Update changelog.

[youtubedl] / youtube_dl / extractor / mitele.py

diff --git a/youtube_dl/extractor/mitele.py b/youtube_dl/extractor/mitele.py
index d8897eb90d526b7b7d2e5a5ace5bec84ebb40031..5a00cd3975b46fbd0f191ecd8e82fb09ea78490a 100644 (file)
--- a/youtube_dl/extractor/mitele.py
+++ b/youtube_dl/extractor/mitele.py
@@ -1,68 +1,132 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import json
+import re

 
 from .common import InfoExtractor
 from ..compat import (
 
 from .common import InfoExtractor
 from ..compat import (

-    compat_urllib_parse,
+    compat_urllib_parse_urlencode,

     compat_urlparse,
 )
 from ..utils import (
     get_element_by_attribute,
     compat_urlparse,
 )
 from ..utils import (
     get_element_by_attribute,

-    parse_duration,
-    strip_jsonp,
+    int_or_none,
+    remove_start,

 )
 
 
 class MiTeleIE(InfoExtractor):
 )
 
 
 class MiTeleIE(InfoExtractor):

-    IE_NAME = 'mitele.es'
-    _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
+    IE_DESC = 'mitele.es'
+    _VALID_URL = r'https?://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'

 
     _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
 
     _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',

-        'md5': '6a75fe9d0d3275bead0cb683c616fddb',
+        # MD5 is unstable

         'info_dict': {
         'info_dict': {

-            'id': '0fce117d',
-            'ext': 'mp4',
-            'title': 'Programa 144 - Tor, la web invisible',
-            'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
+            'id': '0NF1jJnxS1Wu3pHrmvFyw2',

             'display_id': 'programa-144',
             'display_id': 'programa-144',

+            'ext': 'flv',
+            'title': 'Tor, la web invisible',
+            'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
+            'series': 'Diario de',
+            'season': 'La redacción',
+            'episode': 'Programa 144',
+            'thumbnail': 're:(?i)^https?://.*\.jpg$',

             'duration': 2913,
         },
             'duration': 2913,
         },

+    }, {
+        # no explicit title
+        'url': 'http://www.mitele.es/programas-tv/cuarto-milenio/temporada-6/programa-226/',
+        'info_dict': {
+            'id': 'eLZSwoEd1S3pVyUm8lc6F',
+            'display_id': 'programa-226',
+            'ext': 'flv',
+            'title': 'Cuarto Milenio - Temporada 6 - Programa 226',
+            'description': 'md5:50daf9fadefa4e62d9fc866d0c015701',
+            'series': 'Cuarto Milenio',
+            'season': 'Temporada 6',
+            'episode': 'Programa 226',
+            'thumbnail': 're:(?i)^https?://.*\.jpg$',
+            'duration': 7312,
+        },
+        'params': {
+            'skip_download': True,
+        },

     }]
 
     def _real_extract(self, url):
     }]
 
     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
-        )
+        display_id = self._match_id(url)
+
+        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)
+
+        title = self._search_regex(
+            r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>',
+            webpage, 'title', default=None)
+
+        mobj = re.search(r'''(?sx)
+                            class="Destacado-text"[^>]*>.*?<h1>\s*
+                            <span>(?P<series>[^<]+)</span>\s*
+                            <span>(?P<season>[^<]+)</span>\s*
+                            <span>(?P<episode>[^<]+)</span>''', webpage)
+        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 ')
+
+        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 {
 
         return {

-            'id': embed_data['videoId'],
-            'display_id': episode,
-            'title': info_el.find('title').text,
-            'url': token_info['tokenizedUrl'],
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,

             'description': get_element_by_attribute('class', 'text', webpage),
             'description': get_element_by_attribute('class', 'text', webpage),

-            'thumbnail': info_el.find('thumb').text,
-            'duration': parse_duration(info_el.find('duration').text),
+            'series': series,
+            'season': season,
+            'episode': episode,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,

         }
         }