[bbc] Add another description regex

[youtube-dl] / youtube_dl / extractor / facebook.py

diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
index 321eec59ef672eb6f88bf07f78466ba83456afbc..cb5dd57fb4a6cd54400ebd03631dda7e3f07962a 100644 (file)
--- a/youtube_dl/extractor/facebook.py
+++ b/youtube_dl/extractor/facebook.py
@@ -7,11 +7,11 @@ import socket
 from .common import InfoExtractor
 from ..compat import (
     compat_http_client,
-    compat_str,
     compat_urllib_error,
     compat_urllib_parse_unquote,
 )
 from ..utils import (
+    error_to_compat_str,
     ExtractorError,
     limit_length,
     sanitized_Request,
@@ -23,15 +23,23 @@ from ..utils import (
 
 class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
-        https?://(?:\w+\.)?facebook\.com/
-        (?:[^#]*?\#!/)?
-        (?:
-            (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)
-            (?:v|video_id)=|
-            [^/]+/videos/(?:[^/]+/)?
-        )
-        (?P<id>[0-9]+)
-        (?:.*)'''
+                (?:
+                    https?://
+                        (?:\w+\.)?facebook\.com/
+                        (?:[^#]*?\#!/)?
+                        (?:
+                            (?:
+                                video/video\.php|
+                                photo\.php|
+                                video\.php|
+                                video/embed
+                            )\?(?:.*?)(?:v|video_id)=|
+                            [^/]+/videos/(?:[^/]+/)?
+                        )|
+                    facebook:
+                )
+                (?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'
     _NETRC_MACHINE = 'facebook'
@@ -66,6 +74,9 @@ class FacebookIE(InfoExtractor):
     }, {
         'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
         'only_matching': True,
+    }, {
+        'url': 'facebook:544765982287235',
+        'only_matching': True,
     }]
 
     def _login(self):
@@ -74,7 +85,7 @@ class FacebookIE(InfoExtractor):
             return
 
         login_page_req = sanitized_Request(self._LOGIN_URL)
-        login_page_req.add_header('Cookie', 'locale=en_US')
+        self._set_cookie('facebook.com', 'locale', 'en_US')
         login_page = self._download_webpage(login_page_req, None,
                                             note='Downloading login page',
                                             errnote='Unable to download login page')
@@ -100,13 +111,25 @@ class FacebookIE(InfoExtractor):
             login_results = self._download_webpage(request, None,
                                                    note='Logging in', errnote='unable to fetch login page')
             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
-                self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
+                error = self._html_search_regex(
+                    r'(?s)<div[^>]+class=(["\']).*?login_error_box.*?\1[^>]*><div[^>]*>.*?</div><div[^>]*>(?P<error>.+?)</div>',
+                    login_results, 'login error', default=None, group='error')
+                if error:
+                    raise ExtractorError('Unable to login: %s' % error, expected=True)
+                self._downloader.report_warning('unable to log in: bad username/password, or exceeded login rate limit (~3/min). Check credentials or wait.')
+                return
+
+            fb_dtsg = self._search_regex(
+                r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg', default=None)
+            h = self._search_regex(
+                r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h', default=None)
+
+            if not fb_dtsg or not h:
                 return
 
             check_form = {
-                'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
-                'h': self._search_regex(
-                    r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'),
+                'fb_dtsg': fb_dtsg,
+                'h': h,
                 'name_action_selected': 'dont_save',
             }
             check_req = sanitized_Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
@@ -114,9 +137,9 @@ class FacebookIE(InfoExtractor):
             check_response = self._download_webpage(check_req, None,
                                                     note='Confirming login')
             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
-                self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
+                self._downloader.report_warning('Unable to confirm login, you have to login in your browser and authorize the login.')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            self._downloader.report_warning('unable to log in: %s' % compat_str(err))
+            self._downloader.report_warning('unable to log in: %s' % error_to_compat_str(err))
             return
 
     def _real_initialize(self):
@@ -176,3 +199,33 @@ class FacebookIE(InfoExtractor):
             'formats': formats,
             'uploader': uploader,
         }
+
+
+class FacebookPostIE(InfoExtractor):
+    IE_NAME = 'facebook:post'
+    _VALID_URL = r'https?://(?:\w+\.)?facebook\.com/[^/]+/posts/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
+        'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
+        'info_dict': {
+            'id': '544765982287235',
+            'ext': 'mp4',
+            'title': '"What are you doing running in the snow?"',
+            'uploader': 'FailArmy',
+        }
+    }
+
+    def _real_extract(self, url):
+        post_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, post_id)
+
+        entries = [
+            self.url_result('facebook:%s' % video_id, FacebookIE.ie_key())
+            for video_id in self._parse_json(
+                self._search_regex(
+                    r'(["\'])video_ids\1\s*:\s*(?P<ids>\[.+?\])',
+                    webpage, 'video ids', group='ids'),
+                post_id)]
+
+        return self.playlist_result(entries, post_id)