youtube-dl

commit b74fa8cd2c9deb412ac277c6cc44847c3839b844
parent 94eae04c94a43847e8ce7c9bf3d88dd029ef62f6
Author: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date:   Fri,  7 Mar 2014 15:25:33 +0100

[facebook] Fix login process

It was broken and didn't work in python 3.
And use `_download_webpage` instead of `compat_urllib_request.urlopen`.

Diffstat:
M
test/test_utils.py
 | 
5
+++++
M
youtube_dl/extractor/facebook.py
 | 
19
+++++++++++--------
M
youtube_dl/utils.py
 | 
4
++++

3 files changed, 20 insertions(+), 8 deletions(-)
diff --git a/test/test_utils.py b/test/test_utils.py
@@ -33,6 +33,7 @@ from youtube_dl.utils import (
     unified_strdate,
     unsmuggle_url,
     url_basename,
+    urlencode_postdata,
     xpath_with_ns,
 )
 
@@ -261,5 +262,9 @@ class TestUtil(unittest.TestCase):
             bam''')
         self.assertEqual(read_batch_urls(f), [u'foo', u'bar', u'baz', u'bam'])
 
+    def test_urlencode_postdata(self):
+        data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
+        self.assertTrue(isinstance(data, bytes))
+
 if __name__ == '__main__':
     unittest.main()
diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
@@ -11,6 +11,7 @@ from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
+    urlencode_postdata,
 
     ExtractorError,
 )
@@ -51,8 +52,8 @@ class FacebookIE(InfoExtractor):
 
         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,
+        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="([^"]*)"',
@@ -70,23 +71,25 @@ class FacebookIE(InfoExtractor):
             '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:
-            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:
                 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 = {
-                'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, 'fb_dtsg'),
+                'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
                 'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, 'nh'),
                 'name_action_selected': 'dont_save',
-                'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, 'continue'),
+                'submit[Continue]': self._search_regex(r'<button[^>]+value="(.*?)"[^>]+name="submit\[Continue\]"', login_results, '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_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:
                 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:
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
@@ -1263,3 +1263,7 @@ def read_batch_urls(batch_fd):
 
     with contextlib.closing(batch_fd) as fd:
         return [url for url in map(fixup, fd) if url]
+
+
+def urlencode_postdata(*args, **kargs):
+    return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii')