from fastapi import FastAPI, Query, HTTPException
from extractous import Extractor, TesseractOcrConfig
from bs4 import BeautifulSoup

app = FastAPI()

@app.get("/")
def accepts_pdf_link(link: str = Query(..., description="The URL to pdf file")):
    if not link.startswith(("http://", "https://")):
        raise HTTPException(status_code=400, detail="Invalid URL format")
    extractor = Extractor().set_ocr_config(TesseractOcrConfig())
    extractor = extractor.set_xml_output(True)
    content, metadata  = extractor.extract_url_to_string(link)
    soup = BeautifulSoup(content, 'html.parser')
    pages = soup.find_all('div', class_='page')
    pages_text = [p.get_text() for p in pages]
    return {"received_link": link, "content": pages_text}