Merge branch 'dcn' of github.com:remitamine/youtube-dl into remitamine-dcn

[youtube-dl] / youtube_dl / extractor / vimeo.py

diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py
index d1daf9672995db3c4490e2d9077955fa2c4d5665..ce08e69559370c1fed3e393cee729fdac40119e6 100644 (file)
--- a/youtube_dl/extractor/vimeo.py
+++ b/youtube_dl/extractor/vimeo.py
@@ -8,7 +8,6 @@ import itertools
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
-    compat_urllib_request_Request,
     compat_urlparse,
 )
 from ..utils import (
@@ -17,12 +16,14 @@ from ..utils import (
     InAdvancePagedList,
     int_or_none,
     RegexNotFoundError,
+    sanitized_Request,
     smuggle_url,
     std_headers,
     unified_strdate,
     unsmuggle_url,
     urlencode_postdata,
     unescapeHTML,
+    parse_filesize,
 )
 
 
@@ -47,7 +48,7 @@ class VimeoBaseInfoExtractor(InfoExtractor):
             'service': 'vimeo',
             'token': token,
         }))
-        login_request = compat_urllib_request_Request(self._LOGIN_URL, data)
+        login_request = sanitized_Request(self._LOGIN_URL, data)
         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         login_request.add_header('Referer', self._LOGIN_URL)
         self._set_vimeo_cookie('vuid', vuid)
@@ -184,6 +185,20 @@ class VimeoIE(VimeoBaseInfoExtractor):
                 'uploader_id': 'user28849593',
             },
         },
+        {
+            # contains original format
+            'url': 'https://vimeo.com/33951933',
+            'md5': '53c688fa95a55bf4b7293d37a89c5c53',
+            'info_dict': {
+                'id': '33951933',
+                'ext': 'mp4',
+                'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
+                'uploader': 'The DMCI',
+                'uploader_id': 'dmci',
+                'upload_date': '20111220',
+                'description': 'md5:ae23671e82d05415868f7ad1aec21147',
+            },
+        },
         {
             'url': 'https://vimeo.com/109815029',
             'note': 'Video not completely processed, "failed" seed status',
@@ -222,7 +237,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
         if url.startswith('http://'):
             # vimeo only supports https now, but the user can give an http url
             url = url.replace('http://', 'https://')
-        password_request = compat_urllib_request_Request(url + '/password', data)
+        password_request = sanitized_Request(url + '/password', data)
         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         password_request.add_header('Referer', url)
         self._set_vimeo_cookie('vuid', vuid)
@@ -236,7 +251,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
             raise ExtractorError('This video is protected by a password, use the --video-password option')
         data = urlencode_postdata(encode_dict({'password': password}))
         pass_url = url + '/check-password'
-        password_request = compat_urllib_request_Request(pass_url, data)
+        password_request = sanitized_Request(pass_url, data)
         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         return self._download_json(
             password_request, video_id,
@@ -265,7 +280,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
             url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
-        request = compat_urllib_request_Request(url, None, headers)
+        request = sanitized_Request(url, None, headers)
         try:
             webpage = self._download_webpage(request, video_id)
         except ExtractorError as ee:
@@ -392,6 +407,21 @@ class VimeoIE(VimeoBaseInfoExtractor):
             comment_count = None
 
         formats = []
+        download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
+            'X-Requested-With': 'XMLHttpRequest'})
+        download_data = self._download_json(download_request, video_id, fatal=False)
+        if download_data:
+            source_file = download_data.get('source_file')
+            if source_file and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
+                formats.append({
+                    'url': source_file['download_url'],
+                    'ext': source_file['extension'].lower(),
+                    'width': int_or_none(source_file.get('width')),
+                    'height': int_or_none(source_file.get('height')),
+                    'filesize': parse_filesize(source_file.get('size')),
+                    'format_id': source_file.get('public_name', 'Original'),
+                    'preference': 1,
+                })
         config_files = config['video'].get('files') or config['request'].get('files', {})
         for f in config_files.get('progressive', []):
             video_url = f.get('url')
@@ -408,12 +438,12 @@ class VimeoIE(VimeoBaseInfoExtractor):
         m3u8_url = config_files.get('hls', {}).get('url')
         if m3u8_url:
             m3u8_formats = self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False)
+                m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
             if m3u8_formats:
                 formats.extend(m3u8_formats)
         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
         # at the same time without actual units specified. This lead to wrong sorting.
-        self._sort_formats(formats, field_preference=('height', 'width', 'fps', 'format_id'))
+        self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'format_id'))
 
         subtitles = {}
         text_tracks = config['request'].get('text_tracks')
@@ -481,7 +511,7 @@ class VimeoChannelIE(VimeoBaseInfoExtractor):
         password_path = self._search_regex(
             r'action="([^"]+)"', login_form, 'password URL')
         password_url = compat_urlparse.urljoin(page_url, password_path)
-        password_request = compat_urllib_request_Request(password_url, post)
+        password_request = sanitized_Request(password_url, post)
         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
         self._set_vimeo_cookie('vuid', vuid)
         self._set_vimeo_cookie('xsrft', token)
@@ -640,7 +670,7 @@ class VimeoWatchLaterIE(VimeoChannelIE):
 
     def _page_url(self, base_url, pagenum):
         url = '%s/page:%d/' % (base_url, pagenum)
-        request = compat_urllib_request_Request(url)
+        request = sanitized_Request(url)
         # Set the header to get a partial html page with the ids,
         # the normal page doesn't contain them.
         request.add_header('X-Requested-With', 'XMLHttpRequest')