Import Upstream version 2020.01.24

[youtubedl] / youtube_dl / extractor / facebook.py

diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
index a3dcdca3e2bd424bd7a24a800bdf457ec37a033f..ce64e26831fdafceb97b6d8ae919c00a78f0f90f 100644 (file)
--- a/youtube_dl/extractor/facebook.py
+++ b/youtube_dl/extractor/facebook.py
@@ -334,7 +334,7 @@ class FacebookIE(InfoExtractor):
         if not video_data:
             server_js_data = self._parse_json(
                 self._search_regex(
-                    r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+(?:stream_pagelet|pagelet_group_mall|permalink_video_pagelet)',
+                    r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+(?:pagelet_group_mall|permalink_video_pagelet|hyperfeed_story_id_\d+)',
                     webpage, 'js data', default='{}'),
                 video_id, transform_source=js_to_json, fatal=False)
             video_data = extract_from_jsmods_instances(server_js_data)
@@ -379,6 +379,7 @@ class FacebookIE(InfoExtractor):
         if not video_data:
             raise ExtractorError('Cannot parse data')
 
+        subtitles = {}
         formats = []
         for f in video_data:
             format_id = f['stream_type']
@@ -402,9 +403,17 @@ class FacebookIE(InfoExtractor):
             if dash_manifest:
                 formats.extend(self._parse_mpd_formats(
                     compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest))))
+            subtitles_src = f[0].get('subtitles_src')
+            if subtitles_src:
+                subtitles.setdefault('en', []).append({'url': subtitles_src})
         if not formats:
             raise ExtractorError('Cannot find video formats')
 
+        # Downloads with browser's User-Agent are rate limited. Working around
+        # with non-browser User-Agent.
+        for f in formats:
+            f.setdefault('http_headers', {})['User-Agent'] = 'facebookexternalhit/1.1'
+
         self._sort_formats(formats)
 
         video_title = self._html_search_regex(
@@ -442,6 +451,7 @@ class FacebookIE(InfoExtractor):
             'timestamp': timestamp,
             'thumbnail': thumbnail,
             'view_count': view_count,
+            'subtitles': subtitles,
         }
 
         return webpage, info_dict