File size: 4,196 Bytes
04e75ed | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 | import os
from utils.central_logging import setup_logging,get_logger
import textwrap
from langchain_openai import OpenAI
from langchain_chroma import Chroma
#from langchain_community.document_loaders import SeleniumURLLoader
from dotenv import load_dotenv
import os
import openai
from langchain_openai import ChatOpenAI
from langchain_core.runnables import RunnableLambda
import chromadb
import gradio as gr
import time
import asyncio
import nest_asyncio
import threading
import re
from openai import OpenAI
#import streamlit as st
from whisper_singleton import get_embedding,save_file,transcribe_content
from extract_text import pdf_to_documents,store_data
from prompt import get_prompt,get_system_prompt
load_dotenv("./.env")
setup_logging()
logger = get_logger("chat")
_embedding = None
_retriever = None
_vectore_store = None
openai_api_key = os.getenv("OPENAI_API_KEY")
if openai_api_key:
logger.info("Open ai api key has been set")
else:
logger.error("No open ai api key has been found")
try:
llm_openai = ChatOpenAI(model='gpt-3.5-turbo',temperature=0)
client = OpenAI()
logger.info("Clients has been initialized")
except Exception as e:
logger.exception(f"An exception occured: {e}")
def handle_upload(file_path):
global _embedding
global _retriever
_embedding = get_embedding()
text_content = ""
status_message = ""
file_name = "./transcribe.txt"
try:
if file_path.lower().endswith(".pdf"):
collection_name = "pdffiles"
pdf_docs,_vectore_store = pdf_to_documents(file_path,"transcribe_db",collection_name,_embedding)
text_content = "\n\n".join([doc.page_content for doc in pdf_docs])
status_message = "📄 PDF file uploaded — extraction implemented."
logger.info(status_message)
#save_file(file_name,text_content)
elif file_path.lower().endswith(".mp3") or file_path.lower().endswith('.mp4'):
print(f"path:{file_path}")
if file_path.lower().endswith(".mp3"):
collection_name = "audios"
status_message = "🎧 MP3 uploaded — transcription implemented."
logger.info(status_message)
else:
collection_name = "videos"
status_message = "🎬 MP4 uploaded — video transcription implemented."
logger.info(status_message)
text_content = transcribe_content(file_path)
_vectore_store = store_data(text_content,"transcribe_db",collection_name,_embedding)
#save_file(file_name,text_content)
else:
status_message = "Invalid file format"
except Exception as e:
status_message = f"❌ Error processing file: {e}"
logger.exception(status_message)
_retriever = _vectore_store.as_retriever()
return status_message,text_content
def stream_response(user_input,history):
history = history or []
history.append({"role": "user", "content": user_input})
history.append({"role": "assistant", "content": ""})
context = ""
if _retriever is not None:
docs = _retriever.invoke(user_input)
context = "\n\n".join([d.page_content for d in docs])
formatted_history = "\n".join(
f"{m['role'].capitalize()}: {m['content']}"
for m in history
)
system_prompt = get_system_prompt().format(
history=formatted_history,
context=context,
user_message=user_input
)
messages = [
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_input},
]
partial_reply = ""
stream = client.chat.completions.create(
model="gpt-4o-mini",
messages=messages,
stream=True,
temperature = 0
)
for event in stream:
delta = event.choices[0].delta
if delta and delta.content:
token = delta.content
partial_reply += token
history[-1]["content"] = partial_reply
yield history, history, ""
history[-1]["content"] = partial_reply
yield history, history, ""
|