Start new release.

[youtubedl] / youtube_dl / extractor / rtve.py

diff --git a/youtube_dl/extractor/rtve.py b/youtube_dl/extractor/rtve.py
index c2228b2f0f6a1fc9bba02cddcb5a1740cc85038d..ce9db0629459d23ea0cc870a1bde348d21e99e9a 100644 (file)
--- a/youtube_dl/extractor/rtve.py
+++ b/youtube_dl/extractor/rtve.py
@@ -1,70 +1,120 @@
-# encoding: utf-8
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import re

 import base64
 import base64

+import re
+import time

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

+from ..compat import (
+    compat_b64decode,
+    compat_struct_unpack,
+)

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

-    struct_unpack,
+    determine_ext,
+    ExtractorError,
+    float_or_none,
+    remove_end,
+    remove_start,
+    sanitized_Request,
+    std_headers,

 )
 
 
 )
 
 

+def _decrypt_url(png):
+    encrypted_data = compat_b64decode(png)
+    text_index = encrypted_data.find(b'tEXt')
+    text_chunk = encrypted_data[text_index - 4:]
+    length = compat_struct_unpack('!I', text_chunk[:4])[0]
+    # Use bytearray to get integers when iterating in both python 2.x and 3.x
+    data = bytearray(text_chunk[8:8 + length])
+    data = [chr(b) for b in data if b != 0]
+    hash_index = data.index('#')
+    alphabet_data = data[:hash_index]
+    url_data = data[hash_index + 1:]
+    if url_data[0] == 'H' and url_data[3] == '%':
+        # remove useless HQ%% at the start
+        url_data = url_data[4:]
+
+    alphabet = []
+    e = 0
+    d = 0
+    for l in alphabet_data:
+        if d == 0:
+            alphabet.append(l)
+            d = e = (e + 1) % 4
+        else:
+            d -= 1
+    url = ''
+    f = 0
+    e = 3
+    b = 1
+    for letter in url_data:
+        if f == 0:
+            l = int(letter) * 10
+            f = 1
+        else:
+            if e == 0:
+                l += int(letter)
+                url += alphabet[l]
+                e = (b + 3) % 4
+                f = 0
+                b += 1
+            else:
+                e -= 1
+
+    return url
+
+

 class RTVEALaCartaIE(InfoExtractor):
     IE_NAME = 'rtve.es:alacarta'
     IE_DESC = 'RTVE a la carta'
 class RTVEALaCartaIE(InfoExtractor):
     IE_NAME = 'rtve.es:alacarta'
     IE_DESC = 'RTVE a la carta'

-    _VALID_URL = r'http://www\.rtve\.es/alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/(m/)?(alacarta/videos|filmoteca)/[^/]+/[^/]+/(?P<id>\d+)'

 
 

-    _TEST = {
+    _TESTS = [{

         'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
         'md5': '1d49b7e1ca7a7502c56a4bf1b60f1b43',
         'info_dict': {
             'id': '2491869',
             'ext': 'mp4',
             'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia',
         'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
         'md5': '1d49b7e1ca7a7502c56a4bf1b60f1b43',
         'info_dict': {
             'id': '2491869',
             'ext': 'mp4',
             'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia',

+            'duration': 5024.566,

         },
         },

-    }
-
-    def _decrypt_url(self, png):
-        encrypted_data = base64.b64decode(png)
-        text_index = encrypted_data.find(b'tEXt')
-        text_chunk = encrypted_data[text_index-4:]
-        length = struct_unpack('!I', text_chunk[:4])[0]
-        # Use bytearray to get integers when iterating in both python 2.x and 3.x
-        data = bytearray(text_chunk[8:8+length])
-        data = [chr(b) for b in data if b != 0]
-        hash_index = data.index('#')
-        alphabet_data = data[:hash_index]
-        url_data = data[hash_index+1:]
-
-        alphabet = []
-        e = 0
-        d = 0
-        for l in alphabet_data:
-            if d == 0:
-                alphabet.append(l)
-                d = e = (e + 1) % 4
-            else:
-                d -= 1
-        url = ''
-        f = 0
-        e = 3
-        b = 1
-        for letter in url_data:
-            if f == 0:
-                l = int(letter)*10
-                f = 1
-            else:
-                if e == 0:
-                    l += int(letter)
-                    url += alphabet[l]
-                    e = (b + 3) % 4
-                    f = 0
-                    b += 1
-                else:
-                    e -= 1
+    }, {
+        'note': 'Live stream',
+        'url': 'http://www.rtve.es/alacarta/videos/television/24h-live/1694255/',
+        'info_dict': {
+            'id': '1694255',
+            'ext': 'flv',
+            'title': 'TODO',
+        },
+        'skip': 'The f4m manifest can\'t be used yet',
+    }, {
+        'url': 'http://www.rtve.es/alacarta/videos/servir-y-proteger/servir-proteger-capitulo-104/4236788/',
+        'md5': 'e55e162379ad587e9640eda4f7353c0f',
+        'info_dict': {
+            'id': '4236788',
+            'ext': 'mp4',
+            'title': 'Servir y proteger - Capítulo 104 ',
+            'duration': 3222.0,
+        },
+        'params': {
+            'skip_download': True,  # requires ffmpeg
+        },
+    }, {
+        'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.rtve.es/filmoteca/no-do/not-1-introduccion-primer-noticiario-espanol/1465256/',
+        'only_matching': True,
+    }]

 
 

-        return url
+    def _real_initialize(self):
+        user_agent_b64 = base64.b64encode(std_headers['User-Agent'].encode('utf-8')).decode('utf-8')
+        manager_info = self._download_json(
+            'http://www.rtve.es/odin/loki/' + user_agent_b64,
+            None, 'Fetching manager info')
+        self._manager = manager_info['manager']

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)


@@ -72,13 +122,171 @@ class RTVEALaCartaIE(InfoExtractor):
         info = self._download_json(
             'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
             video_id)['page']['items'][0]
         info = self._download_json(
             'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
             video_id)['page']['items'][0]

-        png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % video_id
+        if info['state'] == 'DESPU':
+            raise ExtractorError('The video is no longer available', expected=True)
+        title = info['title']
+        png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id)
+        png_request = sanitized_Request(png_url)
+        png_request.add_header('Referer', url)
+        png = self._download_webpage(png_request, video_id, 'Downloading url information')
+        video_url = _decrypt_url(png)
+        ext = determine_ext(video_url)
+
+        formats = []
+        if not video_url.endswith('.f4m') and ext != 'm3u8':
+            if '?' not in video_url:
+                video_url = video_url.replace('resources/', 'auth/resources/')
+            video_url = video_url.replace('.net.rtve', '.multimedia.cdn.rtve')
+
+        if ext == 'm3u8':
+            formats.extend(self._extract_m3u8_formats(
+                video_url, video_id, ext='mp4', entry_protocol='m3u8_native',
+                m3u8_id='hls', fatal=False))
+        elif ext == 'f4m':
+            formats.extend(self._extract_f4m_formats(
+                video_url, video_id, f4m_id='hds', fatal=False))
+        else:
+            formats.append({
+                'url': video_url,
+            })
+        self._sort_formats(formats)
+
+        subtitles = None
+        if info.get('sbtFile') is not None:
+            subtitles = self.extract_subtitles(video_id, info['sbtFile'])
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnail': info.get('image'),
+            'page_url': url,
+            'subtitles': subtitles,
+            'duration': float_or_none(info.get('duration'), scale=1000),
+        }
+
+    def _get_subtitles(self, video_id, sub_file):
+        subs = self._download_json(
+            sub_file + '.json', video_id,
+            'Downloading subtitles info')['page']['items']
+        return dict(
+            (s['lang'], [{'ext': 'vtt', 'url': s['src']}])
+            for s in subs)
+
+
+class RTVEInfantilIE(InfoExtractor):
+    IE_NAME = 'rtve.es:infantil'
+    IE_DESC = 'RTVE infantil'
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/infantil/serie/(?P<show>[^/]*)/video/(?P<short_title>[^/]*)/(?P<id>[0-9]+)/'
+
+    _TESTS = [{
+        'url': 'http://www.rtve.es/infantil/serie/cleo/video/maneras-vivir/3040283/',
+        'md5': '915319587b33720b8e0357caaa6617e6',
+        'info_dict': {
+            'id': '3040283',
+            'ext': 'mp4',
+            'title': 'Maneras de vivir',
+            'thumbnail': 'http://www.rtve.es/resources/jpg/6/5/1426182947956.JPG',
+            'duration': 357.958,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        info = self._download_json(
+            'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
+            video_id)['page']['items'][0]
+
+        webpage = self._download_webpage(url, video_id)
+        vidplayer_id = self._search_regex(
+            r' id="vidplayer([0-9]+)"', webpage, 'internal video ID')
+
+        png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % vidplayer_id

         png = self._download_webpage(png_url, video_id, 'Downloading url information')
         png = self._download_webpage(png_url, video_id, 'Downloading url information')

-        video_url = self._decrypt_url(png)
+        video_url = _decrypt_url(png)

 
         return {
             'id': video_id,
 
         return {
             'id': video_id,

+            'ext': 'mp4',

             'title': info['title'],
             'url': video_url,
             'title': info['title'],
             'url': video_url,

-            'thumbnail': info['image'],
+            'thumbnail': info.get('image'),
+            'duration': float_or_none(info.get('duration'), scale=1000),
+        }
+
+
+class RTVELiveIE(InfoExtractor):
+    IE_NAME = 'rtve.es:live'
+    IE_DESC = 'RTVE.es live streams'
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/directo/(?P<id>[a-zA-Z0-9-]+)'
+
+    _TESTS = [{
+        'url': 'http://www.rtve.es/directo/la-1/',
+        'info_dict': {
+            'id': 'la-1',
+            'ext': 'mp4',
+            'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2}Z[0-9]{6}$',
+        },
+        'params': {
+            'skip_download': 'live stream',
+        }
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        start_time = time.gmtime()
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        title = remove_end(self._og_search_title(webpage), ' en directo en RTVE.es')
+        title = remove_start(title, 'Estoy viendo ')
+        title += ' ' + time.strftime('%Y-%m-%dZ%H%M%S', start_time)
+
+        vidplayer_id = self._search_regex(
+            (r'playerId=player([0-9]+)',
+             r'class=["\'].*?\blive_mod\b.*?["\'][^>]+data-assetid=["\'](\d+)',
+             r'data-id=["\'](\d+)'),
+            webpage, 'internal video ID')
+        png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/amonet/videos/%s.png' % vidplayer_id
+        png = self._download_webpage(png_url, video_id, 'Downloading url information')
+        m3u8_url = _decrypt_url(png)
+        formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'is_live': True,

         }
         }

+
+
+class RTVETelevisionIE(InfoExtractor):
+    IE_NAME = 'rtve.es:television'
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/television/[^/]+/[^/]+/(?P<id>\d+).shtml'
+
+    _TEST = {
+        'url': 'http://www.rtve.es/television/20160628/revolucion-del-movil/1364141.shtml',
+        'info_dict': {
+            'id': '3069778',
+            'ext': 'mp4',
+            'title': 'Documentos TV - La revolución del móvil',
+            'duration': 3496.948,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+        webpage = self._download_webpage(url, page_id)
+
+        alacarta_url = self._search_regex(
+            r'data-location="alacarta_videos"[^<]+url&quot;:&quot;(http://www\.rtve\.es/alacarta.+?)&',
+            webpage, 'alacarta url', default=None)
+        if alacarta_url is None:
+            raise ExtractorError(
+                'The webpage doesn\'t contain any video', expected=True)
+
+        return self.url_result(alacarta_url, ie=RTVEALaCartaIE.ie_key())