main.py

import requests
from bs4 import BeautifulSoup
from fastapi import FastAPI#, Request
from fastapi.responses import StreamingResponse
from pydantic import BaseModel
import re
import replicate


class Item(BaseModel):
    url: str
    max_tokens: int

app = FastAPI()


def extract_article_content(url):
    try:
        
        response = requests.get(url)

        soup = BeautifulSoup(response.text, 'html.parser')
        results = soup.find_all(['h1', 'p'])
        text = [result.text for result in results]
        ARTICLE = ' '.join(text)
        
        return ARTICLE
    except Exception as e:
        return ""


@app.get("/")
async def root():
    return {"status": "OK"}



@app.post("/summarize-v1")
async def root(item: Item):
    try:
        article = extract_article_content(item.url)

        if len(article) == 0:
            return {'summary': ""}

        event_list = []
        for event in replicate.stream("snowflake/snowflake-arctic-instruct", input= {
    "prompt": "summarize this following news article:" + article,
    "temperature": 0.2}):
            # Convert the event to a string and append it to the list
                            event_list.append(str(event))

        # After the event stream ends, process the collected events
        output_variable = "".join(event_list)
        return output_variable

    except requests.RequestException as e:
        return {"error": str(e), "status_code": 500}




@app.post("/summarize-v2")
async def root(item: Item):
    try:
        article = extract_article_content(item.url)

        if len(article) == 0:
            return {'summary': ""}

        def event_generator():
            for event in replicate.stream("snowflake/snowflake-arctic-instruct", input={
                "prompt": "summarize this following news article:" + article,
                "temperature": 0.2,
                "max_new_tokens" : item.max_tokens
            }):
                # Yield the event as a string
                yield str(event)
                print(str(event), end="")

        # Use StreamingResponse to stream the events
        return StreamingResponse(event_generator(), media_type='text/event-stream')

    except requests.RequestException as e:
        return {"error": str(e), "status_code": 500}





@app.post("/extract-content")
async def root(item: Item):
    
    try:

        article = extract_article_content(item.url)

        if len(article) == 0:
            return {'ERROR': "AHHHHHHHHH"}
        
        return {"content":article}
        
    except requests.RequestException as e:
        return {"error": str(e), "status_code": 500}