[extractor/common] simplify using data, headers and query params with _download_* methods
Robin Houtevelts
Patrick Griffis
Aidan Rowe
+mutantmonkey
+Ben Congdon
assertRegexpMatches(self, ydl._format_note({
'vbr': 10,
}), '^\s*10k$')
+ assertRegexpMatches(self, ydl._format_note({
+ 'fps': 30,
+ }), '^30fps$')
def test_postprocessors(self):
filename = 'post-processor-testfile.mp4'
if fdict.get('vbr') is not None:
res += '%4dk' % fdict['vbr']
if fdict.get('fps') is not None:
- res += ', %sfps' % fdict['fps']
+ if res:
+ res += ', '
+ res += '%sfps' % fdict['fps']
if fdict.get('acodec') is not None:
if res:
res += ', '
from .atresplayer import AtresPlayerIE
from .atttechchannel import ATTTechChannelIE
from .audimedia import AudiMediaIE
+from .audioboom import AudioBoomIE
from .audiomack import AudiomackIE, AudiomackAlbumIE
from .azubu import AzubuIE, AzubuLiveIE
from .baidu import BaiduVideoIE
from .exfm import ExfmIE
from .expotv import ExpoTVIE
from .extremetube import ExtremeTubeIE
-from .facebook import (
- FacebookIE,
- FacebookPostIE,
-)
+from .facebook import FacebookIE
from .faz import FazIE
from .fc2 import FC2IE
from .fczenit import FczenitIE
VimeoChannelIE,
VimeoGroupsIE,
VimeoLikesIE,
+ VimeoOndemandIE,
VimeoReviewIE,
VimeoUserIE,
VimeoWatchLaterIE,
json_url = compat_parse_qs(
compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
if json_url:
- return self._extract_from_json_url(json_url, video_id, lang)
- # Differend kind of embed URL (e.g.
+ title = self._search_regex(
+ r'<h3[^>]+title=(["\'])(?P<title>.+?)\1',
+ webpage, 'title', default=None, group='title')
+ return self._extract_from_json_url(json_url, video_id, lang, title=title)
+ # Different kind of embed URL (e.g.
# http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
embed_url = self._search_regex(
r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
webpage, 'embed url', group='url')
return self.url_result(embed_url)
- def _extract_from_json_url(self, json_url, video_id, lang):
+ def _extract_from_json_url(self, json_url, video_id, lang, title=None):
info = self._download_json(json_url, video_id)
player_info = info['videoJsonPlayer']
if not upload_date_str:
upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
- title = player_info['VTI'].strip()
+ title = (player_info.get('VTI') or title or player_info['VID']).strip()
subtitle = player_info.get('VSU', '').strip()
if subtitle:
title += ' - %s' % subtitle
--- /dev/null
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import float_or_none
+
+
+class AudioBoomIE(InfoExtractor):
+ _VALID_URL = r'https?://(?:www\.)?audioboom\.com/boos/(?P<id>[0-9]+)'
+ _TEST = {
+ 'url': 'https://audioboom.com/boos/4279833-3-09-2016-czaban-hour-3?t=0',
+ 'md5': '63a8d73a055c6ed0f1e51921a10a5a76',
+ 'info_dict': {
+ 'id': '4279833',
+ 'ext': 'mp3',
+ 'title': '3/09/2016 Czaban Hour 3',
+ 'description': 'Guest: Nate Davis - NFL free agency, Guest: Stan Gans',
+ 'duration': 2245.72,
+ 'uploader': 'Steve Czaban',
+ 'uploader_url': 're:https?://(?:www\.)?audioboom\.com/channel/steveczabanyahoosportsradio',
+ }
+ }
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, video_id)
+
+ clip = None
+
+ clip_store = self._parse_json(
+ self._search_regex(
+ r'data-new-clip-store=(["\'])(?P<json>{.*?"clipId"\s*:\s*%s.*?})\1' % video_id,
+ webpage, 'clip store', default='{}', group='json'),
+ video_id, fatal=False)
+ if clip_store:
+ clips = clip_store.get('clips')
+ if clips and isinstance(clips, list) and isinstance(clips[0], dict):
+ clip = clips[0]
+
+ def from_clip(field):
+ if clip:
+ clip.get(field)
+
+ audio_url = from_clip('clipURLPriorToLoading') or self._og_search_property(
+ 'audio', webpage, 'audio url')
+ title = from_clip('title') or self._og_search_title(webpage)
+ description = from_clip('description') or self._og_search_description(webpage)
+
+ duration = float_or_none(from_clip('duration') or self._html_search_meta(
+ 'weibo:audio:duration', webpage))
+
+ uploader = from_clip('author') or self._og_search_property(
+ 'audio:artist', webpage, 'uploader', fatal=False)
+ uploader_url = from_clip('author_url') or self._html_search_meta(
+ 'audioboo:channel', webpage, 'uploader url')
+
+ return {
+ 'id': video_id,
+ 'url': audio_url,
+ 'title': title,
+ 'description': description,
+ 'duration': duration,
+ 'uploader': uploader,
+ 'uploader_url': uploader_url,
+ }
continue
representation_attrib = adaptation_set.attrib.copy()
representation_attrib.update(representation.attrib)
- mime_type = representation_attrib.get('mimeType')
- content_type = mime_type.split('/')[0] if mime_type else representation_attrib.get('contentType')
+ # According to page 41 of ISO/IEC 29001-1:2014, @mimeType is mandatory
+ mime_type = representation_attrib['mimeType']
+ content_type = mime_type.split('/')[0]
if content_type == 'text':
# TODO implement WebVTT downloading
pass
f = {
'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
'url': base_url,
+ 'ext': mimetype2ext(mime_type),
'width': int_or_none(representation_attrib.get('width')),
'height': int_or_none(representation_attrib.get('height')),
'tbr': int_or_none(representation_attrib.get('bandwidth'), 1000),
video/embed|
story\.php
)\?(?:.*?)(?:v|video_id|story_fbid)=|
- [^/]+/videos/(?:[^/]+/)?
+ [^/]+/videos/(?:[^/]+/)?|
+ [^/]+/posts/|
+ groups/[^/]+/permalink/
)|
facebook:
)
_CHROME_USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36'
+ _VIDEO_PAGE_TEMPLATE = 'https://www.facebook.com/video/video.php?v=%s'
+
_TESTS = [{
'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
'md5': '6a40d33c0eccbb1af76cf0485a052659',
'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
'uploader': 'Demy de Zeeuw',
},
+ }, {
+ 'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
+ 'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
+ 'info_dict': {
+ 'id': '544765982287235',
+ 'ext': 'mp4',
+ 'title': '"What are you doing running in the snow?"',
+ 'uploader': 'FailArmy',
+ }
+ }, {
+ 'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
+ 'md5': '1deb90b6ac27f7efcf6d747c8a27f5e3',
+ 'info_dict': {
+ 'id': '1035862816472149',
+ 'ext': 'mp4',
+ 'title': 'What the Flock Is Going On In New Zealand Credit: ViralHog',
+ 'uploader': 'S. Saint',
+ },
+ }, {
+ 'note': 'swf params escaped',
+ 'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
+ 'md5': '97ba073838964d12c70566e0085c2b91',
+ 'info_dict': {
+ 'id': '10153664894881749',
+ 'ext': 'mp4',
+ 'title': 'Facebook video #10153664894881749',
+ },
}, {
'url': 'https://www.facebook.com/video.php?v=10204634152394104',
'only_matching': True,
'url': 'facebook:544765982287235',
'only_matching': True,
}, {
- 'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
+ 'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
'only_matching': True,
}]
def _real_initialize(self):
self._login()
- def _real_extract(self, url):
- video_id = self._match_id(url)
- req = sanitized_Request('https://www.facebook.com/video/video.php?v=%s' % video_id)
+ def _extract_from_url(self, url, video_id, fatal_if_no_video=True):
+ req = sanitized_Request(url)
req.add_header('User-Agent', self._CHROME_USER_AGENT)
webpage = self._download_webpage(req, video_id)
video_data = None
- BEFORE = '{swf.addParam(param[0], param[1]);});\n'
+ BEFORE = '{swf.addParam(param[0], param[1]);});'
AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
- m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
+ m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage)
if m:
- data = dict(json.loads(m.group(1)))
+ swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"')
+ data = dict(json.loads(swf_params))
params_raw = compat_urllib_parse_unquote(data['params'])
video_data = json.loads(params_raw)['video_data']
if not video_data:
server_js_data = self._parse_json(self._search_regex(
- r'handleServerJS\(({.+})\);', webpage, 'server js data'), video_id)
+ r'handleServerJS\(({.+})\);', webpage, 'server js data', default='{}'), video_id)
for item in server_js_data.get('instances', []):
if item[1][0] == 'VideoConfig':
video_data = video_data_list2dict(item[2][0]['videoData'])
break
if not video_data:
+ if not fatal_if_no_video:
+ return webpage, False
m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
if m_msg is not None:
raise ExtractorError(
video_title = 'Facebook video #%s' % video_id
uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))
- return {
+ info_dict = {
'id': video_id,
'title': video_title,
'formats': formats,
'uploader': uploader,
}
-
-class FacebookPostIE(InfoExtractor):
- IE_NAME = 'facebook:post'
- _VALID_URL = r'https?://(?:\w+\.)?facebook\.com/[^/]+/posts/(?P<id>\d+)'
- _TEST = {
- 'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
- 'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
- 'info_dict': {
- 'id': '544765982287235',
- 'ext': 'mp4',
- 'title': '"What are you doing running in the snow?"',
- 'uploader': 'FailArmy',
- }
- }
+ return webpage, info_dict
def _real_extract(self, url):
- post_id = self._match_id(url)
+ video_id = self._match_id(url)
+
+ real_url = self._VIDEO_PAGE_TEMPLATE % video_id if url.startswith('facebook:') else url
+ webpage, info_dict = self._extract_from_url(real_url, video_id, fatal_if_no_video=False)
- webpage = self._download_webpage(url, post_id)
+ if info_dict:
+ return info_dict
- entries = [
- self.url_result('facebook:%s' % video_id, FacebookIE.ie_key())
- for video_id in self._parse_json(
- self._search_regex(
- r'(["\'])video_ids\1\s*:\s*(?P<ids>\[.+?\])',
- webpage, 'video ids', group='ids'),
- post_id)]
+ if '/posts/' in url:
+ entries = [
+ self.url_result('facebook:%s' % vid, FacebookIE.ie_key())
+ for vid in self._parse_json(
+ self._search_regex(
+ r'(["\'])video_ids\1\s*:\s*(?P<ids>\[.+?\])',
+ webpage, 'video ids', group='ids'),
+ video_id)]
- return self.playlist_result(entries, post_id)
+ return self.playlist_result(entries, video_id)
+ else:
+ _, info_dict = self._extract_from_url(
+ self._VIDEO_PAGE_TEMPLATE % video_id,
+ video_id, fatal_if_no_video=True)
+ return info_dict
def get_enc_key(self, video_id):
# TODO: automatic key extraction
# last update at 2016-01-22 for Zombie::bite
- enc_key = '6ab6d0280511493ba85594779759d4ed'
+ enc_key = '8ed797d224d043e7ac23d95b70227d32'
return enc_key
def _extract_playlist(self, webpage):
xpath_with_ns,
xpath_text,
orderedSet,
+ update_url_query,
int_or_none,
float_or_none,
parse_iso8601,
for vn in video_nodes:
tbr = int_or_none(vn.attrib.get('system-bitrate'), 1000)
furl = (
- '%s%s?v=3.0.3&fp=WIN%%2014,0,0,145' % (base, vn.attrib['src']))
+ update_url_query(compat_urlparse.urljoin(base, vn.attrib['src']), {
+ 'v': '3.0.3',
+ 'fp': 'WIN% 14,0,0,145',
+ }))
if 'clipBegin' in vn.attrib:
furl += '&ssek=' + vn.attrib['clipBegin']
formats.append({
self._USHER_BASE, item_id,
compat_urllib_parse.urlencode({
'allow_source': 'true',
+ 'allow_audio_only': 'true',
'allow_spectre': 'true',
'player': 'twitchweb',
'nauth': access_token['token'],
query = {
'allow_source': 'true',
+ 'allow_audio_only': 'true',
'p': random.randint(1000000, 10000000),
'player': 'twitchweb',
'segment_preference': '4',
# _VALID_URL matches Vimeo URLs
_VALID_URL = r'''(?x)
- https?://
- (?:(?:www|(?P<player>player))\.)?
- vimeo(?P<pro>pro)?\.com/
- (?!channels/[^/?#]+/?(?:$|[?#])|album/)
- (?:.*?/)?
- (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
- (?:videos?/)?
- (?P<id>[0-9]+)
- /?(?:[?&].*)?(?:[#].*)?$'''
+ https?://
+ (?:
+ (?:
+ www|
+ (?P<player>player)
+ )
+ \.
+ )?
+ vimeo(?P<pro>pro)?\.com/
+ (?!channels/[^/?#]+/?(?:$|[?#])|(?:album|ondemand)/)
+ (?:.*?/)?
+ (?:
+ (?:
+ play_redirect_hls|
+ moogaloop\.swf)\?clip_id=
+ )?
+ (?:videos?/)?
+ (?P<id>[0-9]+)
+ /?(?:[?&].*)?(?:[#].*)?$
+ '''
IE_NAME = 'vimeo'
_TESTS = [
{
}
+class VimeoOndemandIE(VimeoBaseInfoExtractor):
+ IE_NAME = 'vimeo:ondemand'
+ _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
+ _TESTS = [{
+ # ondemand video not available via https://vimeo.com/id
+ 'url': 'https://vimeo.com/ondemand/20704',
+ 'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
+ 'info_dict': {
+ 'id': '105442900',
+ 'ext': 'mp4',
+ 'title': 'המעבדה - במאי יותם פלדמן',
+ 'uploader': 'גם סרטים',
+ 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/gumfilms',
+ 'uploader_id': 'gumfilms',
+ },
+ }, {
+ 'url': 'https://vimeo.com/ondemand/nazmaalik',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://vimeo.com/ondemand/141692381',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
+ 'only_matching': True,
+ }]
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+ webpage = self._download_webpage(url, video_id)
+ return self.url_result(self._og_search_video_url(webpage), VimeoIE.ie_key())
+
+
class VimeoChannelIE(VimeoBaseInfoExtractor):
IE_NAME = 'vimeo:channel'
_VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
projects / youtube-dl / commitdiff