[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / vine.py
index 65c459fad39c36798d3d262d37ee95e52a2c313c..80b896b5638ce5c7835c3a5ec8ab687cacc61ae0 100644 (file)
@@ -1,10 +1,15 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
-import itertools
 
 from .common import InfoExtractor
-from ..utils import unified_strdate
+from ..compat import compat_str
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    unified_timestamp,
+)
 
 
 class VineIE(InfoExtractor):
@@ -16,38 +21,41 @@ class VineIE(InfoExtractor):
             'id': 'b9KOOWX7HUx',
             'ext': 'mp4',
             'title': 'Chicken.',
-            'alt_title': 'Vine by Jack Dorsey',
-            'description': 'Chicken.',
+            'alt_title': 'Vine by Jack',
+            'timestamp': 1368997951,
             'upload_date': '20130519',
-            'uploader': 'Jack Dorsey',
+            'uploader': 'Jack',
             'uploader_id': '76',
+            'view_count': int,
+            'like_count': int,
+            'comment_count': int,
+            'repost_count': int,
         },
     }, {
-        'url': 'https://vine.co/v/MYxVapFvz2z',
-        'md5': '7b9a7cbc76734424ff942eb52c8f1065',
+        'url': 'https://vine.co/v/e192BnZnZ9V',
         'info_dict': {
-            'id': 'MYxVapFvz2z',
+            'id': 'e192BnZnZ9V',
             'ext': 'mp4',
-            'title': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14',
-            'alt_title': 'Vine by Luna',
-            'description': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14',
-            'upload_date': '20140815',
-            'uploader': 'Luna',
-            'uploader_id': '1102363502380728320',
+            'title': 'ยิ้ม~ เขิน~ อาย~ น่าร้ากอ้ะ >//< @n_whitewo @orlameena #lovesicktheseries  #lovesickseason2',
+            'alt_title': 'Vine by Pimry_zaa',
+            'timestamp': 1436057405,
+            'upload_date': '20150705',
+            'uploader': 'Pimry_zaa',
+            'uploader_id': '1135760698325307392',
+            'view_count': int,
+            'like_count': int,
+            'comment_count': int,
+            'repost_count': int,
+        },
+        'params': {
+            'skip_download': True,
         },
+    }, {
+        'url': 'https://vine.co/v/MYxVapFvz2z',
+        'only_matching': True,
     }, {
         'url': 'https://vine.co/v/bxVjBbZlPUH',
-        'md5': 'ea27decea3fa670625aac92771a96b73',
-        'info_dict': {
-            'id': 'bxVjBbZlPUH',
-            'ext': 'mp4',
-            'title': '#mw3 #ac130 #killcam #angelofdeath',
-            'alt_title': 'Vine by Z3k3',
-            'description': '#mw3 #ac130 #killcam #angelofdeath',
-            'upload_date': '20130430',
-            'uploader': 'Z3k3',
-            'uploader_id': '936470460173008896',
-        },
+        'only_matching': True,
     }, {
         'url': 'https://vine.co/oembed/MYxVapFvz2z.json',
         'only_matching': True,
@@ -55,78 +63,92 @@ class VineIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id)
 
-        data = self._parse_json(
-            self._html_search_regex(
-                r'window\.POST_DATA = { %s: ({.+?}) };\s*</script>' % video_id,
-                webpage, 'vine data'),
-            video_id)
+        data = self._download_json(
+            'https://archive.vine.co/posts/%s.json' % video_id, video_id)
 
-        formats = [{
-            'format_id': '%(format)s-%(rate)s' % f,
-            'vcodec': f['format'],
-            'quality': f['rate'],
-            'url': f['videoUrl'],
-        } for f in data['videoUrls']]
+        def video_url(kind):
+            for url_suffix in ('Url', 'URL'):
+                format_url = data.get('video%s%s' % (kind, url_suffix))
+                if format_url:
+                    return format_url
 
+        formats = []
+        for quality, format_id in enumerate(('low', '', 'dash')):
+            format_url = video_url(format_id.capitalize())
+            if not format_url:
+                continue
+            # DASH link returns plain mp4
+            if format_id == 'dash' and determine_ext(format_url) == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    format_url, video_id, mpd_id='dash', fatal=False))
+            else:
+                formats.append({
+                    'url': format_url,
+                    'format_id': format_id or 'standard',
+                    'quality': quality,
+                })
         self._sort_formats(formats)
 
+        username = data.get('username')
+
+        alt_title = 'Vine by %s' % username if username else None
+
         return {
             'id': video_id,
-            'title': self._og_search_title(webpage),
-            'alt_title': self._og_search_description(webpage),
-            'description': data['description'],
-            'thumbnail': data['thumbnailUrl'],
-            'upload_date': unified_strdate(data['created']),
-            'uploader': data['username'],
-            'uploader_id': data['userIdStr'],
-            'like_count': data['likes']['count'],
-            'comment_count': data['comments']['count'],
-            'repost_count': data['reposts']['count'],
+            'title': data.get('description') or alt_title or 'Vine video',
+            'alt_title': alt_title,
+            'thumbnail': data.get('thumbnailUrl'),
+            'timestamp': unified_timestamp(data.get('created')),
+            'uploader': username,
+            'uploader_id': data.get('userIdStr'),
+            'view_count': int_or_none(data.get('loops')),
+            'like_count': int_or_none(data.get('likes')),
+            'comment_count': int_or_none(data.get('comments')),
+            'repost_count': int_or_none(data.get('reposts')),
             'formats': formats,
         }
 
 
 class VineUserIE(InfoExtractor):
     IE_NAME = 'vine:user'
-    _VALID_URL = r'(?:https?://)?vine\.co/(?P<u>u/)?(?P<user>[^/]+)/?(\?.*)?$'
-    _VINE_BASE_URL = "https://vine.co/"
-    _TESTS = [
-        {
-            'url': 'https://vine.co/Visa',
-            'info_dict': {
-                'id': 'Visa',
-            },
-            'playlist_mincount': 46,
-        },
-        {
-            'url': 'https://vine.co/u/941705360593584128',
-            'only_matching': True,
+    _VALID_URL = r'https?://vine\.co/(?P<u>u/)?(?P<user>[^/]+)'
+    _VINE_BASE_URL = 'https://vine.co/'
+    _TESTS = [{
+        'url': 'https://vine.co/itsruthb',
+        'info_dict': {
+            'id': 'itsruthb',
+            'title': 'Ruth B',
+            'description': '| Instagram/Twitter: itsruthb | still a lost boy from neverland',
         },
-    ]
+        'playlist_mincount': 611,
+    }, {
+        'url': 'https://vine.co/u/942914934646415360',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if VineIE.suitable(url) else super(VineUserIE, cls).suitable(url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         user = mobj.group('user')
         u = mobj.group('u')
 
-        profile_url = "%sapi/users/profiles/%s%s" % (
+        profile_url = '%sapi/users/profiles/%s%s' % (
             self._VINE_BASE_URL, 'vanity/' if not u else '', user)
         profile_data = self._download_json(
             profile_url, user, note='Downloading user profile data')
 
-        user_id = profile_data['data']['userId']
-        timeline_data = []
-        for pagenum in itertools.count(1):
-            timeline_url = "%sapi/timelines/users/%s?page=%s&size=100" % (
-                self._VINE_BASE_URL, user_id, pagenum)
-            timeline_page = self._download_json(
-                timeline_url, user, note='Downloading page %d' % pagenum)
-            timeline_data.extend(timeline_page['data']['records'])
-            if timeline_page['data']['nextPage'] is None:
-                break
-
+        data = profile_data['data']
+        user_id = data.get('userId') or data['userIdStr']
+        profile = self._download_json(
+            'https://archive.vine.co/profiles/%s.json' % user_id, user_id)
         entries = [
-            self.url_result(e['permalinkUrl'], 'Vine') for e in timeline_data]
-        return self.playlist_result(entries, user)
+            self.url_result(
+                'https://vine.co/v/%s' % post_id, ie='Vine', video_id=post_id)
+            for post_id in profile['posts']
+            if post_id and isinstance(post_id, compat_str)]
+        return self.playlist_result(
+            entries, user, profile.get('username'), profile.get('description'))