Prepare to upload.

[youtubedl] / youtube_dl / extractor / telecinco.py

diff --git a/youtube_dl/extractor/telecinco.py b/youtube_dl/extractor/telecinco.py
index 2c8e9b9410f09f23349584e7945bed64e5dd3a76..9ba3da341dac65d18a599a790bff9c95b0e52eb8 100644 (file)
--- a/youtube_dl/extractor/telecinco.py
+++ b/youtube_dl/extractor/telecinco.py
@@ -2,93 +2,187 @@
 from __future__ import unicode_literals
 
 import json
 from __future__ import unicode_literals
 
 import json

+import re

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..compat import (
-    compat_urllib_parse,
-    compat_urllib_parse_unquote,
-    compat_urlparse,
-)
+from .ooyala import OoyalaIE

 from ..utils import (
 from ..utils import (

-    get_element_by_attribute,
-    parse_duration,
-    strip_jsonp,
+    clean_html,
+    determine_ext,
+    int_or_none,
+    str_or_none,
+    try_get,
+    urljoin,

 )
 
 
 class TelecincoIE(InfoExtractor):
     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
 )
 
 
 class TelecincoIE(InfoExtractor):
     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'

-    _VALID_URL = r'https?://www\.(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
+    _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',
 
     _TESTS = [{
         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',

-        'md5': '5cbef3ad5ef17bf0d21570332d140729',

         'info_dict': {
         'info_dict': {

-            'id': 'MDSVID20141015_0058',
-            'ext': 'mp4',
-            'title': 'Con Martín Berasategui, hacer un bacalao al ...',
-            'duration': 662,
+            'id': '1876350223',
+            'title': 'Bacalao con kokotxas al pil-pil',
+            'description': 'md5:716caf5601e25c3c5ab6605b1ae71529',

         },
         },

+        'playlist': [{
+            'md5': 'adb28c37238b675dad0f042292f209a7',
+            'info_dict': {
+                'id': 'JEA5ijCnF6p5W08A1rNKn7',
+                'ext': 'mp4',
+                'title': 'Con Martín Berasategui, hacer un bacalao al pil-pil es fácil y divertido',
+                'duration': 662,
+            },
+        }]

     }, {
         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
     }, {
         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',

-        'md5': '0a5b9f3cc8b074f50a0578f823a12694',
+        'md5': '9468140ebc300fbb8b9d65dc6e5c4b43',

         'info_dict': {
         'info_dict': {

-            'id': 'MDSVID20150916_0128',
+            'id': 'jn24Od1zGLG4XUZcnUnZB6',

             'ext': 'mp4',
             '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',
             'duration': 79,
         },
     }, {
         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',

-        'md5': 'ad1bfaaba922dd4a295724b05b68f86a',
+        'md5': 'ae2dc6b7b50b2392076a51c0f70e01f6',

         'info_dict': {
         'info_dict': {

-            'id': 'MDSVID20150513_0220',
+            'id': 'aywerkD2Sv1vGNqq9b85Q2',

             'ext': 'mp4',
             'title': '#DOYLACARA. Con la trata no hay trato',
             'ext': 'mp4',
             'title': '#DOYLACARA. Con la trata no hay trato',

+            'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',

             'duration': 50,
         },
             'duration': 50,
         },

+    }, {
+        # video in opening's content
+        'url': 'https://www.telecinco.es/vivalavida/fiorella-sobrina-edmundo-arrocet-entrevista_18_2907195140.html',
+        'info_dict': {
+            'id': '2907195140',
+            'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
+            'description': 'md5:73f340a7320143d37ab895375b2bf13a',
+        },
+        'playlist': [{
+            'md5': 'adb28c37238b675dad0f042292f209a7',
+            'info_dict': {
+                'id': 'TpI2EttSDAReWpJ1o0NVh2',
+                'ext': 'mp4',
+                'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
+                'duration': 1015,
+            },
+        }],
+        'params': {
+            'skip_download': True,
+        },

     }, {
         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
         'only_matching': True,
     }, {
         'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
         'only_matching': True,
     }, {
         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
         'only_matching': True,
     }, {
         'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
         'only_matching': True,

+    }, {
+        # ooyala video
+        'url': 'http://www.cuatro.com/chesterinlove/a-carta/chester-chester_in_love-chester_edu_2_2331030022.html',
+        'only_matching': True,

     }]
 
     }]
 

-    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)
+    def _parse_content(self, content, url):
+        video_id = content['dataMediaId']
+        if content.get('dataCmsId') == 'ooyala':
+            return self.url_result(
+                'ooyala:%s' % video_id, OoyalaIE.ie_key(), video_id)
+        config_url = urljoin(url, content['dataConfig'])
+        config = self._download_json(
+            config_url, video_id, 'Downloading config JSON')
+        title = config['info']['title']

 
 

-        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')
+        def mmc_url(mmc_type):
+            return re.sub(
+                r'/(?:flash|html5)\.json', '/%s.json' % mmc_type,
+                config['services']['mmc'])

 
 

-        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')
+        duration = None
+        formats = []
+        for mmc_type in ('flash', 'html5'):
+            mmc = self._download_json(
+                mmc_url(mmc_type), video_id,
+                'Downloading %s mmc JSON' % mmc_type, fatal=False)
+            if not mmc:
+                continue
+            if not duration:
+                duration = int_or_none(mmc.get('duration'))
+            for location in mmc['locations']:
+                gat = self._proto_relative_url(location.get('gat'), 'http:')
+                gcp = location.get('gcp')
+                ogn = location.get('ogn')
+                if None in (gat, gcp, ogn):
+                    continue
+                token_data = {
+                    'gcp': gcp,
+                    'ogn': ogn,
+                    'sta': 0,
+                }
+                media = self._download_json(
+                    gat, video_id, data=json.dumps(token_data).encode('utf-8'),
+                    headers={
+                        'Content-Type': 'application/json;charset=utf-8',
+                        'Referer': url,
+                    }, fatal=False) or {}
+                stream = media.get('stream') or media.get('file')
+                if not stream:
+                    continue
+                ext = determine_ext(stream)
+                if ext == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        stream + '&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(
+                        stream, video_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+        self._sort_formats(formats)

 
         return {
 
         return {

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

             'formats': formats,
             'formats': formats,

-            'description': get_element_by_attribute('class', 'text', webpage),
-            'thumbnail': info_el.find('thumb').text,
-            'duration': parse_duration(info_el.find('duration').text),
+            'thumbnail': content.get('dataPoster') or config.get('poster', {}).get('imageUrl'),
+            'duration': duration,

         }
         }

+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        article = self._parse_json(self._search_regex(
+            r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=\s*({.+})',
+            webpage, 'article'), display_id)['article']
+        title = article.get('title')
+        description = clean_html(article.get('leadParagraph')) or ''
+        if article.get('editorialType') != 'VID':
+            entries = []
+            body = [article.get('opening')]
+            body.extend(try_get(article, lambda x: x['body'], list) or [])
+            for p in body:
+                if not isinstance(p, dict):
+                    continue
+                content = p.get('content')
+                if not content:
+                    continue
+                type_ = p.get('type')
+                if type_ == 'paragraph':
+                    content_str = str_or_none(content)
+                    if content_str:
+                        description += content_str
+                    continue
+                if type_ == 'video' and isinstance(content, dict):
+                    entries.append(self._parse_content(content, url))
+            return self.playlist_result(
+                entries, str_or_none(article.get('id')), title, description)
+        content = article['opening']['content']
+        info = self._parse_content(content, url)
+        info.update({
+            'description': description,
+        })
+        return info