Merge update

chat_service/main.py

@@ -8,7 +8,7 @@
     level=logging.INFO, format="%(asctime)s [%(levelname)s] %(name)s: %(message)s"
 )
 
-app = FastAPI()
+app = FastAPI(root_path="/chat")
 
 app.include_router(health.router)
 app.include_router(chat.router)

docker-compose.yml

@@ -4,8 +4,6 @@ services:
       context: .
       dockerfile: ./pdf_processor_service/Dockerfile
     container_name: pdf_processor_service
-    ports:
-      - "8000:8000"
     env_file:
       - ./pdf_processor_service/.env
 
@@ -14,18 +12,14 @@ services:
       context: .
       dockerfile: ./chat_service/Dockerfile
     container_name: chat_service
-    ports:
-      - "8001:8000"
     env_file:
       - ./chat_service/.env
-      
-  pdf_extraction:
+
+  pdf_extraction_service:
     build:
       context: .
       dockerfile: ./pdf_extraction_service/Dockerfile
     container_name: pdf_extraction_service
-    ports:
-      - "8002:8000"
     depends_on:
       - minio
 
@@ -40,7 +34,21 @@ services:
       - ./embedder_service/.env
     depends_on:
       - chromadb
-
+
+  nginx:
+    container_name: nginx
+    build: 
+      context: .
+      dockerfile: ./nginx/Dockerfile
+    ports:
+      - "80:8080"
+    env_file:
+      - ./nginx/.env
+    depends_on:
+      - pdf_extraction_service
+      - pdf_processor_service
+      - chat_service
+
   redis:
     container_name: redis
     image: redis:7.4.4-alpine
@@ -49,7 +57,7 @@ services:
 
   chromadb:
     container_name: chromadb
-    image: chromadb/chroma:1.0.14
+    image: chromadb/chroma:1.0.13
     expose:
       - "5100"
 

embedder_service/models/embed.py

@@ -16,7 +16,6 @@ class ProcessingConfig(BaseModel):
         default=512, description="Target chunk size in characters")
     overlap: int = Field(
         default=50, description="Overlap between chunks in characters")
-    # Default embedding model provided by Sentence Transformers
     embedding_model: str = Field(
         default=EMBEDDING_MODEL_NAME, description="Sentence Transformer model")
     breakpoint_threshold_type: BreakpointThresholdType = Field(

embedder_service/models/helper.py

@@ -25,9 +25,6 @@ def embed_query(self, text: str) -> List[float]:
         """Embed a single query"""
         embedding = self.model.encode([text], convert_to_tensor=False)
         return embedding[0].tolist()
-
-    # def name(self) -> str:
-    #     return "sentence_transformer"
 
 
 @lru_cache(maxsize=5)

embedder_service/routers/embed.py

@@ -52,7 +52,7 @@ async def data_chunking(request:DataRequest, chunker) -> List[Dict[str, Any]]:
             # First iteration: Extract first chunk of doc.page_content
             chunk_content = chunk.page_content
             logger.info(f"Length of chunk {i+1}: {len(chunk_content.strip())}")
-            # First iteration: Start from first chunk of doc.page_context
+
             chunk_start = request.text.find(chunk_content, current_pos)
 
             if chunk_start == -1:
@@ -105,7 +105,7 @@ async def data_chunking(request:DataRequest, chunker) -> List[Dict[str, Any]]:
         return chunk_data
     except Exception as e:
         logger.error(f"Semantic chunking failed: {e}")
-        raise HTTPException(status_code=500, detail=f"Processing failed: {str(e)}")
+        raise HTTPException(status_code=500, detail="Data chunking failed.")
 
 
 async def vectorize_chromadb(chunk_data: List[Dict[str, Any]], config: ProcessingConfig, emb_model):
@@ -120,15 +120,7 @@ async def vectorize_chromadb(chunk_data: List[Dict[str, Any]], config: Processin
             logger.info(f"Using existing collection: {config.collection_name}")
         except Exception as e:
             logger.error(f"Collection retrieval failed: {e}")
-            raise HTTPException(status_code=500, detail=f"Collection retrieval failed: {str(e)}")
-
-        # for chunk in chunk_data:
-        #     collection.add(
-        #         ids=[chunk['chunk_id']],
-        #         documents=[chunk["content"]],
-        #         metadatas=[chunk["metadata"]]
-        #     )
-        #     logger.info(f"Added chunk {chunk['chunk_id']} to collection")
+            raise HTTPException(status_code=500, detail="Collection retrieval failed.")
 
         ids = [chunk['chunk_id'] for chunk in chunk_data]
         documents = [chunk['content'] for chunk in chunk_data]
@@ -167,7 +159,7 @@ async def vectorize_chromadb(chunk_data: List[Dict[str, Any]], config: Processin
 
     except Exception as e:
         logger.error(f"Embedding process failed: {e}")
-        raise HTTPException(status_code=500, detail=f"Embedding failed: {str(e)}")
+        raise HTTPException(status_code=500, detail="Embedding failed")
 
 
 # def serialize_chroma_results(results: Dict[str, Any]) -> Dict[str, Any]:
@@ -239,7 +231,7 @@ async def pdf_embedder_service(request: DataRequest):
             }
     except Exception as e:
         logger.error(f"PDF embedder service failed: {e}")
-        raise HTTPException(status_code=500, detail=f"Service failed: {str(e)}")
+        raise HTTPException(status_code=500, detail="PDF embedder service failed")
 
 
 @router.get("/status/{doc_id}")
@@ -277,4 +269,4 @@ async def verify_document_embedding(doc_id: str, collection_name: str = "my_docu
 
     except Exception as e:
         logger.error(f"Document verification failed: {e}")
-        raise HTTPException(status_code=500, detail=f"Verification failed: {str(e)}")
+        raise HTTPException(status_code=500, detail="Document verification failed")

nginx/Dockerfile

@@ -0,0 +1,5 @@
+FROM nginx:1.29.0-alpine
+
+EXPOSE 8080
+
+COPY ./nginx/nginx.conf.template /etc/nginx/templates/default.conf.template

nginx/example.env

@@ -0,0 +1,3 @@
+PDF_PROCESSOR_URL=http://pdf_processor_service:8000
+PDF_EXTRACTOR_URL=http://pdf_extraction_service:8000
+CHAT_URL=http://chat_service:8000

nginx/nginx.conf.template

@@ -0,0 +1,22 @@
+server { 
+    listen 8080;
+    server_name localhost;
+
+    proxy_set_header X-Real-IP $remote_addr;
+    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+    proxy_set_header X-Forwarded-Proto $scheme;
+    proxy_set_header X-NginX-Proxy true;
+    proxy_set_header Host $host;
+
+    location /pdf_processor/ {
+        proxy_pass ${PDF_PROCESSOR_URL}/;
+    }
+
+    location /pdf_extraction/ {
+        proxy_pass ${PDF_EXTRACTOR_URL}/;
+    }
+
+    location /chat/ {
+        proxy_pass ${CHAT_URL}/;
+    }
+}

pdf_extraction_service/main.py

@@ -8,7 +8,7 @@
     format="%(asctime)s [%(levelname)s] %(name)s: %(message)s"
 )
 
-app = FastAPI()
+app = FastAPI(root_path="/pdf_extraction")
 
 app.include_router(health.router)
 app.include_router(extractor.router)

pdf_processor_service/main.py

@@ -9,7 +9,7 @@
     format="%(asctime)s [%(levelname)s] %(name)s: %(message)s"
 )
 
-app = FastAPI()
+app = FastAPI(root_path="/pdf_processor")
 
 app.include_router(health.router)
 app.include_router(document.router)