Merge branch 'master' of github.com-rndusr:rg3/youtube-dl into fix/str-item-assignment

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

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 9b73a948ce0f14426e90582c49ef02c243020402..6c3c095f78cec4f44951f0424f20c6828e2462d7 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -36,34 +36,35 @@ from ..utils import (
     clean_html,
     compiled_regex_type,
     determine_ext,
+    determine_protocol,
     error_to_compat_str,
     ExtractorError,
+    extract_attributes,
     fix_xml_ampersands,
     float_or_none,
     GeoRestrictedError,
     GeoUtils,
     int_or_none,
     js_to_json,
+    mimetype2ext,
+    orderedSet,
+    parse_codecs,
+    parse_duration,
     parse_iso8601,
+    parse_m3u8_attributes,
     RegexNotFoundError,
-    sanitize_filename,
     sanitized_Request,
+    sanitize_filename,
     unescapeHTML,
     unified_strdate,
     unified_timestamp,
+    update_Request,
+    update_url_query,
+    urljoin,
     url_basename,
     xpath_element,
     xpath_text,
     xpath_with_ns,
-    determine_protocol,
-    parse_duration,
-    mimetype2ext,
-    update_Request,
-    update_url_query,
-    parse_m3u8_attributes,
-    extract_attributes,
-    parse_codecs,
-    urljoin,
 )
 
 
@@ -714,6 +715,13 @@ class InfoExtractor(object):
             video_info['title'] = video_title
         return video_info
 
+    def playlist_from_matches(self, matches, video_id, video_title, getter=None, ie=None):
+        urlrs = orderedSet(
+            self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
+            for m in matches)
+        return self.playlist_result(
+            urlrs, playlist_id=video_id, playlist_title=video_title)
+
     @staticmethod
     def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
         """Returns a playlist"""
@@ -2161,18 +2169,24 @@ class InfoExtractor(object):
                     })
         return formats
 
-    @staticmethod
-    def _find_jwplayer_data(webpage):
+    def _find_jwplayer_data(self, webpage, video_id=None, transform_source=js_to_json):
         mobj = re.search(
             r'jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)\.setup\s*\((?P<options>[^)]+)\)',
             webpage)
         if mobj:
-            return mobj.group('options')
+            try:
+                jwplayer_data = self._parse_json(mobj.group('options'),
+                                                 video_id=video_id,
+                                                 transform_source=transform_source)
+            except ExtractorError:
+                pass
+            else:
+                if isinstance(jwplayer_data, dict):
+                    return jwplayer_data
 
     def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
-        jwplayer_data = self._parse_json(
-            self._find_jwplayer_data(webpage), video_id,
-            transform_source=js_to_json)
+        jwplayer_data = self._find_jwplayer_data(
+            webpage, video_id, transform_source=js_to_json)
         return self._parse_jwplayer_data(
             jwplayer_data, video_id, *args, **kwargs)
 
@@ -2198,7 +2212,9 @@ class InfoExtractor(object):
 
             this_video_id = video_id or video_data['mediaid']
 
-            formats = self._parse_jwplayer_formats(video_data['sources'], this_video_id)
+            formats = self._parse_jwplayer_formats(
+                video_data['sources'], video_id=this_video_id, m3u8_id=m3u8_id,
+                mpd_id=mpd_id, rtmp_params=rtmp_params, base_url=base_url)
             self._sort_formats(formats)
 
             subtitles = {}
@@ -2232,7 +2248,7 @@ class InfoExtractor(object):
     def _parse_jwplayer_formats(self, jwplayer_sources_data, video_id=None,
                                 m3u8_id=None, mpd_id=None, rtmp_params=None, base_url=None):
         formats = []
-        for source in jwplayer_sources_data :
+        for source in jwplayer_sources_data:
             source_url = self._proto_relative_url(source['file'])
             if base_url:
                 source_url = compat_urlparse.urljoin(base_url, source_url)
@@ -2240,12 +2256,17 @@ class InfoExtractor(object):
             ext = mimetype2ext(source_type) or determine_ext(source_url)
             if source_type == 'hls' or ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    source_url, video_id, 'mp4', 'm3u8_native', m3u8_id=m3u8_id, fatal=False))
+                    source_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id=m3u8_id, fatal=False))
             elif ext == 'mpd':
                 formats.extend(self._extract_mpd_formats(
                     source_url, video_id, mpd_id=mpd_id, fatal=False))
+            elif ext == 'smil':
+                formats.extend(self._extract_smil_formats(
+                    source_url, video_id, fatal=False))
             # https://github.com/jwplayer/jwplayer/blob/master/src/js/providers/default.js#L67
-            elif source_type.startswith('audio') or ext in ('oga', 'aac', 'mp3', 'mpeg', 'vorbis'):
+            elif source_type.startswith('audio') or ext in (
+                    'oga', 'aac', 'mp3', 'mpeg', 'vorbis'):
                 formats.append({
                     'url': source_url,
                     'vcodec': 'none',
@@ -2255,19 +2276,19 @@ class InfoExtractor(object):
                 height = int_or_none(source.get('height'))
                 if height is None:
                     # Often no height is provided but there is a label in
-                    # format like 1080p.
+                    # format like "1080p", "720p SD", or 1080.
                     height = int_or_none(self._search_regex(
-                        r'^(\d{3,})[pP]$', source.get('label') or '',
+                        r'^(\d{3,4})[pP]?(?:\b|$)', compat_str(source.get('label') or ''),
                         'height', default=None))
                 a_format = {
                     'url': source_url,
                     'width': int_or_none(source.get('width')),
                     'height': height,
+                    'tbr': int_or_none(source.get('bitrate')),
                     'ext': ext,
                 }
                 if source_url.startswith('rtmp'):
                     a_format['ext'] = 'flv'
-
                     # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as
                     # of jwplayer.flash.swf
                     rtmp_url_parts = re.split(
@@ -2283,7 +2304,6 @@ class InfoExtractor(object):
                 formats.append(a_format)
         return formats
 
-
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()