Register new version in the changelog.

[youtubedl] / youtube_dl / extractor / facebook.py

diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
index 2139f68aa3cb16facdc45b5fd9e014621e1c6674..f53c5161501df162a1480d5cbb8a82b18682e4cc 100644 (file)
--- a/youtube_dl/extractor/facebook.py
+++ b/youtube_dl/extractor/facebook.py
@@ -9,13 +9,15 @@ from ..compat import (
     compat_http_client,
     compat_str,
     compat_urllib_error,
     compat_http_client,
     compat_str,
     compat_urllib_error,

-    compat_urllib_parse,
+    compat_urllib_parse_unquote,

     compat_urllib_request,
 )
 from ..utils import (
     compat_urllib_request,
 )
 from ..utils import (

-    urlencode_postdata,

     ExtractorError,
     limit_length,
     ExtractorError,
     limit_length,

+    urlencode_postdata,
+    get_element_by_id,
+    clean_html,

 )
 
 
 )
 
 


@@ -23,8 +25,12 @@ class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
         https?://(?:\w+\.)?facebook\.com/
         (?:[^#]*?\#!/)?
     _VALID_URL = r'''(?x)
         https?://(?:\w+\.)?facebook\.com/
         (?:[^#]*?\#!/)?

-        (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)
-        (?:v|video_id)=(?P<id>[0-9]+)
+        (?:
+            (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)
+            (?:v|video_id)=|
+            [^/]+/videos/(?:[^/]+/)?
+        )
+        (?P<id>[0-9]+)

         (?:.*)'''
     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
         (?:.*)'''
     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'


@@ -36,8 +42,8 @@ class FacebookIE(InfoExtractor):
         'info_dict': {
             'id': '637842556329505',
             'ext': 'mp4',
         'info_dict': {
             'id': '637842556329505',
             'ext': 'mp4',

-            'duration': 38,

             'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
             'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',

+            'uploader': 'Tennis on Facebook',

         }
     }, {
         'note': 'Video without discernible title',
         }
     }, {
         'note': 'Video without discernible title',


@@ -46,10 +52,20 @@ class FacebookIE(InfoExtractor):
             'id': '274175099429670',
             'ext': 'mp4',
             'title': 'Facebook video #274175099429670',
             'id': '274175099429670',
             'ext': 'mp4',
             'title': 'Facebook video #274175099429670',

-        }
+            'uploader': 'Asif Nawab Butt',
+        },
+        'expected_warnings': [
+            'title'
+        ]

     }, {
         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
         'only_matching': True,
     }, {
         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
         'only_matching': True,

+    }, {
+        'url': 'https://www.facebook.com/amogood/videos/1618742068337349/?fref=nf',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
+        'only_matching': True,

     }]
 
     def _login(self):
     }]
 
     def _login(self):


@@ -107,9 +123,7 @@ class FacebookIE(InfoExtractor):
         self._login()
 
     def _real_extract(self, url):
         self._login()
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)

         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
         webpage = self._download_webpage(url, video_id)
 
         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
         webpage = self._download_webpage(url, video_id)
 


@@ -125,30 +139,40 @@ class FacebookIE(InfoExtractor):
             else:
                 raise ExtractorError('Cannot parse data')
         data = dict(json.loads(m.group(1)))
             else:
                 raise ExtractorError('Cannot parse data')
         data = dict(json.loads(m.group(1)))

-        params_raw = compat_urllib_parse.unquote(data['params'])
+        params_raw = compat_urllib_parse_unquote(data['params'])

         params = json.loads(params_raw)
         params = json.loads(params_raw)

-        video_data = params['video_data'][0]
-        video_url = video_data.get('hd_src')
-        if not video_url:
-            video_url = video_data['sd_src']
-        if not video_url:
-            raise ExtractorError('Cannot find video URL')
+
+        formats = []
+        for format_id, f in params['video_data'].items():
+            if not f or not isinstance(f, list):
+                continue
+            for quality in ('sd', 'hd'):
+                for src_type in ('src', 'src_no_ratelimit'):
+                    src = f[0].get('%s_%s' % (quality, src_type))
+                    if src:
+                        formats.append({
+                            'format_id': '%s_%s_%s' % (format_id, quality, src_type),
+                            'url': src,
+                            'preference': -10 if format_id == 'progressive' else 0,
+                        })
+        if not formats:
+            raise ExtractorError('Cannot find video formats')

 
         video_title = self._html_search_regex(
 
         video_title = self._html_search_regex(

-            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title',
-            fatal=False)
+            r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
+            default=None)

         if not video_title:
             video_title = self._html_search_regex(
                 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
         if not video_title:
             video_title = self._html_search_regex(
                 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',

-                webpage, 'alternative title', default=None)
+                webpage, 'alternative title', fatal=False)

             video_title = limit_length(video_title, 80)
         if not video_title:
             video_title = 'Facebook video #%s' % video_id
             video_title = limit_length(video_title, 80)
         if not video_title:
             video_title = 'Facebook video #%s' % video_id

+        uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))

 
         return {
             'id': video_id,
             'title': video_title,
 
         return {
             'id': video_id,
             'title': video_title,

-            'url': video_url,
-            'duration': int(video_data['video_duration']),
-            'thumbnail': video_data['thumbnail_src'],
+            'formats': formats,
+            'uploader': uploader,

         }
         }