1
0
Fork 0
mirror of https://gitlab.com/dstftw/youtube-dl.git synced 2020-11-16 09:42:26 +00:00
youtube-dl/youtube_dl/extractor/wat.py

116 lines
4.3 KiB
Python
Raw Normal View History

# coding: utf-8
2014-03-29 14:15:16 +00:00
from __future__ import unicode_literals
2013-06-28 20:01:47 +00:00
import re
import time
import hashlib
2013-06-28 20:01:47 +00:00
from .common import InfoExtractor
from ..utils import (
ExtractorError,
unified_strdate,
2013-06-28 20:01:47 +00:00
)
class WatIE(InfoExtractor):
_VALID_URL = r'http://www\.wat\.tv/video/(?P<display_id>.*)-(?P<short_id>.*?)_.*?\.html'
2013-06-28 20:01:47 +00:00
IE_NAME = 'wat.tv'
_TEST = {
'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
'md5': 'ce70e9223945ed26a8056d413ca55dc9',
2014-03-29 14:15:16 +00:00
'info_dict': {
'id': '11713067',
'display_id': 'soupe-figues-l-orange-aux-epices',
2014-03-29 14:15:16 +00:00
'ext': 'mp4',
'title': 'Soupe de figues à l\'orange et aux épices',
'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.',
'upload_date': '20140819',
'duration': 120,
},
2013-06-28 20:01:47 +00:00
}
2014-03-29 14:15:16 +00:00
def download_video_info(self, real_id):
# 'contentv4' is used in the website, but it also returns the related
# videos, we don't need them
2014-03-29 14:15:16 +00:00
info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id)
return info['media']
2013-06-28 20:01:47 +00:00
def _real_extract(self, url):
def real_id_for_chapter(chapter):
return chapter['tc_start'].split('-')[0]
2013-06-28 20:01:47 +00:00
mobj = re.match(self._VALID_URL, url)
short_id = mobj.group('short_id')
display_id = mobj.group('display_id')
webpage = self._download_webpage(url, display_id or short_id)
real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
video_info = self.download_video_info(real_id)
if video_info.get('geolock'):
raise ExtractorError('This content is not available in your area', expected=True)
chapters = video_info['chapters']
first_chapter = chapters[0]
files = video_info['files']
first_file = files[0]
2013-06-28 20:01:47 +00:00
if real_id_for_chapter(first_chapter) != real_id:
self.to_screen('Multipart video detected')
chapter_urls = []
for chapter in chapters:
chapter_id = real_id_for_chapter(chapter)
# Yes, when we this chapter is processed by WatIE,
# it will download the info again
chapter_info = self.download_video_info(chapter_id)
chapter_urls.append(chapter_info['url'])
entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
return self.playlist_result(entries, real_id, video_info['title'])
2014-03-29 14:15:16 +00:00
upload_date = None
if 'date_diffusion' in first_chapter:
upload_date = unified_strdate(first_chapter['date_diffusion'])
# Otherwise we can continue and extract just one part, we have to use
# the short id for getting the video url
formats = [{
'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
'format_id': 'Mobile',
}]
fmts = [('SD', 'web')]
if first_file.get('hasHD'):
fmts.append(('HD', 'webhd'))
def compute_token(param):
timestamp = '%08x' % int(time.time())
magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564'
return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp)
for fmt in fmts:
webid = '/%s/%s' % (fmt[1], real_id)
video_url = self._download_webpage(
'http://www.wat.tv/get%s?token=%s&getURL=1' % (webid, compute_token(webid)),
real_id,
'Downloding %s video URL' % fmt[0],
'Failed to download %s video URL' % fmt[0],
False)
if not video_url:
continue
formats.append({
'url': video_url,
'ext': 'mp4',
'format_id': fmt[0],
})
2014-03-29 14:15:16 +00:00
return {
'id': real_id,
'display_id': display_id,
2014-03-29 14:15:16 +00:00
'title': first_chapter['title'],
'thumbnail': first_chapter['preview'],
'description': first_chapter['description'],
'view_count': video_info['views'],
'upload_date': upload_date,
'duration': first_file['duration'],
'formats': formats,
2014-03-29 14:15:16 +00:00
}