Merge branch 'subtitles-rework'

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

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 7d8ce18085758469e833c555d9dd675264600ae7..87fce9cd89425150baff91577199f706db2a1e81 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -27,7 +27,6 @@ from ..utils import (
     compiled_regex_type,
     ExtractorError,
     float_or_none,
-    HEADRequest,
     int_or_none,
     RegexNotFoundError,
     sanitize_filename,
@@ -398,6 +397,16 @@ class InfoExtractor(object):
             if blocked_iframe:
                 msg += ' Visit %s for more details' % blocked_iframe
             raise ExtractorError(msg, expected=True)
+        if '<title>The URL you requested has been blocked</title>' in content[:512]:
+            msg = (
+                'Access to this webpage has been blocked by Indian censorship. '
+                'Use a VPN or proxy server (with --proxy) to route around it.')
+            block_msg = self._html_search_regex(
+                r'</h1><p>(.*?)</p>',
+                content, 'block message', default=None)
+            if block_msg:
+                msg += ' (Message: "%s")' % block_msg.replace('\n', ' ')
+            raise ExtractorError(msg, expected=True)
 
         return content
 
@@ -735,6 +744,7 @@ class InfoExtractor(object):
                 f.get('language_preference') if f.get('language_preference') is not None else -1,
                 f.get('quality') if f.get('quality') is not None else -1,
                 f.get('tbr') if f.get('tbr') is not None else -1,
+                f.get('filesize') if f.get('filesize') is not None else -1,
                 f.get('vbr') if f.get('vbr') is not None else -1,
                 f.get('height') if f.get('height') is not None else -1,
                 f.get('width') if f.get('width') is not None else -1,
@@ -742,7 +752,6 @@ class InfoExtractor(object):
                 f.get('abr') if f.get('abr') is not None else -1,
                 audio_ext_preference,
                 f.get('fps') if f.get('fps') is not None else -1,
-                f.get('filesize') if f.get('filesize') is not None else -1,
                 f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
                 f.get('source_preference') if f.get('source_preference') is not None else -1,
                 f.get('format_id'),
@@ -759,9 +768,7 @@ class InfoExtractor(object):
 
     def _is_valid_url(self, url, video_id, item='video'):
         try:
-            self._request_webpage(
-                HEADRequest(url), video_id,
-                'Checking %s URL' % item)
+            self._request_webpage(url, video_id, 'Checking %s URL' % item)
             return True
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError):
@@ -807,8 +814,8 @@ class InfoExtractor(object):
             media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media')
         for i, media_el in enumerate(media_nodes):
             if manifest_version == '2.0':
-                manifest_url = ('/'.join(manifest_url.split('/')[:-1]) + '/'
-                                + (media_el.attrib.get('href') or media_el.attrib.get('url')))
+                manifest_url = ('/'.join(manifest_url.split('/')[:-1]) + '/' +
+                                (media_el.attrib.get('href') or media_el.attrib.get('url')))
             tbr = int_or_none(media_el.attrib.get('bitrate'))
             formats.append({
                 'format_id': '-'.join(filter(None, [f4m_id, 'f4m-%d' % (i if tbr is None else tbr)])),
@@ -832,7 +839,7 @@ class InfoExtractor(object):
             'url': m3u8_url,
             'ext': ext,
             'protocol': 'm3u8',
-            'preference': -1,
+            'preference': preference - 1 if preference else -1,
             'resolution': 'multiple',
             'format_note': 'Quality selection URL',
         }]
@@ -847,6 +854,7 @@ class InfoExtractor(object):
             note='Downloading m3u8 information',
             errnote='Failed to download m3u8 information')
         last_info = None
+        last_media = None
         kv_rex = re.compile(
             r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
         for line in m3u8_doc.splitlines():
@@ -857,6 +865,13 @@ class InfoExtractor(object):
                     if v.startswith('"'):
                         v = v[1:-1]
                     last_info[m.group('key')] = v
+            elif line.startswith('#EXT-X-MEDIA:'):
+                last_media = {}
+                for m in kv_rex.finditer(line):
+                    v = m.group('val')
+                    if v.startswith('"'):
+                        v = v[1:-1]
+                    last_media[m.group('key')] = v
             elif line.startswith('#') or not line.strip():
                 continue
             else:
@@ -885,6 +900,9 @@ class InfoExtractor(object):
                     width_str, height_str = resolution.split('x')
                     f['width'] = int(width_str)
                     f['height'] = int(height_str)
+                if last_media is not None:
+                    f['m3u8_media'] = last_media
+                    last_media = None
                 formats.append(f)
                 last_info = {}
         self._sort_formats(formats)