[Bug] Fix VDMS retriever and apply fix to VDMS dataprep (#928)

* Update requirements to pin protobuf version and fix grpc conflict, and limit vdms version

Signed-off-by: Lacewell, Chaunte W <[email protected]>

* Update fix by removing grpcio pin and pinning opentelemetry-proto to 1.23.0

Signed-off-by: Lacewell, Chaunte W <[email protected]>

---------

Signed-off-by: Lacewell, Chaunte W <[email protected]>

(cherry picked from commit 750e501)
Signed-off-by: chensuyue <[email protected]>

comps/dataprep/vdms/langchain/requirements.txt

@@ -1,39 +1,41 @@
-beautifulsoup4==4.12.3
-cairosvg==2.7.1
-decord==0.6.0
-docarray[full]==0.40.0
-docx2txt==0.8
-easyocr==1.7.2
-einops==0.8.0
-fastapi==0.115.5
-html2text==2024.2.26
-huggingface-hub==0.26.2
-langchain==0.3.7
-langchain-community==0.3.7
-langchain-core==0.3.19
-langchain-text-splitters==0.3.2
-langsmith==0.1.143
-markdown==3.7
-numpy==1.26.4
-opencv-python==4.10.0.84
-opentelemetry-api==1.28.2
-opentelemetry-exporter-otlp==1.28.2
-opentelemetry-sdk==1.28.2
-pandas==2.2.3
-pillow==11.0.0
-prometheus-fastapi-instrumentator==7.0.0
-pymupdf==1.24.14
-pyspark==3.5.3
-pytesseract==0.3.13
+beautifulsoup4
+cairosvg
+decord
+docarray[full]
+docx2txt
+easyocr
+einops
+fastapi
+html2text
+huggingface_hub
+langchain
+langchain-community
+langchain-core
+langchain-text-splitters
+langsmith
+markdown
+numpy
+opencv-python
+opentelemetry-api
+opentelemetry-exporter-otlp
+opentelemetry-proto==1.23.0
+opentelemetry-sdk
+pandas
+Pillow
+prometheus-fastapi-instrumentator
+protobuf==4.24.2
+pymupdf
+pyspark
+pytesseract
 python-bidi==0.4.2
-python-docx==1.1.2
-python-pptx==0.6.23
-pyyaml==6.0.2
-sentence-transformers==3.3.1
-shortuuid==1.0.13
-tqdm==4.67.0
-typing==3.7.4.3
-tzlocal==5.2
+python-docx
+python-pptx
+PyYAML
+sentence_transformers
+shortuuid
+tqdm
+typing
+tzlocal
 unstructured[all-docs]==0.11.5
-uvicorn==0.32.0
-vdms==0.0.17
+uvicorn
+vdms>=0.0.20

comps/dataprep/vdms/multimodal_langchain/requirements.txt

@@ -1,37 +1,39 @@
-beautifulsoup4==4.12.3
-cairosvg==2.7.1
-decord==0.6.0
-docarray[full]==0.40.0
-docx2txt==0.8
-easyocr==1.7.2
-einops==0.8.0
-fastapi==0.115.5
-huggingface-hub==0.26.2
-langchain==0.3.7
-langchain-community==0.3.7
-langchain-core==0.3.19
-langchain-text-splitters==0.3.2
-langsmith==0.1.143
-markdown==3.7
-numpy==1.26.4
-opencv-python==4.10.0.84
-opentelemetry-api==1.28.2
-opentelemetry-exporter-otlp==1.28.2
-opentelemetry-sdk==1.28.2
-pandas==2.2.3
-pillow==11.0.0
-prometheus-fastapi-instrumentator==7.0.0
-pymupdf==1.24.14
-pyspark==3.5.3
+beautifulsoup4
+cairosvg
+decord
+docarray[full]
+docx2txt
+easyocr
+einops
+fastapi
+huggingface_hub
+langchain
+langchain-community
+langchain-core
+langchain-text-splitters
+langsmith
+markdown
+numpy
+opencv-python
+opentelemetry-api
+opentelemetry-exporter-otlp
+opentelemetry-proto==1.23.0
+opentelemetry-sdk
+pandas
+Pillow
+prometheus-fastapi-instrumentator
+protobuf==4.24.2
+pymupdf
+pyspark
 python-bidi==0.4.2
-python-docx==1.1.2
-python-pptx==0.6.23
-pyyaml==6.0.2
-sentence-transformers==3.3.1
-shortuuid==1.0.13
-tqdm==4.67.0
-typing==3.7.4.3
-tzlocal==5.2
+python-docx
+python-pptx
+PyYAML
+sentence_transformers
+shortuuid
+tqdm
+typing
+tzlocal
 unstructured[all-docs]==0.11.5
-uvicorn==0.32.0
-vdms==0.0.17
+uvicorn
+vdms>=0.0.20

comps/retrievers/vdms/langchain/requirements.txt

@@ -1,16 +1,18 @@
-docarray[full]==0.40.0
-easyocr==1.7.2
-einops==0.8.0
-fastapi==0.115.5
-langchain-community==0.3.7
-langchain-core==0.3.19
-langchain-huggingface==0.1.2
-opentelemetry-api==1.28.2
-opentelemetry-exporter-otlp==1.28.2
-opentelemetry-sdk==1.28.2
-prometheus-fastapi-instrumentator==7.0.0
-pymupdf==1.24.14
-sentence-transformers==3.3.1
-shortuuid==1.0.13
-uvicorn==0.32.0
-vdms==0.0.17
+docarray[full]
+easyocr
+einops
+fastapi
+langchain-community
+langchain-core
+langchain-huggingface
+opentelemetry-api
+opentelemetry-exporter-otlp
+opentelemetry-proto==1.23.0
+opentelemetry-sdk
+prometheus-fastapi-instrumentator
+protobuf==4.24.2
+pymupdf
+sentence_transformers
+shortuuid
+uvicorn
+vdms>=0.0.20