[youtube] Remove unused variable

[youtube-dl] / youtube_dl / extractor / youtube.py

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 79ce39aa40bdbf59f7afef403f0afc301333138b..dc9e15e98d655e9e53c294035f467569cd307283 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -788,33 +788,41 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             errnote='Could not download DASH manifest')
 
         formats = []
-        for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
-            url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
-            if url_el is None:
-                continue
-            format_id = r.attrib['id']
-            video_url = url_el.text
-            filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
-            f = {
-                'format_id': format_id,
-                'url': video_url,
-                'width': int_or_none(r.attrib.get('width')),
-                'height': int_or_none(r.attrib.get('height')),
-                'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
-                'asr': int_or_none(r.attrib.get('audioSamplingRate')),
-                'filesize': filesize,
-                'fps': int_or_none(r.attrib.get('frameRate')),
-            }
-            try:
-                existing_format = next(
-                    fo for fo in formats
-                    if fo['format_id'] == format_id)
-            except StopIteration:
-                full_info = self._formats.get(format_id, {}).copy()
-                full_info.update(f)
-                formats.append(full_info)
-            else:
-                existing_format.update(f)
+        for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
+            mime_type = a.attrib.get('mimeType')
+            for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
+                url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
+                if url_el is None:
+                    continue
+                if mime_type == 'text/vtt':
+                    # TODO implement WebVTT downloading
+                    pass
+                elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
+                    format_id = r.attrib['id']
+                    video_url = url_el.text
+                    filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
+                    f = {
+                        'format_id': format_id,
+                        'url': video_url,
+                        'width': int_or_none(r.attrib.get('width')),
+                        'height': int_or_none(r.attrib.get('height')),
+                        'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
+                        'asr': int_or_none(r.attrib.get('audioSamplingRate')),
+                        'filesize': filesize,
+                        'fps': int_or_none(r.attrib.get('frameRate')),
+                    }
+                    try:
+                        existing_format = next(
+                            fo for fo in formats
+                            if fo['format_id'] == format_id)
+                    except StopIteration:
+                        full_info = self._formats.get(format_id, {}).copy()
+                        full_info.update(f)
+                        formats.append(full_info)
+                    else:
+                        existing_format.update(f)
+                else:
+                    self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
     def _real_extract(self, url):
@@ -1362,15 +1370,22 @@ class YoutubeChannelIE(InfoExtractor):
 
     def extract_videos_from_page(self, page):
         ids_in_page = []
-        for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
-            if mobj.group(1) not in ids_in_page:
-                ids_in_page.append(mobj.group(1))
-        return ids_in_page
+        titles_in_page = []
+        for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
+            video_id = mobj.group('id')
+            video_title = unescapeHTML(mobj.group('title'))
+            try:
+                idx = ids_in_page.index(video_id)
+                if video_title and not titles_in_page[idx]:
+                    titles_in_page[idx] = video_title
+            except ValueError:
+                ids_in_page.append(video_id)
+                titles_in_page.append(video_title)
+        return zip(ids_in_page, titles_in_page)
 
     def _real_extract(self, url):
         channel_id = self._match_id(url)
 
-        video_ids = []
         url = 'https://www.youtube.com/channel/%s/videos' % channel_id
         channel_page = self._download_webpage(url, channel_id)
         autogenerated = re.search(r'''(?x)
@@ -1382,10 +1397,12 @@ class YoutubeChannelIE(InfoExtractor):
         if autogenerated:
             # The videos are contained in a single page
             # the ajax pages can't be used, they are empty
-            video_ids = self.extract_videos_from_page(channel_page)
+            videos = self.extract_videos_from_page(channel_page)
             entries = [
-                self.url_result(video_id, 'Youtube', video_id=video_id)
-                for video_id in video_ids]
+                self.url_result(
+                    video_id, 'Youtube', video_id=video_id,
+                    video_title=video_title)
+                for video_id, video_title in videos]
             return self.playlist_result(entries, channel_id)
 
         def _entries():
@@ -1393,9 +1410,10 @@ class YoutubeChannelIE(InfoExtractor):
             for pagenum in itertools.count(1):
 
                 ids_in_page = self.extract_videos_from_page(content_html)
-                for video_id in ids_in_page:
+                for video_id, video_title in ids_in_page:
                     yield self.url_result(
-                        video_id, 'Youtube', video_id=video_id)
+                        video_id, 'Youtube', video_id=video_id,
+                        video_title=video_title)
 
                 mobj = re.search(
                     r'data-uix-load-more-href="/?(?P<more>[^"]+)"',