from userio import *
import requests
import re
import newsParser
from requests_html import HTML
from requests_html import HTMLSession

def article(url):
  say("Article: "+url)
  if not "/amphtml" in url:
    say("Trying AMP")
    url = url.replace("buzzfeednews.com/article","buzzfeednews.com/amphtml")
    url = url.replace("buzzfeed.com/","buzzfeed.com/amphtml/")
  url.replace("?origin=web-hf","")

  session = HTMLSession()
  response = session.get(url,timeout=20,headers={'Accept-Encoding': 'deflate', 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/90.0.4430.212 Safari/537.36'})
  pageContent=""
  article_only=""
  with response as r:
    #articleStrTitle = r.html.xpath('//meta[@property="og:title"]/@content')[0]
    #articleStrDescription = r.html.xpath('//meta[@property="og:description"]/@content')[0]
    #articleStrImageUrl = r.html.xpath('//meta[@property="og:image"]/@content')[0]
    #articleStrAuthor = r.html.xpath('//div[@class="author_wrapper"]/@content')
    #print(articleStrAuthor)
    article=r.html.find("main")[0]
    #article=r.html.find("body")[0]
    article_only+=article.html
    lenBefore=len(article_only)
    say("LengthBefore: "+str(lenBefore))
    pageContent += "<meta property=\"og:type\" content=\"article\" />\n"
    #pageContent += "<meta property=\"og:title\" content=\""+articleStrTitle+"\" />\n"
    #pageContent += "<meta property=\"og:description\" content=\""+articleStrDescription+"\" />\n"
    pageContent += "<meta property=\"og:url\" content=\""+url+"\" />\n"
    #pageContent += "<meta property=\"og:image\" content=\""+articleStrImageUrl+"\" />\n"
    pageContent += "<meta property=\"og:image:type\" content=\"image/jpeg\" />\n"
    #pageContent += "<meta name=\"author\" content=\""+articleStrAuthor+"\" />\n"

  article_only = re.sub(r"<amp-img", '<img', article_only)
  article_only = re.sub(r"</amp-img>", '', article_only)
  article_only = re.sub(r"<h2", '<h3', article_only)
  article_only = re.sub(r"</h2>", '</h3>', article_only)
  article_only = re.sub(r"<h1", '<h2', article_only)
  article_only = re.sub(r"</h1>", '</h2>', article_only)
  article_only = re.sub(r'<nav class="newsblock-trending-tags-bar__nav"(.+?)</nav>','',article_only,flags=re.M|re.S)
  article_only = re.sub(r'<div class="newsblock-trending-tags-bar__title-wrapper"(.+?)</div>','',article_only,flags=re.M|re.S)
  article_only = re.sub(r'<amp-ad type(.+?)</amp-ad>','',article_only,flags=re.M|re.S)
  article_only = re.sub(r'<amp-social-share(.+?)</amp-social-share>','',article_only,flags=re.M|re.S)
  article_only = re.sub(r'<p class="xs-pb1 xs-pt1 xs-text-center caps xs-text-6 text-gray-lightest ad__disclosure">(.+?)</p>','',article_only,flags=re.M|re.S)
  #article_only = re.sub(r'','',article_only)
  #article_only = re.sub(r"href=\"/",'href="https://bfmtv.com/',article_only)
  #article_only = re.sub(r"src=\"/",'src="https://bfmtv.com/',article_only)
  article_only = re.sub(r"^$",'',article_only)
  article_only = re.sub(r'^\s*$', '',article_only,flags=re.M|re.S)
  article_only = re.sub(r"><",'>\n<',article_only)

  pageContent += "<article>\n"+article_only+"\n</article>\n"
  lenAfter=len(article_only)
  lenGain=float(10000-int(float(100*lenAfter/lenBefore)*100))/100
  say("LengthAfter : "+str(lenAfter))
  say("Gain        : "+str(lenGain)+"%")
  return pageContent

def articleOld(url):
  say("Article: "+url)
  if not "/amphtml" in url:
    say("Trying AMP")
    url = url.replace("buzzfeednews.com/article","buzzfeednews.com/amphtml")
    url = url.replace("buzzfeed.com/","buzzfeed.com/amphtml/")
  url.replace("?origin=web-hf","")

  r = requests.get(url, allow_redirects=True)
  content = r.text
  pageContent = ""
  articleCstBegin = "<article "
  articleCstEnd   = "<div class=\"subbuzz subbuzz-bfp\">"
  articleCstEnd2   = "</article>"
  articleCstEnd3   = "<div class=\"shares shares--inline"
  articleStrTitle = newsParser.articleTitle(content)
  articleStrDescription = newsParser.articleDescription(content)
  articleStrImageUrl = newsParser.articleImage(content)
  
  indexBegin = content.index(articleCstBegin)
  try:
    indexEnd   = content.index(articleCstEnd)
  except:
    try:
      indexEnd   = content.index(articleCstEnd2)
    except:
      indexEnd   = content.index(articleCstEnd3)
      
  pageContent = ""
  pageContent += "<meta property=\"og:type\" content=\"article\">\n"
  pageContent += "<meta property=\"og:title\" content=\""+articleStrTitle+"\">\n"
  pageContent += "<meta property=\"og:description\" content=\""+articleStrDescription+"\">\n"
  pageContent += "<meta property=\"og:url\" content=\""+url+"\">\n"
  pageContent += "<meta property=\"og:image\" content=\""+articleStrImageUrl+"\">\n"
  pageContent += "<meta property=\"og:image:type\" content=\"image/jpeg\">"
  article_only = ""
  article_only += "<h1>"+articleStrTitle+"</h1>\n"
  article_only += "<em>"+articleStrDescription+"</em>\n"
  article_only += content[indexBegin:indexEnd]
  article_only = re.sub(r"<amp-img", '<img', article_only)
  article_only = re.sub(r"</amp-img>", '', article_only)
  article_only = re.sub(r"<h2", '<h3', article_only)
  article_only = re.sub(r"</h2>", '</h3>', article_only)
  article_only = re.sub(r"<h1", '<h2', article_only)
  article_only = re.sub(r"</h1>", '</h2>', article_only)
  
  # ~ article_only = re.sub(r"<picture><source media=\"(.*?)\" srcSet=\"(.*?)\?(.*?)</picture>", "<img src=\"\g<2>\">", article_only)
  article_only = re.sub(r"<amp-social-share (.*?)>", "<amp-social-share>", article_only)
  article_only = re.sub(r"<span class=\"icon icon--primary flex\">", "<span>", article_only)
  article_only = re.sub(r"<title>(.*?)</title>", "", article_only)
  article_only = re.sub(r"<use xlink:href=\"(.*?)\">", "<use>", article_only)
  article_only = re.sub(r"<svg class=\"svg-(.*?)\">", "<svg height=\"1px\">", article_only)
  article_only = re.sub(r"Share on Facebook", "", article_only)
  article_only = re.sub(r"Share on Pinterest", "", article_only)
  article_only = article_only.replace("><", ">\n<")
  
  if "buzzfeed.com" in url:
    article_only = re.sub(r"href=\"\/", 'href=\"//www.buzzfeed.com/', article_only)
  elif "buzzfeednews.com" in url:
    article_only = re.sub(r"href=\"\/", 'href=\"//www.buzzfeednews.com/', article_only)
  pageContent += "<article>"+article_only+"</article>"
  return pageContent