[BostonGlobe] New. Nonstandard version of Brightcove.

Has a "data-brightcove-video-id" instead of a "data-video-id," otherwise
pretty much just Brightcove. Except the Globe isn't all Brightcove
videos, so fallback to Generic, too.

Also, abstract playlist_from_matches() from generic.py to common.py, and use
it here.

History of these changes can be found in
51170427d4b1143572a498dedaee61863a5b2c5b.

youtube_dl/extractor/bostonglobe.py

@@ -0,0 +1,72 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+from ..utils import (
+    extract_attributes,
+)
+
+
+class BostonGlobeIE(InfoExtractor):
+    _VALID_URL = r'(?i)https?://(?:www\.)?bostonglobe\.com/.*/(?P<id>[^/]+)/\w+(?:\.html)?'
+    _TESTS = [
+        {
+            'url': 'http://www.bostonglobe.com/metro/2017/02/11/tree-finally-succumbs-disease-leaving-hole-neighborhood/h1b4lviqzMTIn9sVy8F3gP/story.html',
+            'md5': '0a62181079c85c2d2b618c9a738aedaf',
+            'info_dict': {
+                'title': 'A tree finally succumbs to disease, leaving a hole in a neighborhood',
+                'id': '5320421710001',
+                'ext': 'mp4',
+                'description': 'It arrived as a sapling when the Back Bay was in its infancy, a spindly American elm tamped down into a square of dirt cut into the brick sidewalk of 1880s Marlborough Street, no higher than the first bay window of the new brownstone behind it.',
+                'timestamp': 1486877593,
+                'upload_date': '20170212',
+                'uploader_id': '245991542',
+            },
+        },
+        {
+            # Embedded youtube video; we hand it off to the Generic extractor.
+            'url': 'https://www.bostonglobe.com/lifestyle/names/2017/02/17/does-ben-affleck-play-matt-damon-favorite-version-batman/ruqkc9VxKBYmh5txn1XhSI/story.html',
+            'md5': '582b40327089d5c0c949b3c54b13c24b',
+            'info_dict': {
+                'title': "Who Is Matt Damon's Favorite Batman?",
+                'id': 'ZW1QCnlA6Qc',
+                'ext': 'mp4',
+                'upload_date': '20170217',
+                'description': 'md5:3b3dccb9375867e0b4d527ed87d307cb',
+                'uploader': 'The Late Late Show with James Corden',
+                'uploader_id': 'TheLateLateShow',
+            },
+            'expected_warnings': ['404'],
+        },
+    ]
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+        webpage = self._download_webpage(url, page_id)
+
+        page_title = self._og_search_title(webpage, default=None)
+
+        # <video data-brightcove-video-id="5320421710001" data-account="245991542" data-player="SJWAiyYWg" data-embed="default" class="video-js" controls itemscope itemtype="http://schema.org/VideoObject">
+        entries = []
+        for video in re.findall(r'(?i)(<video[^>]+>)', webpage):
+            attrs = extract_attributes(video)
+
+            video_id = attrs.get('data-brightcove-video-id')
+            account_id = attrs.get('data-account')
+            player_id = attrs.get('data-player')
+            embed = attrs.get('data-embed')
+
+            if video_id and account_id and player_id and embed:
+                entries.append(
+                    'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
+                    % (account_id, player_id, embed, video_id))
+
+        if len(entries) == 0:
+            return self.url_result(url, 'Generic')
+        elif len(entries) == 1:
+            return self.url_result(entries[0], 'BrightcoveNew')
+        else:
+            return self.playlist_from_matches(entries, page_id, page_title, ie='BrightcoveNew')

youtube_dl/extractor/common.py

@@ -36,34 +36,35 @@
     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 @@ def url_result(url, ie=None, video_id=None, video_title=None):
             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"""

youtube_dl/extractor/extractors.py

@@ -117,6 +117,7 @@
 from .blinkx import BlinkxIE
 from .bloomberg import BloombergIE
 from .bokecc import BokeCCIE
+from .bostonglobe import BostonGlobeIE
 from .bpb import BpbIE
 from .br import BRIE
 from .bravotv import BravoTVIE

youtube_dl/extractor/generic.py

@@ -1841,14 +1841,6 @@ def _real_extract(self, url):
         video_description = self._og_search_description(webpage, default=None)
         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
 
-        # Helper method
-        def _playlist_from_matches(matches, 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)
-
         # Look for Brightcove Legacy Studio embeds
         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
         if bc_urls:
@@ -1869,28 +1861,28 @@ def _playlist_from_matches(matches, getter=None, ie=None):
         # Look for Brightcove New Studio embeds
         bc_urls = BrightcoveNewIE._extract_urls(webpage)
         if bc_urls:
-            return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
+            return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
 
         # Look for ThePlatform embeds
         tp_urls = ThePlatformIE._extract_urls(webpage)
         if tp_urls:
-            return _playlist_from_matches(tp_urls, ie='ThePlatform')
+            return self.playlist_from_matches(tp_urls, video_id, video_title, ie='ThePlatform')
 
         # Look for Vessel embeds
         vessel_urls = VesselIE._extract_urls(webpage)
         if vessel_urls:
-            return _playlist_from_matches(vessel_urls, ie=VesselIE.ie_key())
+            return self.playlist_from_matches(vessel_urls, video_id, video_title, ie=VesselIE.ie_key())
 
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
             webpage)
         if matches:
-            return _playlist_from_matches(matches, ie='RtlNl')
+            return self.playlist_from_matches(matches, video_id, video_title, ie='RtlNl')
 
         vimeo_urls = VimeoIE._extract_urls(url, webpage)
         if vimeo_urls:
-            return _playlist_from_matches(vimeo_urls, ie=VimeoIE.ie_key())
+            return self.playlist_from_matches(vimeo_urls, video_id, video_title, ie=VimeoIE.ie_key())
 
         vid_me_embed_url = self._search_regex(
             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
@@ -1912,25 +1904,25 @@ def _playlist_from_matches(matches, getter=None, ie=None):
                 (?:embed|v|p)/.+?)
             \1''', webpage)
         if matches:
-            return _playlist_from_matches(
+            return self.playlist_from_matches(
-                matches, lambda m: unescapeHTML(m[1]))
+                matches, video_id, video_title, lambda m: unescapeHTML(m[1]))
 
         # Look for lazyYT YouTube embed
         matches = re.findall(
             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
         if matches:
-            return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
+            return self.playlist_from_matches(matches, video_id, video_title, lambda m: unescapeHTML(m))
 
         # Look for Wordpress "YouTube Video Importer" plugin
         matches = re.findall(r'''(?x)<div[^>]+
             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
         if matches:
-            return _playlist_from_matches(matches, lambda m: m[-1])
+            return self.playlist_from_matches(matches, video_id, video_title, lambda m: m[-1])
 
         matches = DailymotionIE._extract_urls(webpage)
         if matches:
-            return _playlist_from_matches(matches)
+            return self.playlist_from_matches(matches, video_id, video_title)
 
         # Look for embedded Dailymotion playlist player (#3822)
         m = re.search(
@@ -1939,8 +1931,8 @@ def _playlist_from_matches(matches, getter=None, ie=None):
             playlists = re.findall(
                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
             if playlists:
-                return _playlist_from_matches(
+                return self.playlist_from_matches(
-                    playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
+                    playlists, video_id, video_title, lambda p: '//dailymotion.com/playlist/%s' % p)
 
         # Look for embedded Wistia player
         match = re.search(
@@ -2047,8 +2039,9 @@ def _playlist_from_matches(matches, getter=None, ie=None):
         if mobj is not None:
             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
             if embeds:
-                return _playlist_from_matches(
+                return self.playlist_from_matches(
-                    embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
+                    embeds, video_id, video_title,
+                    getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
 
         # Look for Aparat videos
         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
@@ -2110,13 +2103,13 @@ def _playlist_from_matches(matches, getter=None, ie=None):
         # Look for funnyordie embed
         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
         if matches:
-            return _playlist_from_matches(
+            return self.playlist_from_matches(
-                matches, getter=unescapeHTML, ie='FunnyOrDie')
+                matches, video_id, video_title, getter=unescapeHTML, ie='FunnyOrDie')
 
         # Look for BBC iPlayer embed
         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
         if matches:
-            return _playlist_from_matches(matches, ie='BBCCoUk')
+            return self.playlist_from_matches(matches, video_id, video_title, ie='BBCCoUk')
 
         # Look for embedded RUTV player
         rutv_url = RUTVIE._extract_url(webpage)
@@ -2131,32 +2124,32 @@ def _playlist_from_matches(matches, getter=None, ie=None):
         # Look for embedded SportBox player
         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
         if sportbox_urls:
-            return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
+            return self.playlist_from_matches(sportbox_urls, video_id, video_title, ie='SportBoxEmbed')
 
         # Look for embedded XHamster player
         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
         if xhamster_urls:
-            return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
+            return self.playlist_from_matches(xhamster_urls, video_id, video_title, ie='XHamsterEmbed')
 
         # Look for embedded TNAFlixNetwork player
         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
         if tnaflix_urls:
-            return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key())
+            return self.playlist_from_matches(tnaflix_urls, video_id, video_title, ie=TNAFlixNetworkEmbedIE.ie_key())
 
         # Look for embedded PornHub player
         pornhub_urls = PornHubIE._extract_urls(webpage)
         if pornhub_urls:
-            return _playlist_from_matches(pornhub_urls, ie=PornHubIE.ie_key())
+            return self.playlist_from_matches(pornhub_urls, video_id, video_title, ie=PornHubIE.ie_key())
 
         # Look for embedded DrTuber player
         drtuber_urls = DrTuberIE._extract_urls(webpage)
         if drtuber_urls:
-            return _playlist_from_matches(drtuber_urls, ie=DrTuberIE.ie_key())
+            return self.playlist_from_matches(drtuber_urls, video_id, video_title, ie=DrTuberIE.ie_key())
 
         # Look for embedded RedTube player
         redtube_urls = RedTubeIE._extract_urls(webpage)
         if redtube_urls:
-            return _playlist_from_matches(redtube_urls, ie=RedTubeIE.ie_key())
+            return self.playlist_from_matches(redtube_urls, video_id, video_title, ie=RedTubeIE.ie_key())
 
         # Look for embedded Tvigle player
         mobj = re.search(
@@ -2202,12 +2195,12 @@ def _playlist_from_matches(matches, getter=None, ie=None):
         # Look for embedded soundcloud player
         soundcloud_urls = SoundcloudIE._extract_urls(webpage)
         if soundcloud_urls:
-            return _playlist_from_matches(soundcloud_urls, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
+            return self.playlist_from_matches(soundcloud_urls, video_id, video_title, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
 
         # Look for tunein player
         tunein_urls = TuneInBaseIE._extract_urls(webpage)
         if tunein_urls:
-            return _playlist_from_matches(tunein_urls)
+            return self.playlist_from_matches(tunein_urls, video_id, video_title)
 
         # Look for embedded mtvservices player
         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
@@ -2490,35 +2483,35 @@ def _playlist_from_matches(matches, getter=None, ie=None):
         # Look for DBTV embeds
         dbtv_urls = DBTVIE._extract_urls(webpage)
         if dbtv_urls:
-            return _playlist_from_matches(dbtv_urls, ie=DBTVIE.ie_key())
+            return self.playlist_from_matches(dbtv_urls, video_id, video_title, ie=DBTVIE.ie_key())
 
         # Look for Videa embeds
         videa_urls = VideaIE._extract_urls(webpage)
         if videa_urls:
-            return _playlist_from_matches(videa_urls, ie=VideaIE.ie_key())
+            return self.playlist_from_matches(videa_urls, video_id, video_title, ie=VideaIE.ie_key())
 
         # Look for 20 minuten embeds
         twentymin_urls = TwentyMinutenIE._extract_urls(webpage)
         if twentymin_urls:
-            return _playlist_from_matches(
+            return self.playlist_from_matches(
-                twentymin_urls, ie=TwentyMinutenIE.ie_key())
+                twentymin_urls, video_id, video_title, ie=TwentyMinutenIE.ie_key())
 
         # Look for Openload embeds
         openload_urls = OpenloadIE._extract_urls(webpage)
         if openload_urls:
-            return _playlist_from_matches(
+            return self.playlist_from_matches(
-                openload_urls, ie=OpenloadIE.ie_key())
+                openload_urls, video_id, video_title, ie=OpenloadIE.ie_key())
 
         # Look for VideoPress embeds
         videopress_urls = VideoPressIE._extract_urls(webpage)
         if videopress_urls:
-            return _playlist_from_matches(
+            return self.playlist_from_matches(
-                videopress_urls, ie=VideoPressIE.ie_key())
+                videopress_urls, video_id, video_title, ie=VideoPressIE.ie_key())
 
         # Look for Rutube embeds
         rutube_urls = RutubeIE._extract_urls(webpage)
         if rutube_urls:
-            return _playlist_from_matches(
+            return self.playlist_from_matches(
                 rutube_urls, ie=RutubeIE.ie_key())
 
         # Looking for http://schema.org/VideoObject