From f57f84f606b246db4f102fc5bc55e64e4f7a3d60 Mon Sep 17 00:00:00 2001 From: fnord Date: Tue, 21 Jul 2015 16:38:40 -0500 Subject: [PATCH 1/3] Twitter: get and describe video from status urls --- youtube_dl/extractor/twitter.py | 44 +++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index 1aaa06305..a65252cc6 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -70,3 +70,47 @@ def _real_extract(self, url): 'duration': duration, 'formats': formats, } + + +class TwitterIE(TwitterCardIE): + _VALID_URL = r'https?://(?:www|m|mobile)?\.?twitter\.com/(?P[^/]+/status/\d+)' + + _TESTS = [{ + 'url': 'https://m.twitter.com/thereaIbanksy/status/614301758345490432', + 'md5': '8bbccb487bd7a31349b775915fcd412f', + 'info_dict': { + 'id': '614301758345490432', + 'ext': 'mp4', + 'title': 'thereaIbanksy - This time lapse is so pretty \U0001f60d\U0001f60d', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 29.5, + 'description': 'banksy on Twitter: "This time lapse is so pretty \U0001f60d\U0001f60d http://t.co/QB8DDbqiR1"', + 'uploader': 'banksy', + 'uploader_id': 'thereaIbanksy', + }, + }] + + def _real_extract(self, url): + id = self._match_id(url) + username, twid = re.match(r'([^/]+)/status/(\d+)', id).groups() + name = username + url = re.sub(r'https?://(m|mobile)\.', 'https://', url) + webpage = self._download_webpage(url, 'tweet: ' + url) + description = unescapeHTML(self._search_regex('\s*(.+?)\s*', webpage, 'title')) + title = description.replace('\n', ' ') + splitdesc = re.match(r'^(.+?)\s*on Twitter:\s* "(.+?)"$', title) + if splitdesc: + name, title = splitdesc.groups() + title = re.sub(r'\s*https?://[^ ]+', '', title) # strip 'https -_t.co_BJYgOjSeGA' junk from filenames + card_id = self._search_regex(r'["\']/i/cards/tfw/v1/(\d+)', webpage, '/i/card/...') + card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id + return { + '_type': 'url_transparent', + 'ie_key': 'TwitterCard', + 'uploader_id': username, + 'uploader': name, + 'url': card_url, + 'webpage_url': url, + 'description': description, + 'title': username + ' - ' + title, + } From c3dea3f878133f3cbdad9e548609d3077572af66 Mon Sep 17 00:00:00 2001 From: fnord Date: Tue, 21 Jul 2015 16:45:36 -0500 Subject: [PATCH 2/3] Twittercard: support vmapurl method --- youtube_dl/extractor/twitter.py | 47 ++++++++++++++++++++++++++------- 1 file changed, 37 insertions(+), 10 deletions(-) diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index a65252cc6..1dd43ff3c 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -12,17 +12,30 @@ class TwitterCardIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P\d+)' - _TEST = { - 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', - 'md5': 'a74f50b310c83170319ba16de6955192', - 'info_dict': { - 'id': '560070183650213889', - 'ext': 'mp4', - 'title': 'TwitterCard', - 'thumbnail': 're:^https?://.*\.jpg$', - 'duration': 30.033, + _TESTS = [ + { + 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', + 'md5': 'a74f50b310c83170319ba16de6955192', + 'info_dict': { + 'id': '560070183650213889', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 30.033, + } }, - } + { + 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768', + 'md5': '7ee2a553b63d1bccba97fbed97d9e1c8', + 'info_dict': { + 'id': '623160978427936768', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 80.155, + }, + } + ] def _real_extract(self, url): video_id = self._match_id(url) @@ -44,6 +57,20 @@ def _real_extract(self, url): unescapeHTML(self._search_regex( r'data-player-config="([^"]+)"', webpage, 'data player config')), video_id) + if 'playlist' not in config: + if 'vmapUrl' in config: + webpage = self._download_webpage(config['vmapUrl'], video_id + ' (xml)') + video_url = self._search_regex( + r'\s*', webpage, 'data player config (xml)') + f = { + 'url': video_url, + } + ext = re.search(r'\.([a-z0-9]{2,4})(\?.+)?$', video_url) + if ext: + f['ext'] = ext.group(1) + formats.append(f) + break # same video regardless of UA + continue video_url = config['playlist'][0]['source'] From 9e7e0dffd5e3e3c959e8d99a5e236b9099886fe9 Mon Sep 17 00:00:00 2001 From: fnord Date: Tue, 21 Jul 2015 16:56:35 -0500 Subject: [PATCH 3/3] Actually add the extractor --- youtube_dl/extractor/__init__.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 50da08830..5c03bf8e8 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -651,7 +651,7 @@ TwitchBookmarksIE, TwitchStreamIE, ) -from .twitter import TwitterCardIE +from .twitter import TwitterCardIE, TwitterIE from .ubu import UbuIE from .udemy import ( UdemyIE,