Merge tag 'upstream/2016.08.17'

[youtubedl] / youtube_dl / extractor / mitele.py

diff --git a/youtube_dl/extractor/mitele.py b/youtube_dl/extractor/mitele.py
index c595f20775efd8e4aed348e4886ac55209e2c6e2..cd169f3616729871fcad9e6c619e67f392f73312 100644 (file)
--- a/youtube_dl/extractor/mitele.py
+++ b/youtube_dl/extractor/mitele.py
@@ -1,33 +1,114 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

+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 (
     compat_urlparse,
 )
 from ..utils import (

-    encode_dict,

     get_element_by_attribute,
     int_or_none,
     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'
     IE_DESC = 'mitele.es'

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

-        'md5': '0ff1a13aebb35d9bc14081ff633dd324',
+        # MD5 is unstable

         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',

-            'ext': 'flv',
+            'ext': 'mp4',

             'title': 'Tor, la web invisible',
             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
             'title': 'Tor, la web invisible',
             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',

+            'series': 'Diario de',
+            'season': 'La redacción',
+            'episode': 'Programa 144',

             'thumbnail': 're:(?i)^https?://.*\.jpg$',
             'duration': 2913,
         },
             'thumbnail': 're:(?i)^https?://.*\.jpg$',
             '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': 'mp4',
+            '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):


@@ -35,55 +116,32 @@ class MiTeleIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
 
         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(encode_dict(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))
+        info = self._get_player_info(url, webpage)

 
         title = self._search_regex(
 
         title = self._search_regex(

-            r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', webpage, 'title')
+            r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>',
+            webpage, 'title', default=None)

 
 

-        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'))
+        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

 
 

-        return {
-            'id': video_id,
+        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,
             'description': get_element_by_attribute('class', 'text', webpage),
             'display_id': display_id,
             'title': title,
             'description': get_element_by_attribute('class', 'text', webpage),

-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-        }
+            'series': series,
+            'season': season,
+            'episode': episode,
+        })
+        return info