2013-10-13 14:21:13 +02:00
|
|
|
|
# encoding: utf-8
|
2014-01-07 10:04:48 +01:00
|
|
|
|
from __future__ import unicode_literals
|
|
|
|
|
|
2013-06-23 20:24:07 +02:00
|
|
|
|
import re
|
|
|
|
|
|
|
|
|
|
from .common import InfoExtractor
|
|
|
|
|
from ..utils import (
|
2013-07-11 16:12:16 +02:00
|
|
|
|
find_xpath_attr,
|
2013-06-23 20:24:07 +02:00
|
|
|
|
unified_strdate,
|
2013-10-13 14:21:13 +02:00
|
|
|
|
get_element_by_id,
|
2013-12-08 14:02:14 +01:00
|
|
|
|
get_element_by_attribute,
|
2014-10-20 15:27:59 +02:00
|
|
|
|
int_or_none,
|
2014-11-20 12:06:33 +01:00
|
|
|
|
qualities,
|
2013-06-23 20:24:07 +02:00
|
|
|
|
)
|
|
|
|
|
|
2014-11-23 20:41:03 +01:00
|
|
|
|
# There are different sources of video in arte.tv, the extraction process
|
2013-10-13 13:54:31 +02:00
|
|
|
|
# is different for each one. The videos usually expire in 7 days, so we can't
|
|
|
|
|
# add tests.
|
|
|
|
|
|
2013-06-23 20:24:07 +02:00
|
|
|
|
|
2014-03-24 21:30:50 +01:00
|
|
|
|
class ArteTvIE(InfoExtractor):
|
2014-03-24 21:36:26 +01:00
|
|
|
|
_VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
|
2014-01-07 10:04:48 +01:00
|
|
|
|
IE_NAME = 'arte.tv'
|
2013-06-23 20:24:07 +02:00
|
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
2014-03-24 22:34:03 +01:00
|
|
|
|
mobj = re.match(self._VALID_URL, url)
|
|
|
|
|
lang = mobj.group('lang')
|
|
|
|
|
video_id = mobj.group('id')
|
|
|
|
|
|
2013-07-04 18:06:47 +02:00
|
|
|
|
ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
|
|
|
|
|
ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
|
2014-03-08 16:04:03 +01:00
|
|
|
|
ref_xml_doc = self._download_xml(
|
|
|
|
|
ref_xml_url, video_id, note='Downloading metadata')
|
2013-07-11 16:12:16 +02:00
|
|
|
|
config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
|
2013-07-04 18:06:47 +02:00
|
|
|
|
config_xml_url = config_node.attrib['ref']
|
2014-03-24 21:36:26 +01:00
|
|
|
|
config = self._download_xml(
|
2014-03-08 16:04:03 +01:00
|
|
|
|
config_xml_url, video_id, note='Downloading configuration')
|
2013-06-30 13:38:22 +02:00
|
|
|
|
|
2014-03-24 21:36:26 +01:00
|
|
|
|
formats = [{
|
2014-12-28 10:42:29 +01:00
|
|
|
|
'format_id': q.attrib['quality'],
|
2014-07-07 14:10:57 +02:00
|
|
|
|
# The playpath starts at 'mp4:', if we don't manually
|
|
|
|
|
# split the url, rtmpdump will incorrectly parse them
|
|
|
|
|
'url': q.text.split('mp4:', 1)[0],
|
|
|
|
|
'play_path': 'mp4:' + q.text.split('mp4:', 1)[1],
|
2014-03-24 22:38:51 +01:00
|
|
|
|
'ext': 'flv',
|
2014-03-24 21:36:26 +01:00
|
|
|
|
'quality': 2 if q.attrib['quality'] == 'hd' else 1,
|
2014-03-24 22:38:51 +01:00
|
|
|
|
} for q in config.findall('./urls/url')]
|
2014-03-24 21:36:26 +01:00
|
|
|
|
self._sort_formats(formats)
|
|
|
|
|
|
|
|
|
|
title = config.find('.//name').text
|
|
|
|
|
thumbnail = config.find('.//firstThumbnailUrl').text
|
|
|
|
|
return {
|
|
|
|
|
'id': video_id,
|
|
|
|
|
'title': title,
|
|
|
|
|
'thumbnail': thumbnail,
|
2014-03-24 22:34:03 +01:00
|
|
|
|
'formats': formats,
|
2014-03-24 21:30:50 +01:00
|
|
|
|
}
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
class ArteTVPlus7IE(InfoExtractor):
|
2014-01-07 10:04:48 +01:00
|
|
|
|
IE_NAME = 'arte.tv:+7'
|
2014-03-19 15:23:55 +01:00
|
|
|
|
_VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
2013-10-13 14:21:13 +02:00
|
|
|
|
@classmethod
|
|
|
|
|
def _extract_url_info(cls, url):
|
|
|
|
|
mobj = re.match(cls._VALID_URL, url)
|
2013-10-13 13:54:31 +02:00
|
|
|
|
lang = mobj.group('lang')
|
|
|
|
|
# This is not a real id, it can be for example AJT for the news
|
|
|
|
|
# http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
|
|
|
|
|
video_id = mobj.group('id')
|
2013-10-13 14:21:13 +02:00
|
|
|
|
return video_id, lang
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
2013-10-13 14:21:13 +02:00
|
|
|
|
def _real_extract(self, url):
|
|
|
|
|
video_id, lang = self._extract_url_info(url)
|
2013-10-13 13:54:31 +02:00
|
|
|
|
webpage = self._download_webpage(url, video_id)
|
2013-10-13 14:21:13 +02:00
|
|
|
|
return self._extract_from_webpage(webpage, video_id, lang)
|
|
|
|
|
|
|
|
|
|
def _extract_from_webpage(self, webpage, video_id, lang):
|
2014-04-21 07:59:16 +02:00
|
|
|
|
json_url = self._html_search_regex(
|
2014-09-04 20:17:16 +02:00
|
|
|
|
[r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],
|
2014-09-04 15:44:51 +02:00
|
|
|
|
webpage, 'json vp url')
|
2013-12-08 14:02:14 +01:00
|
|
|
|
return self._extract_from_json_url(json_url, video_id, lang)
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
2013-12-08 14:02:14 +01:00
|
|
|
|
def _extract_from_json_url(self, json_url, video_id, lang):
|
2014-03-24 22:01:47 +01:00
|
|
|
|
info = self._download_json(json_url, video_id)
|
2013-10-13 13:54:31 +02:00
|
|
|
|
player_info = info['videoJsonPlayer']
|
|
|
|
|
|
2014-09-29 12:45:18 +02:00
|
|
|
|
upload_date_str = player_info.get('shootingDate')
|
|
|
|
|
if not upload_date_str:
|
|
|
|
|
upload_date_str = player_info.get('VDA', '').split(' ')[0]
|
|
|
|
|
|
2014-10-21 15:08:20 +02:00
|
|
|
|
title = player_info['VTI'].strip()
|
|
|
|
|
subtitle = player_info.get('VSU', '').strip()
|
|
|
|
|
if subtitle:
|
|
|
|
|
title += ' - %s' % subtitle
|
|
|
|
|
|
2013-10-13 13:54:31 +02:00
|
|
|
|
info_dict = {
|
|
|
|
|
'id': player_info['VID'],
|
2014-10-21 15:08:20 +02:00
|
|
|
|
'title': title,
|
2013-10-13 13:54:31 +02:00
|
|
|
|
'description': player_info.get('VDE'),
|
2014-09-29 12:45:18 +02:00
|
|
|
|
'upload_date': unified_strdate(upload_date_str),
|
2013-10-13 13:54:31 +02:00
|
|
|
|
'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
|
|
|
|
|
}
|
2014-11-20 12:06:33 +01:00
|
|
|
|
qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
2014-11-20 12:06:33 +01:00
|
|
|
|
formats = []
|
2014-10-20 15:27:59 +02:00
|
|
|
|
for format_id, format_dict in player_info['VSR'].items():
|
2014-11-20 12:06:33 +01:00
|
|
|
|
f = dict(format_dict)
|
|
|
|
|
versionCode = f.get('versionCode')
|
|
|
|
|
|
|
|
|
|
langcode = {
|
|
|
|
|
'fr': 'F',
|
|
|
|
|
'de': 'A',
|
|
|
|
|
}.get(lang, lang)
|
|
|
|
|
lang_rexs = [r'VO?%s' % langcode, r'VO?.-ST%s' % langcode]
|
|
|
|
|
lang_pref = (
|
|
|
|
|
None if versionCode is None else (
|
|
|
|
|
10 if any(re.match(r, versionCode) for r in lang_rexs)
|
|
|
|
|
else -10))
|
|
|
|
|
source_pref = 0
|
|
|
|
|
if versionCode is not None:
|
|
|
|
|
# The original version with subtitles has lower relevance
|
|
|
|
|
if re.match(r'VO-ST(F|A)', versionCode):
|
|
|
|
|
source_pref -= 10
|
|
|
|
|
# The version with sourds/mal subtitles has also lower relevance
|
|
|
|
|
elif re.match(r'VO?(F|A)-STM\1', versionCode):
|
|
|
|
|
source_pref -= 9
|
|
|
|
|
format = {
|
|
|
|
|
'format_id': format_id,
|
|
|
|
|
'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
|
|
|
|
|
'language_preference': lang_pref,
|
|
|
|
|
'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
|
|
|
|
|
'width': int_or_none(f.get('width')),
|
|
|
|
|
'height': int_or_none(f.get('height')),
|
|
|
|
|
'tbr': int_or_none(f.get('bitrate')),
|
2014-12-28 10:41:52 +01:00
|
|
|
|
'quality': qfunc(f.get('quality')),
|
2014-11-20 12:06:33 +01:00
|
|
|
|
'source_preference': source_pref,
|
2013-10-13 13:54:31 +02:00
|
|
|
|
}
|
2014-11-20 12:06:33 +01:00
|
|
|
|
|
|
|
|
|
if f.get('mediaType') == 'rtmp':
|
|
|
|
|
format['url'] = f['streamer']
|
|
|
|
|
format['play_path'] = 'mp4:' + f['url']
|
|
|
|
|
format['ext'] = 'flv'
|
2013-10-13 13:54:31 +02:00
|
|
|
|
else:
|
2014-11-20 12:06:33 +01:00
|
|
|
|
format['url'] = f['url']
|
|
|
|
|
|
|
|
|
|
formats.append(format)
|
|
|
|
|
|
2015-03-17 14:42:50 +01:00
|
|
|
|
self._check_formats(formats, video_id)
|
2014-11-20 12:06:33 +01:00
|
|
|
|
self._sort_formats(formats)
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
2014-11-20 12:06:33 +01:00
|
|
|
|
info_dict['formats'] = formats
|
2013-10-13 13:54:31 +02:00
|
|
|
|
return info_dict
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# It also uses the arte_vp_url url from the webpage to extract the information
|
|
|
|
|
class ArteTVCreativeIE(ArteTVPlus7IE):
|
2014-01-07 10:04:48 +01:00
|
|
|
|
IE_NAME = 'arte.tv:creative'
|
2014-08-24 02:57:31 +02:00
|
|
|
|
_VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/(?:magazine?/)?(?P<id>[^?#]+)'
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
2014-08-24 02:57:31 +02:00
|
|
|
|
_TESTS = [{
|
2014-01-07 10:04:48 +01:00
|
|
|
|
'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
|
|
|
|
|
'info_dict': {
|
2014-08-24 02:57:31 +02:00
|
|
|
|
'id': '72176',
|
2014-03-20 09:14:43 +01:00
|
|
|
|
'ext': 'mp4',
|
2014-08-24 02:57:31 +02:00
|
|
|
|
'title': 'Folge 2 - Corporate Design',
|
|
|
|
|
'upload_date': '20131004',
|
2013-10-13 13:54:31 +02:00
|
|
|
|
},
|
2014-08-24 02:57:31 +02:00
|
|
|
|
}, {
|
|
|
|
|
'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'id': '160676',
|
|
|
|
|
'ext': 'mp4',
|
|
|
|
|
'title': 'Monty Python live (mostly)',
|
2014-08-24 03:11:00 +02:00
|
|
|
|
'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
|
|
|
|
|
'upload_date': '20140805',
|
2014-08-24 02:57:31 +02:00
|
|
|
|
}
|
|
|
|
|
}]
|
2013-10-13 13:54:31 +02:00
|
|
|
|
|
2013-10-13 14:21:13 +02:00
|
|
|
|
|
|
|
|
|
class ArteTVFutureIE(ArteTVPlus7IE):
|
2014-01-07 10:04:48 +01:00
|
|
|
|
IE_NAME = 'arte.tv:future'
|
2013-10-13 14:21:13 +02:00
|
|
|
|
_VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
|
|
|
|
|
|
|
|
|
|
_TEST = {
|
2014-01-07 10:04:48 +01:00
|
|
|
|
'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
|
|
|
|
|
'info_dict': {
|
2014-06-28 10:22:53 +02:00
|
|
|
|
'id': '5201',
|
2014-03-20 09:14:43 +01:00
|
|
|
|
'ext': 'mp4',
|
2014-01-07 10:04:48 +01:00
|
|
|
|
'title': 'Les champignons au secours de la planète',
|
2014-06-28 10:22:53 +02:00
|
|
|
|
'upload_date': '20131101',
|
2013-10-13 14:21:13 +02:00
|
|
|
|
},
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
|
anchor_id, lang = self._extract_url_info(url)
|
|
|
|
|
webpage = self._download_webpage(url, anchor_id)
|
2015-05-24 17:25:00 +02:00
|
|
|
|
row = self._search_regex(
|
|
|
|
|
r'(?s)id="%s"[^>]*>.+?(<div[^>]*arte_vp_url[^>]*>)' % anchor_id,
|
|
|
|
|
webpage, 'row')
|
2013-10-13 14:21:13 +02:00
|
|
|
|
return self._extract_from_webpage(row, anchor_id, lang)
|
2013-12-08 14:02:14 +01:00
|
|
|
|
|
2013-12-08 16:34:57 +01:00
|
|
|
|
|
2013-12-08 14:02:14 +01:00
|
|
|
|
class ArteTVDDCIE(ArteTVPlus7IE):
|
2014-01-07 10:04:48 +01:00
|
|
|
|
IE_NAME = 'arte.tv:ddc'
|
2014-03-20 09:14:43 +01:00
|
|
|
|
_VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
|
2013-12-08 14:02:14 +01:00
|
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
|
video_id, lang = self._extract_url_info(url)
|
|
|
|
|
if lang == 'folge':
|
|
|
|
|
lang = 'de'
|
|
|
|
|
elif lang == 'emission':
|
|
|
|
|
lang = 'fr'
|
|
|
|
|
webpage = self._download_webpage(url, video_id)
|
|
|
|
|
scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
|
|
|
|
|
script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
|
|
|
|
|
javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
|
|
|
|
|
json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
|
|
|
|
|
return self._extract_from_json_url(json_url, video_id, lang)
|
2014-03-20 09:09:56 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
class ArteTVConcertIE(ArteTVPlus7IE):
|
|
|
|
|
IE_NAME = 'arte.tv:concert'
|
|
|
|
|
_VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
|
|
|
|
|
|
|
|
|
|
_TEST = {
|
|
|
|
|
'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
|
|
|
|
|
'md5': '9ea035b7bd69696b67aa2ccaaa218161',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'id': '186',
|
|
|
|
|
'ext': 'mp4',
|
|
|
|
|
'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
|
|
|
|
|
'upload_date': '20140128',
|
2014-03-22 14:20:05 +01:00
|
|
|
|
'description': 'md5:486eb08f991552ade77439fe6d82c305',
|
2014-03-20 09:09:56 +01:00
|
|
|
|
},
|
|
|
|
|
}
|
2014-03-24 22:01:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
class ArteTVEmbedIE(ArteTVPlus7IE):
|
|
|
|
|
IE_NAME = 'arte.tv:embed'
|
|
|
|
|
_VALID_URL = r'''(?x)
|
|
|
|
|
http://www\.arte\.tv
|
|
|
|
|
/playerv2/embed\.php\?json_url=
|
|
|
|
|
(?P<json_url>
|
|
|
|
|
http://arte\.tv/papi/tvguide/videos/stream/player/
|
|
|
|
|
(?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
|
|
|
|
|
)
|
|
|
|
|
'''
|
|
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
|
mobj = re.match(self._VALID_URL, url)
|
|
|
|
|
video_id = mobj.group('id')
|
|
|
|
|
lang = mobj.group('lang')
|
|
|
|
|
json_url = mobj.group('json_url')
|
|
|
|
|
return self._extract_from_json_url(json_url, video_id, lang)
|