extracción iniciativas constituyentes

iniciativas.py

import pickle, requests, time
from bs4 import BeautifulSoup
import pandas as pd

def tit(html):
    tt0=BeautifulSoup(html,'lxml').find_all('meta', attrs={'property': 'og:title'})[0]
    return tt0['content']
    
def apo(html):
    idx = html.index('Cuenta con')
    cola = html[(idx+29):]
    jdx = cola.index('<')
    return int(cola[:jdx].replace('.',''))    
####################################################################################
url = 'https://iniciativas.chileconvencion.cl/m/iniciativa_popular/'
ubs = BeautifulSoup(requests.get(url).text,'lxml')
lbs = ubs.find_all('a')
lbs = sorted(set([link['href'] for link in lbs]))

print('LINKS:', len(lbs))
props = {}
t0 = time.time()

for ix, link in enumerate(lbs):
    if 'o/' in link:
        if ix%100==50:
            print(lurl, ix, round(time.time()-t0,2))
        lt = requests.get(url+link).text
        props[link[2:]] = lt

pdf = pd.DataFrame(dict(id=props.keys(), texto=props.values()))
pdf['apoyos'] = pdf.texto.apply(apo)
pdf['titulo'] = pdf.texto.apply(tit)

top20 = pdf.sort_values('apoyos',ascending=False).head(20)[['titulo','apoyos','id']]