[telecinco] fix info extraction

youtube_dl/extractor/mitele.py

@@ -12,12 +12,69 @@ from ..utils import (
     get_element_by_attribute,
     int_or_none,
     remove_start,
+    extract_attributes,
+    determine_ext,
 )
 
 
-class MiTeleIE(InfoExtractor):
+class MiTeleBaseIE(InfoExtractor):
+    def _get_player_info(self, url, webpage):
+        player_data = extract_attributes(self._search_regex(
+            r'(?s)(<ms-video-player.+?</ms-video-player>)',
+            webpage, 'ms video player'))
+        video_id = player_data['data-media-id']
+        config_url = compat_urlparse.urljoin(url, player_data['data-config'])
+        config = self._download_json(
+            config_url, video_id, 'Downloading config JSON')
+        mmc_url = config['services']['mmc']
+
+        duration = None
+        formats = []
+        for m_url in (mmc_url, mmc_url.replace('/flash.json', '/html5.json')):
+            mmc = self._download_json(
+                m_url, video_id, 'Downloading mmc JSON')
+            if not duration:
+                duration = int_or_none(mmc.get('duration'))
+            for location in mmc['locations']:
+                gat = self._proto_relative_url(location.get('gat'), 'http:')
+                bas = location.get('bas')
+                loc = location.get('loc')
+                ogn = location.get('ogn')
+                if None in (gat, bas, loc, ogn):
+                    continue
+                token_data = {
+                    'bas': bas,
+                    'icd': loc,
+                    'ogn': ogn,
+                    'sta': '0',
+                }
+                media = self._download_json(
+                    '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
+                    video_id, 'Downloading %s JSON' % location['loc'])
+                file_ = media.get('file')
+                if not file_:
+                    continue
+                ext = determine_ext(file_)
+                if ext == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
+                        video_id, f4m_id='hds', fatal=False))
+                elif ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        file_, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'thumbnail': player_data.get('data-poster') or config.get('poster', {}).get('imageUrl'),
+            'duration': duration,
+        }
+
+
+class MiTeleIE(MiTeleBaseIE):
     IE_DESC = 'mitele.es'
-    _VALID_URL = r'https?://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
+    _VALID_URL = r'https?://www\.mitele\.es/(?:[^/]+/){3}(?P<id>[^/]+)/'
 
     _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
@@ -25,7 +82,7 @@ class MiTeleIE(InfoExtractor):
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Tor, la web invisible',
             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
             'series': 'Diario de',
@@ -40,7 +97,7 @@ class MiTeleIE(InfoExtractor):
         'info_dict': {
             'id': 'eLZSwoEd1S3pVyUm8lc6F',
             'display_id': 'programa-226',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Cuarto Milenio - Temporada 6 - Programa 226',
             'description': 'md5:50daf9fadefa4e62d9fc866d0c015701',
             'series': 'Cuarto Milenio',
@@ -59,40 +116,7 @@ class MiTeleIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        config_url = self._search_regex(
-            r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url')
-        config_url = compat_urlparse.urljoin(url, config_url)
-
-        config = self._download_json(
-            config_url, display_id, 'Downloading config JSON')
-
-        mmc = self._download_json(
-            config['services']['mmc'], display_id, 'Downloading mmc JSON')
-
-        formats = []
-        for location in mmc['locations']:
-            gat = self._proto_relative_url(location.get('gat'), 'http:')
-            bas = location.get('bas')
-            loc = location.get('loc')
-            ogn = location.get('ogn')
-            if None in (gat, bas, loc, ogn):
-                continue
-            token_data = {
-                'bas': bas,
-                'icd': loc,
-                'ogn': ogn,
-                'sta': '0',
-            }
-            media = self._download_json(
-                '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
-                display_id, 'Downloading %s JSON' % location['loc'])
-            file_ = media.get('file')
-            if not file_:
-                continue
-            formats.extend(self._extract_f4m_formats(
-                file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
-                display_id, f4m_id=loc))
-        self._sort_formats(formats)
+        info = self._get_player_info(url, webpage)
 
         title = self._search_regex(
             r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>',
@@ -112,21 +136,12 @@ class MiTeleIE(InfoExtractor):
                 title = remove_start(self._search_regex(
                     r'<title>([^<]+)</title>', webpage, 'title'), 'Ver online ')
 
-        video_id = self._search_regex(
-            r'data-media-id\s*=\s*"([^"]+)"', webpage,
-            'data media id', default=None) or display_id
-        thumbnail = config.get('poster', {}).get('imageUrl')
-        duration = int_or_none(mmc.get('duration'))
-
-        return {
-            'id': video_id,
+        info.update({
             'display_id': display_id,
             'title': title,
             'description': get_element_by_attribute('class', 'text', webpage),
             'series': series,
             'season': season,
             'episode': episode,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-        }
+        })
+        return info

youtube_dl/extractor/telecinco.py

@@ -1,50 +1,41 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import json
+from .mitele import MiTeleBaseIE
 
-from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_unquote,
-    compat_urllib_parse_urlencode,
-    compat_urlparse,
-)
-from ..utils import (
-    get_element_by_attribute,
-    parse_duration,
-    strip_jsonp,
-)
 
-
-class TelecincoIE(InfoExtractor):
+class TelecincoIE(MiTeleBaseIE):
     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
     _VALID_URL = r'https?://www\.(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
 
     _TESTS = [{
         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
-        'md5': '5cbef3ad5ef17bf0d21570332d140729',
+        'md5': '8d7b2d5f699ee2709d992a63d5cd1712',
         'info_dict': {
-            'id': 'MDSVID20141015_0058',
+            'id': 'JEA5ijCnF6p5W08A1rNKn7',
             'ext': 'mp4',
-            'title': 'Con Martín Berasategui, hacer un bacalao al ...',
+            'title': 'Bacalao con kokotxas al pil-pil',
+            'description': 'md5:1382dacd32dd4592d478cbdca458e5bb',
             'duration': 662,
         },
     }, {
         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
-        'md5': '0a5b9f3cc8b074f50a0578f823a12694',
+        'md5': '284393e5387b3b947b77c613ef04749a',
         'info_dict': {
-            'id': 'MDSVID20150916_0128',
+            'id': 'jn24Od1zGLG4XUZcnUnZB6',
             'ext': 'mp4',
-            'title': '¿Quién es este ex futbolista con el que hablan ...',
+            'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?',
+            'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
             'duration': 79,
         },
     }, {
         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
-        'md5': 'ad1bfaaba922dd4a295724b05b68f86a',
+        'md5': '749afab6ea5a136a8806855166ae46a2',
         'info_dict': {
-            'id': 'MDSVID20150513_0220',
+            'id': 'aywerkD2Sv1vGNqq9b85Q2',
             'ext': 'mp4',
             'title': '#DOYLACARA. Con la trata no hay trato',
+            'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',
             'duration': 50,
         },
     }, {
@@ -56,40 +47,16 @@ class TelecincoIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        episode = self._match_id(url)
-        webpage = self._download_webpage(url, episode)
-        embed_data_json = self._search_regex(
-            r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
-        ).replace('\'', '"')
-        embed_data = json.loads(embed_data_json)
-
-        domain = embed_data['mediaUrl']
-        if not domain.startswith('http'):
-            # only happens in telecinco.es videos
-            domain = 'http://' + domain
-        info_url = compat_urlparse.urljoin(
-            domain,
-            compat_urllib_parse_unquote(embed_data['flashvars']['host'])
-        )
-        info_el = self._download_xml(info_url, episode).find('./video/info')
-
-        video_link = info_el.find('videoUrl/link').text
-        token_query = compat_urllib_parse_urlencode({'id': video_link})
-        token_info = self._download_json(
-            embed_data['flashvars']['ov_tk'] + '?' + token_query,
-            episode,
-            transform_source=strip_jsonp
-        )
-        formats = self._extract_m3u8_formats(
-            token_info['tokenizedUrl'], episode, ext='mp4', entry_protocol='m3u8_native')
-        self._sort_formats(formats)
-
-        return {
-            'id': embed_data['videoId'],
-            'display_id': episode,
-            'title': info_el.find('title').text,
-            'formats': formats,
-            'description': get_element_by_attribute('class', 'text', webpage),
-            'thumbnail': info_el.find('thumb').text,
-            'duration': parse_duration(info_el.find('duration').text),
-        }
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        title = self._html_search_meta(
+            ['og:title', 'twitter:title'], webpage, 'title')
+        info = self._get_player_info(url, webpage)
+        info.update({
+            'display_id': display_id,
+            'title': title,
+            'description': self._html_search_meta(
+                ['og:description', 'twitter:description'],
+                webpage, 'title', fatal=False),
+        })
+        return info