Merge remote-tracking branch 'Boris-de/wdrmaus_fix#8562'

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

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index a4dd628a1114d5aeb156b025db75ee31544da5f7..6c9f77d95370e740676faef9e3d5cf1231629af6 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -125,6 +125,12 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         if login_results is False:
             return False
 
+        error_msg = self._html_search_regex(
+            r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
+            login_results, 'error message', default=None)
+        if error_msg:
+            raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
+
         if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
             raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
 
@@ -338,6 +344,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
         '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
         '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
+        '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
+        '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
 
         # Dash webm
         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
@@ -1320,9 +1328,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         if video_description:
             video_description = re.sub(r'''(?x)
                 <a\s+
-                    (?:[a-zA-Z-]+="[^"]+"\s+)*?
+                    (?:[a-zA-Z-]+="[^"]*"\s+)*?
                     (?:title|href)="([^"]+)"\s+
-                    (?:[a-zA-Z-]+="[^"]+"\s+)*?
+                    (?:[a-zA-Z-]+="[^"]*"\s+)*?
                     class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
                 [^<]+\.{3}\s*
                 </a>
@@ -2133,10 +2141,11 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
     _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
 
 
-class YoutubeSearchURLIE(InfoExtractor):
+class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
     IE_DESC = 'YouTube.com search URLs'
     IE_NAME = 'youtube:search_url'
     _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
+    _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
     _TESTS = [{
         'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
         'playlist_mincount': 5,
@@ -2151,32 +2160,8 @@ class YoutubeSearchURLIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         query = compat_urllib_parse_unquote_plus(mobj.group('query'))
-
         webpage = self._download_webpage(url, query)
-        result_code = self._search_regex(
-            r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
-
-        part_codes = re.findall(
-            r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
-        entries = []
-        for part_code in part_codes:
-            part_title = self._html_search_regex(
-                [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
-            part_url_snippet = self._html_search_regex(
-                r'(?s)href="([^"]+)"', part_code, 'item URL')
-            part_url = compat_urlparse.urljoin(
-                'https://www.youtube.com/', part_url_snippet)
-            entries.append({
-                '_type': 'url',
-                'url': part_url,
-                'title': part_title,
-            })
-
-        return {
-            '_type': 'playlist',
-            'entries': entries,
-            'title': query,
-        }
+        return self.playlist_result(self._process_page(webpage), playlist_title=query)
 
 
 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):