Merge tag 'upstream/2017.05.18.1'

[youtubedl] / youtube_dl / extractor / facebook.py

diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
index b4d38e5c258b830e192bcfa2639f2074d9217434..b69c1ede0046d73e31df2098f78cf6dc20c254d4 100644 (file)
--- a/youtube_dl/extractor/facebook.py
+++ b/youtube_dl/extractor/facebook.py
@@ -1,3 +1,4 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re


@@ -12,14 +13,16 @@ from ..compat import (
     compat_urllib_parse_unquote_plus,
 )
 from ..utils import (
     compat_urllib_parse_unquote_plus,
 )
 from ..utils import (

+    clean_html,

     error_to_compat_str,
     ExtractorError,
     error_to_compat_str,
     ExtractorError,

+    get_element_by_id,

     int_or_none,
     int_or_none,

+    js_to_json,

     limit_length,
     sanitized_Request,
     limit_length,
     sanitized_Request,

+    try_get,

     urlencode_postdata,
     urlencode_postdata,

-    get_element_by_id,
-    clean_html,

 )
 
 
 )
 
 


@@ -27,7 +30,7 @@ class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
                 (?:
                     https?://
     _VALID_URL = r'''(?x)
                 (?:
                     https?://

-                        (?:[\w-]+\.)?facebook\.com/
+                        (?:[\w-]+\.)?(?:facebook\.com|facebookcorewwwi\.onion)/

                         (?:[^#]*?\#!/)?
                         (?:
                             (?:
                         (?:[^#]*?\#!/)?
                         (?:
                             (?:


@@ -71,7 +74,7 @@ class FacebookIE(InfoExtractor):
         'info_dict': {
             'id': '274175099429670',
             'ext': 'mp4',
         'info_dict': {
             'id': '274175099429670',
             'ext': 'mp4',

-            'title': 'Facebook video #274175099429670',
+            'title': 'Asif Nawab Butt posted a video to his Timeline.',

             'uploader': 'Asif Nawab Butt',
             'upload_date': '20140506',
             'timestamp': 1399398998,
             'uploader': 'Asif Nawab Butt',
             'upload_date': '20140506',
             'timestamp': 1399398998,


@@ -132,6 +135,46 @@ class FacebookIE(InfoExtractor):
             'upload_date': '20161030',
             'uploader': 'CNN',
         },
             'upload_date': '20161030',
             'uploader': 'CNN',
         },

+    }, {
+        # bigPipe.onPageletArrive ... onPageletArrive pagelet_group_mall
+        'url': 'https://www.facebook.com/yaroslav.korpan/videos/1417995061575415/',
+        'info_dict': {
+            'id': '1417995061575415',
+            'ext': 'mp4',
+            'title': 'md5:a7b86ca673f51800cd54687b7f4012fe',
+            'timestamp': 1486648217,
+            'upload_date': '20170209',
+            'uploader': 'Yaroslav Korpan',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.facebook.com/LaGuiaDelVaron/posts/1072691702860471',
+        'info_dict': {
+            'id': '1072691702860471',
+            'ext': 'mp4',
+            'title': 'md5:ae2d22a93fbb12dad20dc393a869739d',
+            'timestamp': 1477305000,
+            'upload_date': '20161024',
+            'uploader': 'La Guía Del Varón',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.facebook.com/groups/1024490957622648/permalink/1396382447100162/',
+        'info_dict': {
+            'id': '1396382447100162',
+            'ext': 'mp4',
+            'title': 'md5:e2d2700afdf84e121f5d0f999bad13a3',
+            'timestamp': 1486035494,
+            'upload_date': '20170202',
+            'uploader': 'Elisabeth Ahtn',
+        },
+        'params': {
+            'skip_download': True,
+        },

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


@@ -150,6 +193,13 @@ class FacebookIE(InfoExtractor):
     }, {
         'url': 'https://zh-hk.facebook.com/peoplespower/videos/1135894589806027/',
         'only_matching': True,
     }, {
         'url': 'https://zh-hk.facebook.com/peoplespower/videos/1135894589806027/',
         'only_matching': True,

+    }, {
+        'url': 'https://www.facebookcorewwwi.onion/video.php?v=274175099429670',
+        'only_matching': True,
+    }, {
+        # no title
+        'url': 'https://www.facebook.com/onlycleverentertainment/videos/1947995502095005/',
+        'only_matching': True,

     }]
 
     @staticmethod
     }]
 
     @staticmethod


@@ -240,12 +290,30 @@ class FacebookIE(InfoExtractor):
 
         video_data = None
 
 
         video_data = None
 

+        def extract_video_data(instances):
+            for item in instances:
+                if item[1][0] == 'VideoConfig':
+                    video_item = item[2][0]
+                    if video_item.get('video_id'):
+                        return video_item['videoData']
+

         server_js_data = self._parse_json(self._search_regex(
         server_js_data = self._parse_json(self._search_regex(

-            r'handleServerJS\(({.+})(?:\);|,")', webpage, 'server js data', default='{}'), video_id)
-        for item in server_js_data.get('instances', []):
-            if item[1][0] == 'VideoConfig':
-                video_data = item[2][0]['videoData']
-                break
+            r'handleServerJS\(({.+})(?:\);|,")', webpage,
+            'server js data', default='{}'), video_id, fatal=False)
+
+        if server_js_data:
+            video_data = extract_video_data(server_js_data.get('instances', []))
+
+        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)',
+                    webpage, 'js data', default='{}'),
+                video_id, transform_source=js_to_json, fatal=False)
+            if server_js_data:
+                video_data = extract_video_data(try_get(
+                    server_js_data, lambda x: x['jsmods']['instances'],
+                    list) or [])

 
         if not video_data:
             if not fatal_if_no_video:
 
         if not video_data:
             if not fatal_if_no_video:


@@ -255,6 +323,8 @@ class FacebookIE(InfoExtractor):
                 raise ExtractorError(
                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
                     expected=True)
                 raise ExtractorError(
                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
                     expected=True)

+            elif '>You must log in to continue' in webpage:
+                self.raise_login_required()

             else:
                 raise ExtractorError('Cannot parse data')
 
             else:
                 raise ExtractorError('Cannot parse data')
 


@@ -287,16 +357,22 @@ class FacebookIE(InfoExtractor):
         self._sort_formats(formats)
 
         video_title = self._html_search_regex(
         self._sort_formats(formats)
 
         video_title = self._html_search_regex(

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

-            video_title = limit_length(video_title, 80)

         if not video_title:
         if not video_title:

+            video_title = self._html_search_meta(
+                'description', webpage, 'title', default=None)
+        if video_title:
+            video_title = limit_length(video_title, 80)
+        else:

             video_title = 'Facebook video #%s' % video_id
             video_title = 'Facebook video #%s' % video_id

-        uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))
+        uploader = clean_html(get_element_by_id(
+            'fbPhotoPageAuthorName', webpage)) or self._search_regex(
+            r'ownerName\s*:\s*"([^"]+)"', webpage, 'uploader', fatal=False)

         timestamp = int_or_none(self._search_regex(
             r'<abbr[^>]+data-utime=["\'](\d+)', webpage,
             'timestamp', default=None))
         timestamp = int_or_none(self._search_regex(
             r'<abbr[^>]+data-utime=["\'](\d+)', webpage,
             'timestamp', default=None))