Create app.py

app.py

@@ -0,0 +1,83 @@
+import gradio as gr
+import os
+import zipfile
+import uuid
+from langchain_huggingface import HuggingFaceEmbeddings
+from pinecone import Pinecone, ServerlessSpec
+from langchain_community.document_loaders import WhatsAppChatLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+# Initialize Hugging Face embeddings
+embeddings = HuggingFaceEmbeddings()
+
+def load_chat_content(pinecone_key, file) -> str:
+    """Load chat content from the uploaded zip file and store it in Pinecone."""
+    # Initialize Pinecone
+    pc = Pinecone(api_key=pinecone_key)
+    index_name = "whatsapp-group-chat-index"
+    
+    if index_name not in pc.list_indexes().names():
+        pc.create_index(
+            name=index_name,
+            dimension=768,  # change as per embedding model
+            metric="cosine",
+            spec=ServerlessSpec(
+                cloud='aws',
+                region='us-east-1'
+            )
+        )
+
+    index = pc.Index(index_name)
+
+    # Load and process the ZIP file
+    temp_dir = 'temp_extracted_files'
+    os.makedirs(temp_dir, exist_ok=True)
+    
+    with zipfile.ZipFile(file, 'r') as z:
+        z.extractall(temp_dir)
+        
+    chat_files = [f for f in os.listdir(temp_dir) if f.endswith('.txt')]
+    if not chat_files:
+        raise ValueError("No chat files found in the zip archive.")
+
+    chat_file_path = os.path.join(temp_dir, chat_files[0])
+    
+    loader = WhatsAppChatLoader(path=chat_file_path)
+    raw_messages = loader.lazy_load()
+    messages = list(raw_messages)
+    
+    chat_content = "\n".join([doc.page_content for doc in messages])
+    
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=3000,
+        chunk_overlap=150,
+        length_function=len,
+    )
+    
+    chunks = text_splitter.create_documents([chat_content])
+    
+    # Store chunks in Pinecone with unique IDs
+    vectors_to_upsert = []
+    for i, chunk in enumerate(chunks):
+        vector = embeddings.embed_documents([chunk.page_content])[0]
+        unique_id = str(uuid.uuid4())  # Generate a unique ID
+        vectors_to_upsert.append((unique_id, vector, {"text": chunk.page_content}))
+    
+    index.upsert(vectors_to_upsert)
+
+    return "Chat content has been successfully upserted to Pinecone."
+
+# Define the Gradio interface
+interface = gr.Interface(
+    fn=load_chat_content,
+    inputs=[
+        gr.Textbox(label="Enter Pinecone API Key", type="password"),
+        gr.File(label="Upload WhatsApp Chat Zip File")
+    ],
+    outputs="text",
+    title="WhatsApp Chat Upsert to Pinecone",
+    description="Upload a zip file containing a WhatsApp chat file and upsert its content to Pinecone.",
+)
+
+if __name__ == "__main__":
+    interface.launch()