debian/changelog: Annotate bugs to close.

[youtubedl] / youtube_dl / extractor / vine.py

diff --git a/youtube_dl/extractor/vine.py b/youtube_dl/extractor/vine.py
index 651ba317dcd8fffefb2ac938c9f09d5de8356865..c733a48fa26edce6b219d3bf2404267b8c346bf6 100644 (file)
--- a/youtube_dl/extractor/vine.py
+++ b/youtube_dl/extractor/vine.py
@@ -1,40 +1,132 @@
+from __future__ import unicode_literals
+
 import re
+import itertools
 
 from .common import InfoExtractor
+from ..utils import unified_strdate
 
 
 class VineIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?vine\.co/v/(?P<id>\w+)'
-    _TEST = {
-        u'url': u'https://vine.co/v/b9KOOWX7HUx',
-        u'file': u'b9KOOWX7HUx.mp4',
-        u'md5': u'2f36fed6235b16da96ce9b4dc890940d',
-        u'info_dict': {
-            u"uploader": u"Jack Dorsey", 
-            u"title": u"Chicken."
-        }
-    }
+    _VALID_URL = r'https?://(?:www\.)?vine\.co/(?:v|oembed)/(?P<id>\w+)'
+    _TESTS = [{
+        'url': 'https://vine.co/v/b9KOOWX7HUx',
+        'md5': '2f36fed6235b16da96ce9b4dc890940d',
+        'info_dict': {
+            'id': 'b9KOOWX7HUx',
+            'ext': 'mp4',
+            'title': 'Chicken.',
+            'alt_title': 'Vine by Jack Dorsey',
+            'description': 'Chicken.',
+            'upload_date': '20130519',
+            'uploader': 'Jack Dorsey',
+            'uploader_id': '76',
+        },
+    }, {
+        'url': 'https://vine.co/v/MYxVapFvz2z',
+        'md5': '7b9a7cbc76734424ff942eb52c8f1065',
+        'info_dict': {
+            'id': 'MYxVapFvz2z',
+            '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',
+        },
+    }, {
+        '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',
+        },
+    }, {
+        'url': 'https://vine.co/oembed/MYxVapFvz2z.json',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
+        webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id)
 
-        video_id = mobj.group('id')
-        webpage_url = 'https://vine.co/v/' + video_id
-        webpage = self._download_webpage(webpage_url, video_id)
+        data = self._parse_json(
+            self._html_search_regex(
+                r'window\.POST_DATA = { %s: ({.+?}) };\s*</script>' % video_id,
+                webpage, 'vine data'),
+            video_id)
 
-        self.report_extraction(video_id)
+        formats = [{
+            'format_id': '%(format)s-%(rate)s' % f,
+            'vcodec': f['format'],
+            'quality': f['rate'],
+            'url': f['videoUrl'],
+        } for f in data['videoUrls']]
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'alt_title': self._og_search_description(webpage, default=None),
+            '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'],
+            '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,
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user = mobj.group('user')
+        u = mobj.group('u')
 
-        video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"',
-            webpage, u'video URL')
+        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')
 
-        uploader = self._html_search_regex(r'<p class="username">(.*?)</p>',
-            webpage, u'uploader', fatal=False, flags=re.DOTALL)
+        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
 
-        return [{
-            'id':        video_id,
-            'url':       video_url,
-            'ext':       'mp4',
-            'title':     self._og_search_title(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'uploader':  uploader,
-        }]
+        entries = [
+            self.url_result(e['permalinkUrl'], 'Vine') for e in timeline_data]
+        return self.playlist_result(entries, user)