[mitele] Fix extraction after website redesign (fixes #10824)

This commit is contained in:
DarkZeros 2016-10-04 02:04:24 +01:00 committed by Jaime Marquínez Ferrándiz
parent 98708e6cbd
commit cb882540e8
3 changed files with 63 additions and 43 deletions

View File

@ -1,3 +1,9 @@
version <unreleased>
Extractors
* [mitele] Fix extraction after website redesign (#10824)
version 2016.11.04 version 2016.11.04
Core Core

View File

@ -1,19 +1,20 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re import uuid
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import (
compat_str,
compat_urllib_parse_urlencode, compat_urllib_parse_urlencode,
compat_urlparse, compat_urlparse,
) )
from ..utils import ( from ..utils import (
get_element_by_attribute,
int_or_none, int_or_none,
remove_start,
extract_attributes, extract_attributes,
determine_ext, determine_ext,
smuggle_url,
parse_duration,
) )
@ -72,16 +73,14 @@ class MiTeleBaseIE(InfoExtractor):
} }
class MiTeleIE(MiTeleBaseIE): class MiTeleIE(InfoExtractor):
IE_DESC = 'mitele.es' IE_DESC = 'mitele.es'
_VALID_URL = r'https?://(?:www\.)?mitele\.es/(?:[^/]+/){3}(?P<id>[^/]+)/' _VALID_URL = r'https?://(?:www\.)?mitele\.es/programas-tv/(?:[^/]+/)(?P<id>[^/]+)/player'
_TESTS = [{ _TESTS = [{
'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/', 'url': 'http://www.mitele.es/programas-tv/diario-de/57b0dfb9c715da65618b4afa/player',
# MD5 is unstable
'info_dict': { 'info_dict': {
'id': '0NF1jJnxS1Wu3pHrmvFyw2', 'id': '57b0dfb9c715da65618b4afa',
'display_id': 'programa-144',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Tor, la web invisible', 'title': 'Tor, la web invisible',
'description': 'md5:3b6fce7eaa41b2d97358726378d9369f', 'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
@ -91,57 +90,71 @@ class MiTeleIE(MiTeleBaseIE):
'thumbnail': 're:(?i)^https?://.*\.jpg$', 'thumbnail': 're:(?i)^https?://.*\.jpg$',
'duration': 2913, 'duration': 2913,
}, },
'add_ie': ['Ooyala'],
}, { }, {
# no explicit title # no explicit title
'url': 'http://www.mitele.es/programas-tv/cuarto-milenio/temporada-6/programa-226/', 'url': 'http://www.mitele.es/programas-tv/cuarto-milenio/57b0de3dc915da14058b4876/player',
'info_dict': { 'info_dict': {
'id': 'eLZSwoEd1S3pVyUm8lc6F', 'id': '57b0de3dc915da14058b4876',
'display_id': 'programa-226',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Cuarto Milenio - Temporada 6 - Programa 226', 'title': 'Cuarto Milenio Temporada 6 Programa 226',
'description': 'md5:50daf9fadefa4e62d9fc866d0c015701', 'description': 'md5:5ff132013f0cd968ffbf1f5f3538a65f',
'series': 'Cuarto Milenio', 'series': 'Cuarto Milenio',
'season': 'Temporada 6', 'season': 'Temporada 6',
'episode': 'Programa 226', 'episode': 'Programa 226',
'thumbnail': 're:(?i)^https?://.*\.jpg$', 'thumbnail': 're:(?i)^https?://.*\.jpg$',
'duration': 7312, 'duration': 7313,
}, },
'params': { 'params': {
'skip_download': True, 'skip_download': True,
}, },
'add_ie': ['Ooyala'],
}] }]
def _real_extract(self, url): def _real_extract(self, url):
display_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
webpage = self._download_webpage(url, display_id) gigya_url = self._search_regex(r'<gigya-api>[^>]*</gigya-api>[^>]*<script\s*src="([^"]*)">[^>]*</script>', webpage, 'gigya', default=None)
gigya_sc = self._download_webpage(compat_urlparse.urljoin(r'http://www.mitele.es/', gigya_url), video_id, 'Downloading gigya script')
# Get a appKey/uuid for getting the session key
appKey_var = self._search_regex(r'value\("appGridApplicationKey",([0-9a-f]+)\)', gigya_sc, 'appKey variable')
appKey = self._search_regex(r'var %s="([0-9a-f]+)"' % appKey_var, gigya_sc, 'appKey')
uid = compat_str(uuid.uuid4())
session_url = 'https://appgrid-api.cloud.accedo.tv/session?appKey=%s&uuid=%s' % (appKey, uid)
session_json = self._download_json(session_url, video_id, 'Downloading session keys')
sessionKey = compat_str(session_json['sessionKey'])
info = self._get_player_info(url, webpage) paths_url = 'https://appgrid-api.cloud.accedo.tv/metadata/general_configuration,%20web_configuration?sessionKey=' + sessionKey
paths = self._download_json(paths_url, video_id, 'Downloading paths JSON')
ooyala_s = paths['general_configuration']['api_configuration']['ooyala_search']
data_p = (
'http://' + ooyala_s['base_url'] + ooyala_s['full_path'] + ooyala_s['provider_id'] +
'/docs/' + video_id + '?include_titles=Series,Season&product_name=test&format=full')
data = self._download_json(data_p, video_id, 'Downloading data JSON')
source = data['hits']['hits'][0]['_source']
embedCode = source['offers'][0]['embed_codes'][0]
title = self._search_regex( titles = source['localizable_titles'][0]
r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', title = titles.get('title_medium') or titles['title_long']
webpage, 'title', default=None) episode = titles['title_sort_name']
description = titles['summary_long']
titles_series = source['localizable_titles_series'][0]
series = titles_series['title_long']
titles_season = source['localizable_titles_season'][0]
season = titles_season['title_medium']
duration = parse_duration(source['videos'][0]['duration'])
mobj = re.search(r'''(?sx) return {
class="Destacado-text"[^>]*>.*?<h1>\s* '_type': 'url_transparent',
<span>(?P<series>[^<]+)</span>\s* # for some reason only HLS is supported
<span>(?P<season>[^<]+)</span>\s* 'url': smuggle_url('ooyala:' + embedCode, {'supportedformats': 'm3u8'}),
<span>(?P<episode>[^<]+)</span>''', webpage) 'id': video_id,
series, season, episode = mobj.groups() if mobj else [None] * 3
if not title:
if mobj:
title = '%s - %s - %s' % (series, season, episode)
else:
title = remove_start(self._search_regex(
r'<title>([^<]+)</title>', webpage, 'title'), 'Ver online ')
info.update({
'display_id': display_id,
'title': title, 'title': title,
'description': get_element_by_attribute('class', 'text', webpage), 'description': description,
'series': series, 'series': series,
'season': season, 'season': season,
'episode': episode, 'episode': episode,
}) 'duration': duration,
return info 'thumbnail': source['images'][0]['url'],
}

View File

@ -18,7 +18,7 @@ class OoyalaBaseIE(InfoExtractor):
_CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/' _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
_AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s?' _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s?'
def _extract(self, content_tree_url, video_id, domain='example.org'): def _extract(self, content_tree_url, video_id, domain='example.org', supportedformats=None):
content_tree = self._download_json(content_tree_url, video_id)['content_tree'] content_tree = self._download_json(content_tree_url, video_id)['content_tree']
metadata = content_tree[list(content_tree)[0]] metadata = content_tree[list(content_tree)[0]]
embed_code = metadata['embed_code'] embed_code = metadata['embed_code']
@ -29,7 +29,7 @@ class OoyalaBaseIE(InfoExtractor):
self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) + self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
compat_urllib_parse_urlencode({ compat_urllib_parse_urlencode({
'domain': domain, 'domain': domain,
'supportedFormats': 'mp4,rtmp,m3u8,hds', 'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds',
}), video_id) }), video_id)
cur_auth_data = auth_data['authorization_data'][embed_code] cur_auth_data = auth_data['authorization_data'][embed_code]
@ -145,8 +145,9 @@ class OoyalaIE(OoyalaBaseIE):
url, smuggled_data = unsmuggle_url(url, {}) url, smuggled_data = unsmuggle_url(url, {})
embed_code = self._match_id(url) embed_code = self._match_id(url)
domain = smuggled_data.get('domain') domain = smuggled_data.get('domain')
supportedformats = smuggled_data.get('supportedformats')
content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code) content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
return self._extract(content_tree_url, embed_code, domain) return self._extract(content_tree_url, embed_code, domain, supportedformats)
class OoyalaExternalIE(OoyalaBaseIE): class OoyalaExternalIE(OoyalaBaseIE):