Add bug number that we are closing.

[youtubedl] / youtube_dl / extractor / vimeo.py

diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py
index 677cf4e1c46c7e08ad1c72ac300d4e83c8229a64..4a7d82b7adfb90e337ff17d9f90d731612650ec6 100644 (file)
--- a/youtube_dl/extractor/vimeo.py
+++ b/youtube_dl/extractor/vimeo.py
@@ -1,5 +1,6 @@
 import json
 import re
 import json
 import re

+import itertools

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (


@@ -16,13 +17,68 @@ class VimeoIE(InfoExtractor):
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs

-    _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)'
+    _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)(?:[?].*)?$'
+    _NETRC_MACHINE = 'vimeo'

     IE_NAME = u'vimeo'
     IE_NAME = u'vimeo'

+    _TESTS = [
+        {
+            u'url': u'http://vimeo.com/56015672',
+            u'file': u'56015672.mp4',
+            u'md5': u'8879b6cc097e987f02484baf890129e5',
+            u'info_dict': {
+                u"upload_date": u"20121220", 
+                u"description": u"This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", 
+                u"uploader_id": u"user7108434", 
+                u"uploader": u"Filippo Valsorda", 
+                u"title": u"youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
+            },
+        },
+        {
+            u'url': u'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
+            u'file': u'68093876.mp4',
+            u'md5': u'3b5ca6aa22b60dfeeadf50b72e44ed82',
+            u'note': u'Vimeo Pro video (#1197)',
+            u'info_dict': {
+                u'uploader_id': u'openstreetmapus', 
+                u'uploader': u'OpenStreetMap US', 
+                u'title': u'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
+            },
+        },
+        {
+            u'url': u'http://player.vimeo.com/video/54469442',
+            u'file': u'54469442.mp4',
+            u'md5': u'619b811a4417aa4abe78dc653becf511',
+            u'note': u'Videos that embed the url in the player page',
+            u'info_dict': {
+                u'title': u'Kathy Sierra: Building the minimum Badass User, Business of Software',
+                u'uploader': u'The BLN & Business of Software',
+            },
+        },
+    ]
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+        self.report_login()
+        login_url = 'https://vimeo.com/log_in'
+        webpage = self._download_webpage(login_url, None, False)
+        token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1)
+        data = compat_urllib_parse.urlencode({'email': username,
+                                              'password': password,
+                                              'action': 'login',
+                                              'service': 'vimeo',
+                                              'token': token,
+                                              })
+        login_request = compat_urllib_request.Request(login_url, data)
+        login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        login_request.add_header('Cookie', 'xsrft=%s' % token)
+        self._download_webpage(login_request, None, False, u'Wrong login info')

 
     def _verify_video_password(self, url, video_id, webpage):
 
     def _verify_video_password(self, url, video_id, webpage):

-        password = self._downloader.params.get('password', None)
+        password = self._downloader.params.get('videopassword', None)

         if password is None:
         if password is None:

-            raise ExtractorError(u'This video is protected by a password, use the --password option')
+            raise ExtractorError(u'This video is protected by a password, use the --video-password option')

         token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1)
         data = compat_urllib_parse.urlencode({'password': password,
                                               'token': token})
         token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1)
         data = compat_urllib_parse.urlencode({'password': password,
                                               'token': token})


@@ -38,6 +94,9 @@ class VimeoIE(InfoExtractor):
                                u'Verifying the password',
                                u'Wrong password')
 
                                u'Verifying the password',
                                u'Wrong password')
 

+    def _real_initialize(self):
+        self._login()
+

     def _real_extract(self, url, new_video=True):
         # Extract ID from URL
         mobj = re.match(self._VALID_URL, url)
     def _real_extract(self, url, new_video=True):
         # Extract ID from URL
         mobj = re.match(self._VALID_URL, url)


@@ -47,7 +106,9 @@ class VimeoIE(InfoExtractor):
         video_id = mobj.group('id')
         if not mobj.group('proto'):
             url = 'https://' + url
         video_id = mobj.group('id')
         if not mobj.group('proto'):
             url = 'https://' + url

-        if mobj.group('direct_link') or mobj.group('pro'):
+        elif mobj.group('pro'):
+            url = 'http://player.vimeo.com/video/' + video_id
+        elif mobj.group('direct_link'):

             url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
             url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information


@@ -61,7 +122,8 @@ class VimeoIE(InfoExtractor):
 
         # Extract the config JSON
         try:
 
         # Extract the config JSON
         try:

-            config = webpage.split(' = {config:')[1].split(',assets:')[0]
+            config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'],
+                webpage, u'info section', flags=re.DOTALL)

             config = json.loads(config)
         except:
             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
             config = json.loads(config)
         except:
             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):


@@ -81,12 +143,22 @@ class VimeoIE(InfoExtractor):
         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 
         # Extract video thumbnail
         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 
         # Extract video thumbnail

-        video_thumbnail = config["video"]["thumbnail"]
+        video_thumbnail = config["video"].get("thumbnail")
+        if video_thumbnail is None:
+            _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in config["video"]["thumbs"].items())[-1]

 
         # Extract video description
 
         # Extract video description

-        video_description = get_element_by_attribute("itemprop", "description", webpage)
-        if video_description: video_description = clean_html(video_description)
-        else: video_description = u''
+        video_description = None
+        try:
+            video_description = get_element_by_attribute("itemprop", "description", webpage)
+            if video_description: video_description = clean_html(video_description)
+        except AssertionError as err:
+            # On some pages like (http://player.vimeo.com/video/54469442) the
+            # html tags are not closed, python 2.6 cannot handle it
+            if err.args[0] == 'we should not get here!':
+                pass
+            else:
+                raise

 
         # Extract upload date
         video_upload_date = None
 
         # Extract upload date
         video_upload_date = None


@@ -103,14 +175,15 @@ class VimeoIE(InfoExtractor):
         # TODO bind to format param
         codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
         files = { 'hd': [], 'sd': [], 'other': []}
         # TODO bind to format param
         codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
         files = { 'hd': [], 'sd': [], 'other': []}

+        config_files = config["video"].get("files") or config["request"].get("files")

         for codec_name, codec_extension in codecs:
         for codec_name, codec_extension in codecs:

-            if codec_name in config["video"]["files"]:
-                if 'hd' in config["video"]["files"][codec_name]:
+            if codec_name in config_files:
+                if 'hd' in config_files[codec_name]:

                     files['hd'].append((codec_name, codec_extension, 'hd'))
                     files['hd'].append((codec_name, codec_extension, 'hd'))

-                elif 'sd' in config["video"]["files"][codec_name]:
+                elif 'sd' in config_files[codec_name]:

                     files['sd'].append((codec_name, codec_extension, 'sd'))
                 else:
                     files['sd'].append((codec_name, codec_extension, 'sd'))
                 else:

-                    files['other'].append((codec_name, codec_extension, config["video"]["files"][codec_name][0]))
+                    files['other'].append((codec_name, codec_extension, config_files[codec_name][0]))

 
         for quality in ('hd', 'sd', 'other'):
             if len(files[quality]) > 0:
 
         for quality in ('hd', 'sd', 'other'):
             if len(files[quality]) > 0:


@@ -122,8 +195,12 @@ class VimeoIE(InfoExtractor):
         else:
             raise ExtractorError(u'No known codec found')
 
         else:
             raise ExtractorError(u'No known codec found')
 

-        video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
-                    %(video_id, sig, timestamp, video_quality, video_codec.upper())
+        video_url = None
+        if isinstance(config_files[video_codec], dict):
+            video_url = config_files[video_codec][video_quality].get("url")
+        if video_url is None:
+            video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
+                        %(video_id, sig, timestamp, video_quality, video_codec.upper())

 
         return [{
             'id':       video_id,
 
         return [{
             'id':       video_id,


@@ -136,3 +213,31 @@ class VimeoIE(InfoExtractor):
             'thumbnail':    video_thumbnail,
             'description':  video_description,
         }]
             'thumbnail':    video_thumbnail,
             'description':  video_description,
         }]

+
+
+class VimeoChannelIE(InfoExtractor):
+    IE_NAME = u'vimeo:channel'
+    _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P<id>[^/]+)'
+    _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        channel_id =  mobj.group('id')
+        video_ids = []
+
+        for pagenum in itertools.count(1):
+            webpage = self._download_webpage('http://vimeo.com/channels/%s/videos/page:%d' % (channel_id, pagenum),
+                                             channel_id, u'Downloading page %s' % pagenum)
+            video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
+            if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
+                break
+
+        entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
+                   for video_id in video_ids]
+        channel_title = self._html_search_regex(r'<a href="/channels/%s">(.*?)</a>' % channel_id,
+                                                webpage, u'channel title')
+        return {'_type': 'playlist',
+                'id': channel_id,
+                'title': channel_title,
+                'entries': entries,
+                }