1
0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-11-14 05:17:25 +00:00
youtube-dl/youtube_dl/extractor/tf1.py
Jaime Marquínez Ferrándiz 8244288dfe WatIE: support videos divided in multiple parts (closes #222 and #659)
The id for the videos is now the full id, no the one in the webpage url.
Also extract more information: description, view_count and upload_date
2013-06-29 18:22:03 +02:00

36 lines
1.5 KiB
Python

# coding: utf-8
import json
import re
from .common import InfoExtractor
class TF1IE(InfoExtractor):
"""
TF1 uses the wat.tv player, currently it can only download videos with the
html5 player enabled, it cannot download HD videos.
"""
_VALID_URL = r'http://videos.tf1.fr/.*-(.*?).html'
_TEST = {
u'url': u'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html',
u'file': u'10635995.mp4',
u'md5': u'66789d3e91278d332f75e1feb7aea327',
u'info_dict': {
u'title': u'Citroën Grand C4 Picasso 2013 : présentation officielle',
u'description': u'Vidéo officielle du nouveau Citroën Grand C4 Picasso, lancé à l\'automne 2013.',
}
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
id = mobj.group(1)
webpage = self._download_webpage(url, id)
embed_url = self._html_search_regex(r'"(https://www.wat.tv/embedframe/.*?)"',
webpage, 'embed url')
embed_page = self._download_webpage(embed_url, id, u'Downloading embed player page')
wat_id = self._search_regex(r'UVID=(.*?)&', embed_page, 'wat id')
wat_info = self._download_webpage('http://www.wat.tv/interface/contentv3/%s' % wat_id, id, u'Downloading Wat info')
wat_info = json.loads(wat_info)['media']
wat_url = wat_info['url']
return self.url_result(wat_url, 'Wat')