import gradio as gr
from transformers import pipeline,AutoTokenizer, AutoModelForSeq2SeqLM
from bs4 import BeautifulSoup
from bs4.element import Comment
from urllib.request import urlopen
import urllib.request 
from bs4 import BeautifulSoup

def easyterms(text:str)->str:
    print("In summerizing function of easyterms")
    tokenizer = AutoTokenizer.from_pretrained("EasyTerms/etsummerizer_v2")
    model = AutoModelForSeq2SeqLM.from_pretrained("EasyTerms/etsummerizer_v2")

    inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512)
    summary_ids = model.generate(inputs['input_ids'], attention_mask=inputs['attention_mask'], max_length=128, num_beams=4)
    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
    return summary
def get_paragaph(url:str)-> list:
    parser = 'html.parser'  # or 'lxml' (preferred) or 'html5lib', if installed
    html = urllib.request.urlopen(url)
    # parsing the html file
    soup = BeautifulSoup(html, parser, from_encoding=html.info().get_param('charset'))
    samples = soup.findAll("p")
    samples = [item.text for item in samples]
    return samples

def summerize(Option:str, Text:str)-> str:
    print(Option)
    if Option == "text":
        return easyterms(Text)
    else:
        paragraph = get_paragaph(Text)
        result = []
        for par in paragraph:
            result.append(easyterms(par))
        res = '\n'.join(data for data in result)
        return res


intro = gr.Markdown(
        '''
        <center><h1>A Legal document summerizer.</h1></span>  
        
        If you want to better understand legal text or document, this platform is for you. By choosing the url option you submit a url whose content will in turn be summerized for you.
        Otherwhise you can choose the text option and submit your own text to be summerized.

        '''
    )

interface = gr.Interface(
    fn=summerize,
    inputs=[gr.Radio(["url", "text"]),"text"],
    outputs=["text"]
)

interface.launch()