debian/control: Update list of extractors.

[youtubedl] / youtube_dl / extractor / vimeo.py

diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py
index 20dc255d4c3e8de8dd4dd4cc30a60cb0427135de..f27763ae2ff110051613243ef2543e3349d6b019 100644 (file)
--- a/youtube_dl/extractor/vimeo.py
+++ b/youtube_dl/extractor/vimeo.py
@@ -1,5 +1,7 @@
+# encoding: utf-8

 import json
 import re
 import json
 import re

+import itertools

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


@@ -9,15 +11,87 @@ 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'(?P<proto>https?://)?(?:(?:www|(?P<player>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#at=0',
+            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',
+            },
+        },
+        {
+            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):
+        (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):
         password = self._downloader.params.get('videopassword', None)
 
     def _verify_video_password(self, url, video_id, webpage):
         password = self._downloader.params.get('videopassword', None)


@@ -38,20 +112,29 @@ 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):
     def _real_extract(self, url, new_video=True):

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


@@ -61,17 +144,26 @@ class VimeoIE(InfoExtractor):
 
         # Extract the config JSON
         try:
 
         # Extract the config JSON
         try:

-            config = webpage.split(' = {config:')[1].split(',assets:')[0]
-            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
+                config = self._search_regex([r' = {config:({.+?}),assets:', r'(?:c|b)=({.+?});'],
+                    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"]


@@ -81,12 +173,22 @@ 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


@@ -100,39 +202,93 @@ class VimeoIE(InfoExtractor):
 
         # Vimeo specific: extract video codec and quality information
         # First consider quality, then codecs, then take everything
 
         # 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:
         for codec_name, codec_extension in codecs:

-            if codec_name in config["video"]["files"]:
-                if 'hd' in config["video"]["files"][codec_name]:
-                    files['hd'].append((codec_name, codec_extension, 'hd'))
-                elif 'sd' in config["video"]["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["video"]["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 = "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,
+        }
+
+
+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 _extract_videos(self, list_id, base_url):
+        video_ids = []
+        for pagenum in itertools.count(1):
+            webpage = self._download_webpage(
+                '%s/videos/page:%d/' % (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]
+        list_title = self._html_search_regex(self._TITLE_RE, webpage,
+            u'list title')
+        return {'_type': 'playlist',
+                'id': list_id,
+                'title': list_title,
+                '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):
+            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)