X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvine.py;h=80b896b5638ce5c7835c3a5ec8ab687cacc61ae0;hb=HEAD;hp=a919129a67ae7cd73821a56485c4c879827d223f;hpb=78b373975d21f9573c1b32c53aa3c05ab4a490f2;p=youtube-dl diff --git a/youtube_dl/extractor/vine.py b/youtube_dl/extractor/vine.py index a919129a6..80b896b56 100644 --- a/youtube_dl/extractor/vine.py +++ b/youtube_dl/extractor/vine.py @@ -1,44 +1,154 @@ +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + determine_ext, + int_or_none, + unified_timestamp, +) class VineIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vine\.co/v/(?P\w+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?vine\.co/(?:v|oembed)/(?P\w+)' + _TESTS = [{ 'url': 'https://vine.co/v/b9KOOWX7HUx', 'md5': '2f36fed6235b16da96ce9b4dc890940d', 'info_dict': { 'id': 'b9KOOWX7HUx', 'ext': 'mp4', - 'uploader': 'Jack Dorsey', 'title': 'Chicken.', + 'alt_title': 'Vine by Jack', + 'timestamp': 1368997951, + 'upload_date': '20130519', + 'uploader': 'Jack', + 'uploader_id': '76', + 'view_count': int, + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, - } + }, { + 'url': 'https://vine.co/v/e192BnZnZ9V', + 'info_dict': { + 'id': 'e192BnZnZ9V', + 'ext': 'mp4', + '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', + 'only_matching': True, + }, { + '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) - video_id = mobj.group('id') - webpage_url = 'https://vine.co/v/' + video_id - webpage = self._download_webpage(webpage_url, video_id) + data = self._download_json( + 'https://archive.vine.co/posts/%s.json' % video_id, video_id) - self.report_extraction(video_id) + 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 - video_url = self._html_search_meta('twitter:player:stream', webpage, - 'video 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) - twitter_title = self._html_search_meta('twitter:title', webpage, - 'twitter title') - uploader = re.sub('\'s post on Vine', '', twitter_title) + username = data.get('username') + + alt_title = 'Vine by %s' % username if username else None return { 'id': video_id, - 'url': video_url, - 'ext': 'mp4', - 'title': self._og_search_title(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), - 'uploader': uploader, + '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/(?Pu/)?(?P[^/]+)' + _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' % ( + self._VINE_BASE_URL, 'vanity/' if not u else '', user) + profile_data = self._download_json( + profile_url, user, note='Downloading user profile data') + + 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( + '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'))