midgard/youtube-dl
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[itv] Make SOAP request non fatal and extract metadata from a webpage (closes #16780)

Browse source
This commit is contained in:
Sergey M․ 2018-06-21 23:06:58 +07:00
parent 91aa502d91
commit 30374f4d40
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 69 additions and 57 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

126
youtube_dl/extractor/itv.py
View file

@ -18,6 +18,7 @@ from ..utils import (
xpath_element, xpath_element,
xpath_text, xpath_text,
int_or_none, int_or_none,
merge_dicts,
parse_duration, parse_duration,
smuggle_url, smuggle_url,
ExtractorError, ExtractorError,
@ -129,64 +130,65 @@ class ITVIE(InfoExtractor):
resp_env = self._download_xml( resp_env = self._download_xml(
params['data-playlist-url'], video_id, params['data-playlist-url'], video_id,
headers=headers, data=etree.tostring(req_env)) headers=headers, data=etree.tostring(req_env), fatal=False)
playlist = xpath_element(resp_env, './/Playlist') if resp_env:
if playlist is None: playlist = xpath_element(resp_env, './/Playlist')
fault_code = xpath_text(resp_env, './/faultcode') if playlist is None:
fault_string = xpath_text(resp_env, './/faultstring') fault_code = xpath_text(resp_env, './/faultcode')
if fault_code == 'InvalidGeoRegion': fault_string = xpath_text(resp_env, './/faultstring')
self.raise_geo_restricted( if fault_code == 'InvalidGeoRegion':
msg=fault_string, countries=self._GEO_COUNTRIES) self.raise_geo_restricted(
elif fault_code not in ( msg=fault_string, countries=self._GEO_COUNTRIES)
'InvalidEntity', 'InvalidVodcrid', 'ContentUnavailable'): elif fault_code not in (
raise ExtractorError( 'InvalidEntity', 'InvalidVodcrid', 'ContentUnavailable'):
'%s said: %s' % (self.IE_NAME, fault_string), expected=True) raise ExtractorError(
info.update({ '%s said: %s' % (self.IE_NAME, fault_string), expected=True)
'title': self._og_search_title(webpage), info.update({
'episode_title': params.get('data-video-episode'), 'title': self._og_search_title(webpage),
'series': params.get('data-video-title'), 'episode_title': params.get('data-video-episode'),
}) 'series': params.get('data-video-title'),
else: })
title = xpath_text(playlist, 'EpisodeTitle', default=None) else:
info.update({ title = xpath_text(playlist, 'EpisodeTitle', default=None)
'title': title, info.update({
'episode_title': title, 'title': title,
'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')), 'episode_title': title,
'series': xpath_text(playlist, 'ProgrammeTitle'), 'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')),
'duration': parse_duration(xpath_text(playlist, 'Duration')), 'series': xpath_text(playlist, 'ProgrammeTitle'),
}) 'duration': parse_duration(xpath_text(playlist, 'Duration')),
video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True) })
media_files = xpath_element(video_element, 'MediaFiles', fatal=True) video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True)
rtmp_url = media_files.attrib['base'] media_files = xpath_element(video_element, 'MediaFiles', fatal=True)
rtmp_url = media_files.attrib['base']
for media_file in media_files.findall('MediaFile'): for media_file in media_files.findall('MediaFile'):
play_path = xpath_text(media_file, 'URL') play_path = xpath_text(media_file, 'URL')
if not play_path: if not play_path:
continue continue
tbr = int_or_none(media_file.get('bitrate'), 1000) tbr = int_or_none(media_file.get('bitrate'), 1000)
f = { f = {
'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''), 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''),
'play_path': play_path, 'play_path': play_path,
# Providing this swfVfy allows to avoid truncated downloads # Providing this swfVfy allows to avoid truncated downloads
'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf', 'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf',
'page_url': url, 'page_url': url,
'tbr': tbr, 'tbr': tbr,
'ext': 'flv', 'ext': 'flv',
} }
app = self._search_regex( app = self._search_regex(
'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None) 'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None)
if app: if app:
f.update({ f.update({
'url': rtmp_url.split('?', 1)[0], 'url': rtmp_url.split('?', 1)[0],
'app': app, 'app': app,
}) })
else: else:
f['url'] = rtmp_url f['url'] = rtmp_url
formats.append(f) formats.append(f)
for caption_url in video_element.findall('ClosedCaptioningURIs/URL'): for caption_url in video_element.findall('ClosedCaptioningURIs/URL'):
if caption_url.text: if caption_url.text:
extract_subtitle(caption_url.text) extract_subtitle(caption_url.text)
ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id') ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id')
hmac = params.get('data-video-hmac') hmac = params.get('data-video-hmac')
@ -261,7 +263,17 @@ class ITVIE(InfoExtractor):
'formats': formats, 'formats': formats,
'subtitles': subtitles, 'subtitles': subtitles,
}) })
return info
webpage_info = self._search_json_ld(webpage, video_id, default={})
if not webpage_info.get('title'):
webpage_info['title'] = self._html_search_regex(
r'(?s)<h\d+[^>]+\bclass=["\'][^>]*episode-title["\'][^>]*>([^<]+)<',
webpage, 'title', default=None) or self._og_search_title(
webpage, default=None) or self._html_search_meta(
'twitter:title', webpage, 'title',
default=None) or webpage_info['episode']
return merge_dicts(info, webpage_info)
class ITVBTCCIE(InfoExtractor): class ITVBTCCIE(InfoExtractor):