d/control: Add build-dependency on dh-python as per warning from dh_python2.

[youtubedl] / youtube_dl / extractor / bliptv.py

diff --git a/youtube_dl/extractor/bliptv.py b/youtube_dl/extractor/bliptv.py
index f2b02643de162ff1547bef47a62dca55f2a405a3..fb56cd78d07ab396ae26ff5a15ac7ebdef933237 100644 (file)
--- a/youtube_dl/extractor/bliptv.py
+++ b/youtube_dl/extractor/bliptv.py
@@ -3,18 +3,21 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from .subtitles import SubtitlesInfoExtractor
-from ..utils import (
+
+from ..compat import (
+    compat_str,

     compat_urllib_request,
     compat_urllib_request,

-    unescapeHTML,
-    parse_iso8601,

     compat_urlparse,
     compat_urlparse,

+)
+from ..utils import (

     clean_html,
     clean_html,

-    compat_str,
+    int_or_none,
+    parse_iso8601,
+    unescapeHTML,

 )
 
 
 )
 
 

-class BlipTVIE(SubtitlesInfoExtractor):
+class BlipTVIE(InfoExtractor):

     _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_]+)))'
 
     _TESTS = [
     _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_]+)))'
 
     _TESTS = [


@@ -64,9 +67,50 @@ class BlipTVIE(SubtitlesInfoExtractor):
                 'uploader': 'redvsblue',
                 'uploader_id': '792887',
             }
                 'uploader': 'redvsblue',
                 'uploader_id': '792887',
             }

-        }
+        },
+        {
+            'url': 'http://blip.tv/play/gbk766dkj4Yn',
+            'md5': 'fe0a33f022d49399a241e84a8ea8b8e3',
+            'info_dict': {
+                'id': '1749452',
+                'ext': 'mp4',
+                'upload_date': '20090208',
+                'description': 'Witness the first appearance of the Nostalgia Critic character, as Doug reviews the movie Transformers.',
+                'title': 'Nostalgia Critic: Transformers',
+                'timestamp': 1234068723,
+                'uploader': 'NostalgiaCritic',
+                'uploader_id': '246467',
+            }
+        },
+        {
+            # https://github.com/rg3/youtube-dl/pull/4404
+            'note': 'Audio only',
+            'url': 'http://blip.tv/hilarios-productions/weekly-manga-recap-kingdom-7119982',
+            'md5': '76c0a56f24e769ceaab21fbb6416a351',
+            'info_dict': {
+                'id': '7103299',
+                'ext': 'flv',
+                'title': 'Weekly Manga Recap: Kingdom',
+                'description': 'And then Shin breaks the enemy line, and he&apos;s all like HWAH! And then he slices a guy and it&apos;s all like FWASHING! And... it&apos;s really hard to describe the best parts of this series without breaking down into sound effects, okay?',
+                'timestamp': 1417660321,
+                'upload_date': '20141204',
+                'uploader': 'The Rollo T',
+                'uploader_id': '407429',
+                'duration': 7251,
+                'vcodec': 'none',
+            }
+        },

     ]
 
     ]
 

+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
+        if mobj:
+            return 'http://blip.tv/a/a-' + mobj.group(1)
+        mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
+        if mobj:
+            return mobj.group(1)
+

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


@@ -74,11 +118,13 @@ class BlipTVIE(SubtitlesInfoExtractor):
         # See https://github.com/rg3/youtube-dl/issues/857 and
         # https://github.com/rg3/youtube-dl/issues/4197
         if lookup_id:
         # See https://github.com/rg3/youtube-dl/issues/857 and
         # https://github.com/rg3/youtube-dl/issues/4197
         if lookup_id:

-            info_page = self._download_webpage(
-                'http://blip.tv/play/%s.x?p=1' % lookup_id, lookup_id, 'Resolving lookup id')
-            video_id = self._search_regex(r'config\.id\s*=\s*"([0-9]+)', info_page, 'video_id')
-        else:
-            video_id = mobj.group('id')
+            urlh = self._request_webpage(
+                'http://blip.tv/play/%s' % lookup_id, lookup_id, 'Resolving lookup id')
+            url = compat_urlparse.urlparse(urlh.geturl())
+            qs = compat_urlparse.parse_qs(url.query)
+            mobj = re.match(self._VALID_URL, qs['file'][0])
+
+        video_id = mobj.group('id')

 
         rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
 
 
         rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
 


@@ -105,7 +151,7 @@ class BlipTVIE(SubtitlesInfoExtractor):
         categories = [category.text for category in item.findall('category')]
 
         formats = []
         categories = [category.text for category in item.findall('category')]
 
         formats = []

-        subtitles = {}
+        subtitles_urls = {}

 
         media_group = item.find(media('group'))
         for media_content in media_group.findall(media('content')):
 
         media_group = item.find(media('group'))
         for media_content in media_group.findall(media('content')):


@@ -114,7 +160,7 @@ class BlipTVIE(SubtitlesInfoExtractor):
             msg = self._download_webpage(
                 url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
                 video_id, 'Resolving URL for %s' % role)
             msg = self._download_webpage(
                 url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
                 video_id, 'Resolving URL for %s' % role)

-            real_url = compat_urlparse.parse_qs(msg)['message'][0]
+            real_url = compat_urlparse.parse_qs(msg.strip())['message'][0]

 
             media_type = media_content.get('type')
             if media_type == 'text/srt' or url.endswith('.srt'):
 
             media_type = media_content.get('type')
             if media_type == 'text/srt' or url.endswith('.srt'):


@@ -123,25 +169,22 @@ class BlipTVIE(SubtitlesInfoExtractor):
                 }
                 lang = role.rpartition('-')[-1].strip().lower()
                 langcode = LANGS.get(lang, lang)
                 }
                 lang = role.rpartition('-')[-1].strip().lower()
                 langcode = LANGS.get(lang, lang)

-                subtitles[langcode] = url
+                subtitles_urls[langcode] = url

             elif media_type.startswith('video/'):
                 formats.append({
                     'url': real_url,
                     'format_id': role,
                     'format_note': media_type,
             elif media_type.startswith('video/'):
                 formats.append({
                     'url': real_url,
                     'format_id': role,
                     'format_note': media_type,

-                    'vcodec': media_content.get(blip('vcodec')),
+                    'vcodec': media_content.get(blip('vcodec')) or 'none',

                     'acodec': media_content.get(blip('acodec')),
                     'filesize': media_content.get('filesize'),
                     'acodec': media_content.get(blip('acodec')),
                     'filesize': media_content.get('filesize'),

-                    'width': int(media_content.get('width')),
-                    'height': int(media_content.get('height')),
+                    'width': int_or_none(media_content.get('width')),
+                    'height': int_or_none(media_content.get('height')),

                 })
                 })

+        self._check_formats(formats, video_id)

         self._sort_formats(formats)
 
         self._sort_formats(formats)
 

-        # subtitles
-        video_subtitles = self.extract_subtitles(video_id, subtitles)
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, subtitles)
-            return
+        subtitles = self.extract_subtitles(video_id, subtitles_urls)

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


@@ -154,15 +197,22 @@ class BlipTVIE(SubtitlesInfoExtractor):
             'thumbnail': thumbnail,
             'categories': categories,
             'formats': formats,
             'thumbnail': thumbnail,
             'categories': categories,
             'formats': formats,

-            'subtitles': video_subtitles,
+            'subtitles': subtitles,

         }
 
         }
 

-    def _download_subtitle_url(self, sub_lang, url):
-        # For some weird reason, blip.tv serves a video instead of subtitles
-        # when we request with a common UA
-        req = compat_urllib_request.Request(url)
-        req.add_header('Youtubedl-user-agent', 'youtube-dl')
-        return self._download_webpage(req, None, note=False)
+    def _get_subtitles(self, video_id, subtitles_urls):
+        subtitles = {}
+        for lang, url in subtitles_urls.items():
+            # For some weird reason, blip.tv serves a video instead of subtitles
+            # when we request with a common UA
+            req = compat_urllib_request.Request(url)
+            req.add_header('User-Agent', 'youtube-dl')
+            subtitles[lang] = [{
+                # The extension is 'srt' but it's actually an 'ass' file
+                'ext': 'ass',
+                'data': self._download_webpage(req, None, note=False),
+            }]
+        return subtitles

 
 
 class BlipTVUserIE(InfoExtractor):
 
 
 class BlipTVUserIE(InfoExtractor):