nv-ingest 2026.1.4.dev20260104__tar.gz → 2026.1.5.dev20260105__tar.gz

{nv_ingest-2026.1.4.dev20260104 → nv_ingest-2026.1.5.dev20260105}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nv-ingest
-Version: 2026.1.4.dev20260104
+Version: 2026.1.5.dev20260105
 Summary: Python module for multimodal document ingestion
 Author-email: Jeremy Dyer <jdyer@nvidia.com>
 License:                                  Apache License
@@ -227,7 +227,7 @@ Requires-Dist: httpx>=0.28.1
 Requires-Dist: isodate>=0.7.2
 Requires-Dist: langdetect>=1.0.9
 Requires-Dist: minio>=7.2.12
-Requires-Dist: librosa>=0.10.2
+Requires-Dist: librosa==0.10.2
 Requires-Dist: opentelemetry-api>=1.27.0
 Requires-Dist: opentelemetry-exporter-otlp>=1.27.0
 Requires-Dist: opentelemetry-sdk>=1.27.0

{nv_ingest-2026.1.4.dev20260104 → nv_ingest-2026.1.5.dev20260105}/nv_ingest/api/v2/ingest.py

@@ -122,11 +122,16 @@ def get_pdf_split_page_count(client_override: Optional[int] = None) -> int:
         )
         return DEFAULT_PDF_SPLIT_PAGE_COUNT
 
-    if parsed <= 0:
-        logger.warning("PDF_SPLIT_PAGE_COUNT must be >= 1; received %s. Using 1.", parsed)
-        return 1
-
-    return parsed
+    clamped = max(MIN_PAGES, min(parsed, MAX_PAGES))
+    if clamped != parsed:
+        logger.warning(
+            "Env PDF_SPLIT_PAGE_COUNT=%s clamped to %s (min=%s, max=%s)",
+            parsed,
+            clamped,
+            MIN_PAGES,
+            MAX_PAGES,
+        )
+    return clamped
 
 
 def split_pdf_to_chunks(pdf_content: bytes, pages_per_chunk: int) -> List[Dict[str, Any]]:
@@ -955,7 +960,7 @@ async def submit_job_v2(
                         "subjob_order": subjob_ids,
                     }
                 )
-        elif document_types and payloads and document_types[0].lower() in ["mp4", "mov", "avi", "mp3", "wav"]:
+        elif document_types and payloads and document_types[0].lower() in ["mp4", "mov", "avi", "mp3", "wav", "mkv"]:
             document_type = document_types[0]
             upload_path = f"./{Path(original_source_id).name}"
             # dump the payload to a file, just came from client

{nv_ingest-2026.1.4.dev20260104 → nv_ingest-2026.1.5.dev20260105}/nv_ingest/pipeline/default_libmode_pipeline_impl.py

@@ -243,7 +243,7 @@ stages:
           $OCR_GRPC_ENDPOINT|"",
           $OCR_HTTP_ENDPOINT|"https://ai.api.nvidia.com/v1/cv/nvidia/nemoretriever-ocr-v1"
         ]
-        ocr_infer_protocol: $PADDLE_INFER_PROTOCOL|"http"
+        ocr_infer_protocol: $OCR_INFER_PROTOCOL|"http"
         auth_token: $NGC_API_KEY|$NVIDIA_API_KEY
     replicas:
       min_replicas: 0

{nv_ingest-2026.1.4.dev20260104 → nv_ingest-2026.1.5.dev20260105}/nv_ingest.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nv-ingest
-Version: 2026.1.4.dev20260104
+Version: 2026.1.5.dev20260105
 Summary: Python module for multimodal document ingestion
 Author-email: Jeremy Dyer <jdyer@nvidia.com>
 License:                                  Apache License
@@ -227,7 +227,7 @@ Requires-Dist: httpx>=0.28.1
 Requires-Dist: isodate>=0.7.2
 Requires-Dist: langdetect>=1.0.9
 Requires-Dist: minio>=7.2.12
-Requires-Dist: librosa>=0.10.2
+Requires-Dist: librosa==0.10.2
 Requires-Dist: opentelemetry-api>=1.27.0
 Requires-Dist: opentelemetry-exporter-otlp>=1.27.0
 Requires-Dist: opentelemetry-sdk>=1.27.0

{nv_ingest-2026.1.4.dev20260104 → nv_ingest-2026.1.5.dev20260105}/nv_ingest.egg-info/requires.txt

@@ -12,7 +12,7 @@ httpx>=0.28.1
 isodate>=0.7.2
 langdetect>=1.0.9
 minio>=7.2.12
-librosa>=0.10.2
+librosa==0.10.2
 opentelemetry-api>=1.27.0
 opentelemetry-exporter-otlp>=1.27.0
 opentelemetry-sdk>=1.27.0

{nv_ingest-2026.1.4.dev20260104 → nv_ingest-2026.1.5.dev20260105}/pyproject.toml

@@ -34,7 +34,7 @@ dependencies = [
     "isodate>=0.7.2",
     "langdetect>=1.0.9",
     "minio>=7.2.12",
-    "librosa>=0.10.2",
+    "librosa==0.10.2",
     "opentelemetry-api>=1.27.0",
     "opentelemetry-exporter-otlp>=1.27.0",
     "opentelemetry-sdk>=1.27.0",