From 778f96944785f814a97964be1d6fb3bb78bc13f0 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Tue, 14 Jun 2016 00:06:31 +0700 Subject: [PATCH] [twitch:clips] Add extractor (Closes #9767) --- youtube_dl/extractor/extractors.py | 1 + youtube_dl/extractor/twitch.py | 43 ++++++++++++++++++++++++++++++ 2 files changed, 44 insertions(+) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 36ddc1f73..d2db4d803 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -862,6 +862,7 @@ from .twitch import ( TwitchProfileIE, TwitchPastBroadcastsIE, TwitchStreamIE, + TwitchClipsIE, ) from .twitter import ( TwitterCardIE, diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index d898f14c3..20919774d 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -16,6 +16,7 @@ from ..compat import ( from ..utils import ( ExtractorError, int_or_none, + js_to_json, orderedSet, parse_duration, parse_iso8601, @@ -454,3 +455,45 @@ class TwitchStreamIE(TwitchBaseIE): 'formats': formats, 'is_live': True, } + + +class TwitchClipsIE(InfoExtractor): + IE_NAME = 'twitch:clips' + _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P[^/?#&]+)' + + _TEST = { + 'url': 'https://clips.twitch.tv/ea/AggressiveCobraPoooound', + 'md5': '761769e1eafce0ffebfb4089cb3847cd', + 'info_dict': { + 'id': 'AggressiveCobraPoooound', + 'ext': 'mp4', + 'title': 'EA Play 2016 Live from the Novo Theatre', + 'thumbnail': 're:^https?://.*\.jpg', + 'creator': 'EA', + 'uploader': 'stereotype_', + 'uploader_id': 'stereotype_', + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + clip = self._parse_json( + self._search_regex( + r'(?s)clipInfo\s*=\s*({.+?});', webpage, 'clip info'), + video_id, transform_source=js_to_json) + + video_url = clip['clip_video_url'] + title = clip['channel_title'] + + return { + 'id': video_id, + 'url': video_url, + 'title': title, + 'thumbnail': self._og_search_thumbnail(webpage), + 'creator': clip.get('broadcaster_display_name') or clip.get('broadcaster_login'), + 'uploader': clip.get('curator_login'), + 'uploader_id': clip.get('curator_display_name'), + } -- 2.30.2