Merge tag 'upstream/2015.01.16'

[youtubedl] / youtube_dl / extractor / facebook.py

diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
index 8f9154c0e6864d04ef6c4e0f441ef4a68e6d30d2..1ad4e77a8a334dc0bfec62a0fb4752676e2e1435 100644 (file)
--- a/youtube_dl/extractor/facebook.py
+++ b/youtube_dl/extractor/facebook.py
@@ -1,45 +1,56 @@
+from __future__ import unicode_literals
+

 import json
 import re
 import socket
 
 from .common import InfoExtractor
 import json
 import re
 import socket
 
 from .common import InfoExtractor

-from ..utils import (
+from ..compat import (

     compat_http_client,
     compat_str,
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
     compat_http_client,
     compat_str,
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,

-
+)
+from ..utils import (

     ExtractorError,
     ExtractorError,

+    int_or_none,
+    limit_length,
+    urlencode_postdata,

 )
 
 
 class FacebookIE(InfoExtractor):
 )
 
 
 class FacebookIE(InfoExtractor):

-    """Information Extractor for Facebook"""
-

     _VALID_URL = r'''(?x)
     _VALID_URL = r'''(?x)

-        (?:https?://)?(?:\w+\.)?facebook\.com/
-        (?:[^#?]*\#!/)?
-        (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
+        https?://(?:\w+\.)?facebook\.com/
+        (?:[^#]*?\#!/)?
+        (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)

         (?:v|video_id)=(?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'
         (?:v|video_id)=(?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'

-    IE_NAME = u'facebook'
-    _TEST = {
-        u'url': u'https://www.facebook.com/photo.php?v=120708114770723',
-        u'file': u'120708114770723.mp4',
-        u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
-        u'info_dict': {
-            u"duration": 279,
-            u"title": u"PEOPLE ARE AWESOME 2013"
+    IE_NAME = 'facebook'
+    _TESTS = [{
+        'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
+        'md5': '6a40d33c0eccbb1af76cf0485a052659',
+        'info_dict': {
+            'id': '637842556329505',
+            'ext': 'mp4',
+            'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',

         }
         }

-    }
-
-    def report_login(self):
-        """Report attempt to log in."""
-        self.to_screen(u'Logging in')
+    }, {
+        'note': 'Video without discernible title',
+        'url': 'https://www.facebook.com/video.php?v=274175099429670',
+        'info_dict': {
+            'id': '274175099429670',
+            'ext': 'mp4',
+            'title': 'Facebook video #274175099429670',
+        }
+    }, {
+        'url': 'https://www.facebook.com/video.php?v=10204634152394104',
+        'only_matching': True,
+    }]

 
     def _login(self):
         (useremail, password) = self._get_login_info()
 
     def _login(self):
         (useremail, password) = self._get_login_info()


@@ -48,11 +59,13 @@ class FacebookIE(InfoExtractor):
 
         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
         login_page_req.add_header('Cookie', 'locale=en_US')
 
         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
         login_page_req.add_header('Cookie', 'locale=en_US')

-        self.report_login()
-        login_page = self._download_webpage(login_page_req, None, note=False,
-            errnote=u'Unable to download login page')
-        lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
-        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
+        login_page = self._download_webpage(login_page_req, None,
+                                            note='Downloading login page',
+                                            errnote='Unable to download login page')
+        lsd = self._search_regex(
+            r'<input type="hidden" name="lsd" value="([^"]*)"',
+            login_page, 'lsd')
+        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')

 
         login_form = {
             'email': useremail,
 
         login_form = {
             'email': useremail,


@@ -64,39 +77,37 @@ class FacebookIE(InfoExtractor):
             'legacy_return': '1',
             'timezone': '-60',
             'trynum': '1',
             'legacy_return': '1',
             'timezone': '-60',
             'trynum': '1',

-            }
-        request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+        }
+        request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))

         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:

-            login_results = compat_urllib_request.urlopen(request).read()
+            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:
             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:

-                self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
+                self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')

                 return
 
             check_form = {
                 return
 
             check_form = {

-                'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
-                'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
+                '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'),

                 'name_action_selected': 'dont_save',
                 'name_action_selected': 'dont_save',

-                'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),

             }
             }

-            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
+            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))

             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')

-            check_response = compat_urllib_request.urlopen(check_req).read()
+            check_response = self._download_webpage(check_req, None,
+                                                    note='Confirming login')

             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:

-                self._downloader.report_warning(u'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 brower and authorize the login.')

         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:

-            self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
+            self._downloader.report_warning('unable to log in: %s' % compat_str(err))

             return
 
     def _real_initialize(self):
         self._login()
 
     def _real_extract(self, url):
             return
 
     def _real_initialize(self):
         self._login()
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % 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)
 


@@ -107,10 +118,10 @@ class FacebookIE(InfoExtractor):
             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
             if m_msg is not None:
                 raise ExtractorError(
             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
             if m_msg is not None:
                 raise ExtractorError(

-                    u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
+                    'The video is not available, Facebook said: "%s"' % m_msg.group(1),

                     expected=True)
             else:
                     expected=True)
             else:

-                raise ExtractorError(u'Cannot parse data')
+                raise ExtractorError('Cannot parse data')

         data = dict(json.loads(m.group(1)))
         params_raw = compat_urllib_parse.unquote(data['params'])
         params = json.loads(params_raw)
         data = dict(json.loads(m.group(1)))
         params_raw = compat_urllib_parse.unquote(data['params'])
         params = json.loads(params_raw)


@@ -119,19 +130,23 @@ class FacebookIE(InfoExtractor):
         if not video_url:
             video_url = video_data['sd_src']
         if not video_url:
         if not video_url:
             video_url = video_data['sd_src']
         if not video_url:

-            raise ExtractorError(u'Cannot find video URL')
-        video_duration = int(video_data['video_duration'])
-        thumbnail = video_data['thumbnail_src']
+            raise ExtractorError('Cannot find video URL')

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

-            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, u'title')
-
-        info = {
+            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title',
+            fatal=False)
+        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:
+            video_title = 'Facebook video #%s' % video_id
+
+        return {

             'id': video_id,
             'title': video_title,
             'url': video_url,
             'id': video_id,
             'title': video_title,
             'url': video_url,

-            'ext': 'mp4',
-            'duration': video_duration,
-            'thumbnail': thumbnail,
+            'duration': int_or_none(video_data.get('video_duration')),
+            'thumbnail': video_data.get('thumbnail_src'),

         }
         }

-        return [info]