+                "ocr_endpoints": (ocr_grpc, ocr_http),
+                "ocr_infer_protocol": ocr_protocol,
                 "auth_token": yolox_auth,
             }
         }
@@ -252,15 +252,15 @@ def add_chart_extractor_stage(pipeline, default_cpu_count, stage_name="chart_ext
     yolox_graphic_elements_grpc, yolox_graphic_elements_http, yolox_auth, yolox_graphic_elements_protocol = (
         get_nim_service("yolox_graphic_elements")
     )
-    paddle_grpc, paddle_http, paddle_auth, paddle_protocol = get_nim_service("paddle")
+    ocr_grpc, ocr_http, ocr_auth, ocr_protocol = get_nim_service("ocr")
 
     chart_extractor_config = ChartExtractorSchema(
         **{
             "endpoint_config": {
                 "yolox_endpoints": (yolox_graphic_elements_grpc, yolox_graphic_elements_http),
                 "yolox_infer_protocol": yolox_graphic_elements_protocol,
-                "paddle_endpoints": (paddle_grpc, paddle_http),
-                "paddle_infer_protocol": paddle_protocol,
+                "ocr_endpoints": (ocr_grpc, ocr_http),
+                "ocr_infer_protocol": ocr_protocol,
                 "auth_token": yolox_auth,
             }
         }
@@ -278,14 +278,14 @@ def add_chart_extractor_stage(pipeline, default_cpu_count, stage_name="chart_ext
 
 
 def add_infographic_extractor_stage(pipeline, default_cpu_count, stage_name="infographic_extractor"):
-    paddle_grpc, paddle_http, paddle_auth, paddle_protocol = get_nim_service("paddle")
+    ocr_grpc, ocr_http, ocr_auth, ocr_protocol = get_nim_service("ocr")
 
     infographic_content_extractor_config = InfographicExtractorSchema(
         **{
             "endpoint_config": {
-                "paddle_endpoints": (paddle_grpc, paddle_http),
-                "paddle_infer_protocol": paddle_protocol,
-                "auth_token": paddle_auth,
+                "ocr_endpoints": (ocr_grpc, ocr_http),
+                "ocr_infer_protocol": ocr_protocol,
+                "auth_token": ocr_auth,
             }
         }
     )

{nv_ingest-2025.7.15.dev20250715 → nv_ingest-2025.7.17.dev20250717}/nv_ingest.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nv-ingest
-Version: 2025.7.15.dev20250715
+Version: 2025.7.17.dev20250717
 Summary: Python module for multimodal document ingestion
 Author-email: Jeremy Dyer <jdyer@nvidia.com>
 License:                                  Apache License