Fix Dataprep pinecone CI issue

comps/dataprep/pinecone/langchain/README.md

@@ -41,7 +41,7 @@ docker build -t opea/dataprep-pinecone:latest --build-arg https_proxy=$https_pro
 ### Run Docker with CLI
 
 ```bash
-docker run -d --name="dataprep-pinecone-server" -p 6000:6000 --ipc=host -e http_proxy=$http_proxy -e https_proxy=$https_proxy opea/dataprep-pinecone:latest
+docker run -d --name="dataprep-pinecone-server" -p 6007:6007 --ipc=host -e http_proxy=$http_proxy -e https_proxy=$https_proxy opea/dataprep-pinecone:latest
 ```
 
 ### Setup Environment Variables
@@ -65,5 +65,5 @@ docker compose -f docker-compose-dataprep-pinecone.yaml up -d
 Once document preparation microservice for Pinecone is started, user can use below command to invoke the microservice to convert the document to embedding and save to the database.
 
 ```bash
-curl -X POST -H "Content-Type: application/json" -d '{"path":"/path/to/document"}' http://localhost:6000/v1/dataprep
+curl -X POST -H "Content-Type: application/json" -d '{"path":"/path/to/document"}' http://localhost:6007/v1/dataprep
 ```

comps/dataprep/pinecone/langchain/pinecone_langchain.yaml

@@ -23,8 +23,6 @@ services:
     container_name: dataprep-pinecone-server
     ports:
       - "6007:6007"
-      - "6008:6008"
-      - "6009:6009"
     ipc: host
     environment:
       no_proxy: ${no_proxy}

comps/dataprep/pinecone/langchain/prepare_doc_pinecone.py

@@ -118,7 +118,7 @@ def ingest_data_to_pinecone(doc_path: DocPath):
         table_chunks = get_tables_result(path, doc_path.table_strategy)
         chunks = chunks + table_chunks
     if logflag:
-        logger.info("Done preprocessing. Created ", len(chunks), " chunks of the original file.")
+        logger.info(f"Done preprocessing. Created {len(chunks)} chunks of the original file.")
 
     # Create vectorstore
     if tei_embedding_endpoint:
@@ -135,7 +135,7 @@ def ingest_data_to_pinecone(doc_path: DocPath):
         # Creating the index
         create_index(pc)
         if logflag:
-            logger.info("Successfully created the index", PINECONE_INDEX_NAME)
+            logger.info(f"Successfully created the index {PINECONE_INDEX_NAME}")
 
     # Batch size
     batch_size = 32
@@ -174,7 +174,7 @@ async def ingest_link_to_pinecone(link_list: List[str], chunk_size, chunk_overla
         # Creating the index
         create_index(pc)
         if logflag:
-            logger.info("Successfully created the index", PINECONE_INDEX_NAME)
+            logger.info(f"Successfully created the index {PINECONE_INDEX_NAME}")
 
     # save link contents and doc_ids one by one
     for link in link_list:
@@ -252,7 +252,7 @@ async def ingest_documents(
 
 
 @register_microservice(
-    name="opea_service@prepare_doc_pinecone_file", endpoint="/v1/dataprep/get_file", host="0.0.0.0", port=6008
+    name="opea_service@prepare_doc_pinecone", endpoint="/v1/dataprep/get_file", host="0.0.0.0", port=6007
 )
 async def rag_get_file_structure():
     if logflag:
@@ -270,7 +270,7 @@ async def rag_get_file_structure():
 
 
 @register_microservice(
-    name="opea_service@prepare_doc_pinecone_del", endpoint="/v1/dataprep/delete_file", host="0.0.0.0", port=6009
+    name="opea_service@prepare_doc_pinecone", endpoint="/v1/dataprep/delete_file", host="0.0.0.0", port=6007
 )
 async def delete_all(file_path: str = Body(..., embed=True)):
     """Delete file according to `file_path`.
@@ -288,7 +288,7 @@ async def delete_all(file_path: str = Body(..., embed=True)):
             logger.info("[dataprep - del] successfully delete all files.")
         create_upload_folder(upload_folder)
         if logflag:
-            logger.info({"status": True})
+            logger.info('{"status": True}')
         return {"status": True}
     else:
         raise HTTPException(status_code=404, detail="Single file deletion is not implemented yet")
@@ -297,5 +297,3 @@ async def delete_all(file_path: str = Body(..., embed=True)):
 if __name__ == "__main__":
     create_upload_folder(upload_folder)
     opea_microservices["opea_service@prepare_doc_pinecone"].start()
-    opea_microservices["opea_service@prepare_doc_pinecone_file"].start()
-    opea_microservices["opea_service@prepare_doc_pinecone_del"].start()

tests/dataprep/test_dataprep_pinecone_langchain.sh

@@ -24,7 +24,7 @@ function start_service() {
     export PINECONE_INDEX_NAME="test-index"
     export HUGGINGFACEHUB_API_TOKEN=$HF_TOKEN
 
-    docker run -d --name="test-comps-dataprep-pinecone" -p 5039:6007 -p 5040:6008 -p 5041:6009 --ipc=host -e http_proxy=$http_proxy -e https_proxy=$https_proxy -e no_proxy=$no_proxy -e PINECONE_API_KEY=$PINECONE_API_KEY -e PINECONE_INDEX_NAME=$PINECONE_INDEX_NAME opea/dataprep-pinecone:comps
+    docker run -d --name="test-comps-dataprep-pinecone" -p 5039:6007 --ipc=host -e http_proxy=$http_proxy -e https_proxy=$https_proxy -e no_proxy=$no_proxy -e PINECONE_API_KEY=$PINECONE_API_KEY -e PINECONE_INDEX_NAME=$PINECONE_INDEX_NAME -e LOGFLAG=true opea/dataprep-pinecone:comps
 
     sleep 1m
 }
@@ -41,7 +41,7 @@ function validate_microservice() {
         docker logs test-comps-dataprep-pinecone
         exit 1
     fi
-    DELETE_URL="http://$ip_address:5041/v1/dataprep/delete_file"
+    DELETE_URL="http://$ip_address:5039/v1/dataprep/delete_file"
     result=$(curl --noproxy $ip_address --location --request POST \
       -d '{"file_path": "all"}' -H 'Content-Type: application/json' $DELETE_URL)
     if [[ $result == *"true"* ]]; then