Merge tag 'upstream/2013.12.23'

[youtubedl] / youtube_dl / extractor / vimeo.py

diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py
index 4a7d82b7adfb90e337ff17d9f90d731612650ec6..c3623fcbe6b01493c5ec2115f4fe5f2d32737e59 100644 (file)
--- a/youtube_dl/extractor/vimeo.py
+++ b/youtube_dl/extractor/vimeo.py
@@ -1,3 +1,4 @@
+# encoding: utf-8

 import json
 import re
 import itertools
 import json
 import re
 import itertools


@@ -10,19 +11,30 @@ from ..utils import (
     clean_html,
     get_element_by_attribute,
     ExtractorError,
     clean_html,
     get_element_by_attribute,
     ExtractorError,

+    RegexNotFoundError,

     std_headers,
     std_headers,

+    unsmuggle_url,

 )
 
 )
 

+

 class VimeoIE(InfoExtractor):
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs
 class VimeoIE(InfoExtractor):
     """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]+)(?:[?].*)?$'
+    _VALID_URL = r'''(?x)
+        (?P<proto>https?://)?
+        (?:(?:www|(?P<player>player))\.)?
+        vimeo(?P<pro>pro)?\.com/
+        (?:.*?/)?
+        (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
+        (?:videos?/)?
+        (?P<id>[0-9]+)
+        /?(?:[?&].*)?(?:[#].*)?$'''

     _NETRC_MACHINE = 'vimeo'
     IE_NAME = u'vimeo'
     _TESTS = [
         {
     _NETRC_MACHINE = 'vimeo'
     IE_NAME = u'vimeo'
     _TESTS = [
         {

-            u'url': u'http://vimeo.com/56015672',
+            u'url': u'http://vimeo.com/56015672#at=0',

             u'file': u'56015672.mp4',
             u'md5': u'8879b6cc097e987f02484baf890129e5',
             u'info_dict': {
             u'file': u'56015672.mp4',
             u'md5': u'8879b6cc097e987f02484baf890129e5',
             u'info_dict': {


@@ -54,6 +66,21 @@ class VimeoIE(InfoExtractor):
                 u'uploader': u'The BLN & Business of Software',
             },
         },
                 u'uploader': u'The BLN & Business of Software',
             },
         },

+        {
+            u'url': u'http://vimeo.com/68375962',
+            u'file': u'68375962.mp4',
+            u'md5': u'aaf896bdb7ddd6476df50007a0ac0ae7',
+            u'note': u'Video protected with password',
+            u'info_dict': {
+                u'title': u'youtube-dl password protected test video',
+                u'upload_date': u'20130614',
+                u'uploader_id': u'user18948128',
+                u'uploader': u'Jaime Marquínez Ferrándiz',
+            },
+            u'params': {
+                u'videopassword': u'youtube-dl',
+            },
+        },

     ]
 
     def _login(self):
     ]
 
     def _login(self):


@@ -97,22 +124,26 @@ class VimeoIE(InfoExtractor):
     def _real_initialize(self):
         self._login()
 
     def _real_initialize(self):
         self._login()
 

-    def _real_extract(self, url, new_video=True):
+    def _real_extract(self, url):
+        url, data = unsmuggle_url(url)
+        headers = std_headers
+        if data is not None:
+            headers = headers.copy()
+            headers.update(data)
+

         # Extract ID from URL
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
 
         video_id = mobj.group('id')
         # Extract ID from URL
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
 
         video_id = mobj.group('id')

-        if not mobj.group('proto'):
-            url = 'https://' + url
-        elif mobj.group('pro'):
+        if mobj.group('pro') or mobj.group('player'):

             url = 'http://player.vimeo.com/video/' + video_id
             url = 'http://player.vimeo.com/video/' + video_id

-        elif mobj.group('direct_link'):
+        else:

             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

-        request = compat_urllib_request.Request(url, None, std_headers)
+        request = compat_urllib_request.Request(url, None, headers)

         webpage = self._download_webpage(request, video_id)
 
         # Now we begin extracting as much information as we can from what we
         webpage = self._download_webpage(request, video_id)
 
         # Now we begin extracting as much information as we can from what we


@@ -122,18 +153,32 @@ class VimeoIE(InfoExtractor):
 
         # Extract the config JSON
         try:
 
         # Extract the config JSON
         try:

-            config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'],
-                webpage, u'info section', flags=re.DOTALL)
-            config = json.loads(config)
-        except:
+            try:
+                config_url = self._html_search_regex(
+                    r' data-config-url="(.+?)"', webpage, u'config URL')
+                config_json = self._download_webpage(config_url, video_id)
+                config = json.loads(config_json)
+            except RegexNotFoundError:
+                # For pro videos or player.vimeo.com urls
+                # We try to find out to which variable is assigned the config dic
+                m_variable_name = re.search('(\w)\.video\.id', webpage)
+                if m_variable_name is not None:
+                    config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
+                else:
+                    config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
+                config = self._search_regex(config_re, webpage, u'info section',
+                    flags=re.DOTALL)
+                config = json.loads(config)
+        except Exception as e:

             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
                 raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option')
 
             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
                 raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option')
 

-            if re.search('If so please provide the correct password.', webpage):
+            if re.search('<form[^>]+?id="pw_form"', webpage) is not None:

                 self._verify_video_password(url, video_id, webpage)
                 return self._real_extract(url)
             else:
                 self._verify_video_password(url, video_id, webpage)
                 return self._real_extract(url)
             else:

-                raise ExtractorError(u'Unable to extract info section')
+                raise ExtractorError(u'Unable to extract info section',
+                                     cause=e)

 
         # Extract title
         video_title = config["video"]["title"]
 
         # Extract title
         video_title = config["video"]["title"]


@@ -166,78 +211,143 @@ class VimeoIE(InfoExtractor):
         if mobj is not None:
             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 
         if mobj is not None:
             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 

+        try:
+            view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, u'view count'))
+            like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, u'like count'))
+            comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, u'comment count'))
+        except RegexNotFoundError:
+            # This info is only available in vimeo.com/{id} urls
+            view_count = None
+            like_count = None
+            comment_count = None
+

         # Vimeo specific: extract request signature and timestamp
         sig = config['request']['signature']
         timestamp = config['request']['timestamp']
 
         # Vimeo specific: extract video codec and quality information
         # First consider quality, then codecs, then take everything
         # Vimeo specific: extract request signature and timestamp
         sig = config['request']['signature']
         timestamp = config['request']['timestamp']
 
         # Vimeo specific: extract video codec and quality information
         # First consider quality, then codecs, then take everything

-        # TODO bind to format param
-        codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
-        files = { 'hd': [], 'sd': [], 'other': []}
+        codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
+        files = {'hd': [], 'sd': [], 'other': []}

         config_files = config["video"].get("files") or config["request"].get("files")
         for codec_name, codec_extension in codecs:
         config_files = config["video"].get("files") or config["request"].get("files")
         for codec_name, codec_extension in codecs:

-            if codec_name in config_files:
-                if 'hd' in config_files[codec_name]:
-                    files['hd'].append((codec_name, codec_extension, 'hd'))
-                elif 'sd' in config_files[codec_name]:
-                    files['sd'].append((codec_name, codec_extension, 'sd'))
+            for quality in config_files.get(codec_name, []):
+                format_id = '-'.join((codec_name, quality)).lower()
+                key = quality if quality in files else 'other'
+                video_url = None
+                if isinstance(config_files[codec_name], dict):
+                    file_info = config_files[codec_name][quality]
+                    video_url = file_info.get('url')

                 else:
                 else:

-                    files['other'].append((codec_name, codec_extension, config_files[codec_name][0]))
-
-        for quality in ('hd', 'sd', 'other'):
-            if len(files[quality]) > 0:
-                video_quality = files[quality][0][2]
-                video_codec = files[quality][0][0]
-                video_extension = files[quality][0][1]
-                self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality))
-                break
-        else:
-            raise ExtractorError(u'No known codec found')
+                    file_info = {}
+                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, quality, codec_name.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())
+                files[key].append({
+                    'ext': codec_extension,
+                    'url': video_url,
+                    'format_id': format_id,
+                    'width': file_info.get('width'),
+                    'height': file_info.get('height'),
+                })
+        formats = []
+        for key in ('other', 'sd', 'hd'):
+            formats += files[key]
+        if len(formats) == 0:
+            raise ExtractorError(u'No known codec found')

 
 

-        return [{
+        return {

             'id':       video_id,
             'id':       video_id,

-            'url':      video_url,

             'uploader': video_uploader,
             'uploader_id': video_uploader_id,
             'upload_date':  video_upload_date,
             'title':    video_title,
             'uploader': video_uploader,
             'uploader_id': video_uploader_id,
             'upload_date':  video_upload_date,
             'title':    video_title,

-            'ext':      video_extension,

             'thumbnail':    video_thumbnail,
             'description':  video_description,
             'thumbnail':    video_thumbnail,
             'description':  video_description,

-        }]
+            'formats': formats,
+            'webpage_url': url,
+            'view_count': view_count,
+            'like_count': like_count,
+            'comment_count': comment_count,
+        }

 
 
 class VimeoChannelIE(InfoExtractor):
     IE_NAME = u'vimeo:channel'
     _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P<id>[^/]+)'
     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 
 
 class VimeoChannelIE(InfoExtractor):
     IE_NAME = u'vimeo:channel'
     _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P<id>[^/]+)'
     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'

+    _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'

 
 

-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        channel_id =  mobj.group('id')
-        video_ids = []
+    def _page_url(self, base_url, pagenum):
+        return '%s/videos/page:%d/' % (base_url, pagenum)

 
 

+    def _extract_list_title(self, webpage):
+        return self._html_search_regex(self._TITLE_RE, webpage, u'list title')
+
+    def _extract_videos(self, list_id, base_url):
+        video_ids = []

         for pagenum in itertools.count(1):
         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)
+            webpage = self._download_webpage(
+                self._page_url(base_url, pagenum) ,list_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]
             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',
         return {'_type': 'playlist',

-                'id': channel_id,
-                'title': channel_title,
+                'id': list_id,
+                'title': self._extract_list_title(webpage),

                 'entries': entries,
                 }
                 'entries': entries,
                 }

+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        channel_id =  mobj.group('id')
+        return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
+
+
+class VimeoUserIE(VimeoChannelIE):
+    IE_NAME = u'vimeo:user'
+    _VALID_URL = r'(?:https?://)?vimeo.\com/(?P<name>[^/]+)'
+    _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
+
+    @classmethod
+    def suitable(cls, url):
+        if VimeoChannelIE.suitable(url) or VimeoIE.suitable(url) or VimeoAlbumIE.suitable(url) or VimeoGroupsIE.suitable(url):
+            return False
+        return super(VimeoUserIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        return self._extract_videos(name, 'http://vimeo.com/%s' % name)
+
+
+class VimeoAlbumIE(VimeoChannelIE):
+    IE_NAME = u'vimeo:album'
+    _VALID_URL = r'(?:https?://)?vimeo.\com/album/(?P<id>\d+)'
+    _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
+
+    def _page_url(self, base_url, pagenum):
+        return '%s/page:%d/' % (base_url, pagenum)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        album_id =  mobj.group('id')
+        return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
+
+
+class VimeoGroupsIE(VimeoAlbumIE):
+    IE_NAME = u'vimeo:group'
+    _VALID_URL = r'(?:https?://)?vimeo.\com/groups/(?P<name>[^/]+)'
+
+    def _extract_list_title(self, webpage):
+        return self._og_search_title(webpage)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)