[toutv] fix info extraction(closes #1792)(closes #2082)

This commit is contained in:
remitamine 2016-05-24 16:06:02 +01:00 committed by Remita Amine
parent 47335a0efa
commit 882af14d7d
2 changed files with 29 additions and 59 deletions

View file

@ -22,13 +22,13 @@ class RadioCanadaIE(InfoExtractor):
'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
'info_dict': {
'id': '7184272',
'ext': 'flv',
'ext': 'mp4',
'title': 'Le parcours du tireur capté sur vidéo',
'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
'upload_date': '20141023',
},
'params': {
# rtmp download
# m3u8 download
'skip_download': True,
},
}
@ -36,11 +36,14 @@ class RadioCanadaIE(InfoExtractor):
def _real_extract(self, url):
app_code, video_id = re.match(self._VALID_URL, url).groups()
device_types = ['ipad']
if app_code != 'toutv':
device_types.append('flash')
formats = []
# TODO: extract m3u8 and f4m formats
# m3u8 formats can be extracted using ipad device_type return 403 error code when ffmpeg try to download segements
# TODO: extract f4m formats
# f4m formats can be extracted using flashhd device_type but they produce unplayable file
for device_type in ('flash',):
for device_type in device_types:
v_data = self._download_xml(
'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
video_id, note='Downloading %s XML' % device_type, query={
@ -115,13 +118,13 @@ class RadioCanadaAudioVideoIE(InfoExtractor):
'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
'info_dict': {
'id': '7527184',
'ext': 'flv',
'ext': 'mp4',
'title': 'Barack Obama au Vietnam',
'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
'upload_date': '20160523',
},
'params': {
# rtmp download
# m3u8 download
'skip_download': True,
},
}

View file

@ -1,74 +1,41 @@
# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import (
ExtractorError,
unified_strdate,
)
from ..utils import int_or_none
class TouTvIE(InfoExtractor):
IE_NAME = 'tou.tv'
_VALID_URL = r'https?://www\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/(?P<episode>S[0-9]+E[0-9]+)))'
_VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+/S[0-9]+E[0-9]+)'
_TEST = {
'url': 'http://www.tou.tv/30-vies/S04E41',
'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
'info_dict': {
'id': '30-vies_S04E41',
'id': '122017',
'ext': 'mp4',
'title': '30 vies Saison 4 / Épisode 41',
'description': 'md5:da363002db82ccbe4dafeb9cab039b09',
'age_limit': 8,
'uploader': 'Groupe des Nouveaux Médias',
'duration': 1296,
'upload_date': '20131118',
'thumbnail': 'http://static.tou.tv/medias/images/2013-11-18_19_00_00_30VIES_0341_01_L.jpeg',
'title': 'Saison 2015 Épisode 17',
'description': 'La photo de famille 2',
'upload_date': '20100717',
},
'params': {
'skip_download': True, # Requires rtmpdump
# m3u8 download
'skip_download': True,
},
'skip': 'Only available in Canada'
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id)
mediaId = self._search_regex(
r'"idMedia":\s*"([^"]+)"', webpage, 'media ID')
streams_url = 'http://release.theplatform.com/content.select?pid=' + mediaId
streams_doc = self._download_xml(
streams_url, video_id, note='Downloading stream list')
video_url = next(n.text
for n in streams_doc.findall('.//choice/url')
if '//ad.doubleclick' not in n.text)
if video_url.endswith('/Unavailable.flv'):
raise ExtractorError(
'Access to this video is blocked from outside of Canada',
expected=True)
duration_str = self._html_search_meta(
'video:duration', webpage, 'duration')
duration = int(duration_str) if duration_str else None
upload_date_str = self._html_search_meta(
'video:release_date', webpage, 'upload date')
upload_date = unified_strdate(upload_date_str) if upload_date_str else None
path = self._match_id(url)
metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
video_id = metadata['IdMedia']
details = metadata['Details']
title = details['OriginalTitle']
return {
'_type': 'url_transparent',
'url': 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id),
'id': video_id,
'title': self._og_search_title(webpage),
'url': video_url,
'description': self._og_search_description(webpage),
'uploader': self._dc_search_uploader(webpage),
'thumbnail': self._og_search_thumbnail(webpage),
'age_limit': self._media_rating_search(webpage),
'duration': duration,
'upload_date': upload_date,
'ext': 'mp4',
'title': title,
'thumbnail': details.get('ImageUrl'),
'duration': int_or_none(details.get('LengthInSeconds')),
}