1
0
Fork 0
mirror of https://codeberg.org/polarisfm/youtube-dl.git synced 2024-12-25 01:01:39 +00:00

[nbc:nbcnews] improve extraction and add msnbc to the extractor

This commit is contained in:
Remita Amine 2016-06-23 01:36:19 +01:00
parent 22b7ac13ef
commit 0437307a41
2 changed files with 50 additions and 108 deletions

View file

@ -480,7 +480,6 @@ from .nbc import (
NBCNewsIE, NBCNewsIE,
NBCSportsIE, NBCSportsIE,
NBCSportsVPlayerIE, NBCSportsVPlayerIE,
MSNBCIE,
) )
from .ndr import ( from .ndr import (
NDRIE, NDRIE,

View file

@ -9,10 +9,6 @@ from ..utils import (
lowercase_escape, lowercase_escape,
smuggle_url, smuggle_url,
unescapeHTML, unescapeHTML,
update_url_query,
int_or_none,
HEADRequest,
parse_iso8601,
) )
@ -192,9 +188,9 @@ class CSNNEIE(InfoExtractor):
class NBCNewsIE(ThePlatformIE): class NBCNewsIE(ThePlatformIE):
_VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today)\.com/ _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/
(?:video/.+?/(?P<id>\d+)| (?:video/.+?/(?P<id>\d+)|
([^/]+/)*(?P<display_id>[^/?]+)) ([^/]+/)*(?:.*-)?(?P<mpx_id>[^/?]+))
''' '''
_TESTS = [ _TESTS = [
@ -216,13 +212,16 @@ class NBCNewsIE(ThePlatformIE):
'ext': 'mp4', 'ext': 'mp4',
'title': 'How Twitter Reacted To The Snowden Interview', 'title': 'How Twitter Reacted To The Snowden Interview',
'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64', 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
'uploader': 'NBCU-NEWS',
'timestamp': 1401363060,
'upload_date': '20140529',
}, },
}, },
{ {
'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156', 'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
'md5': 'fdbf39ab73a72df5896b6234ff98518a', 'md5': 'fdbf39ab73a72df5896b6234ff98518a',
'info_dict': { 'info_dict': {
'id': 'Wjf9EDR3A_60', 'id': '529953347624',
'ext': 'mp4', 'ext': 'mp4',
'title': 'FULL EPISODE: Family Business', 'title': 'FULL EPISODE: Family Business',
'description': 'md5:757988edbaae9d7be1d585eb5d55cc04', 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
@ -237,6 +236,9 @@ class NBCNewsIE(ThePlatformIE):
'ext': 'mp4', 'ext': 'mp4',
'title': 'Nightly News with Brian Williams Full Broadcast (February 4)', 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5', 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
'timestamp': 1423104900,
'uploader': 'NBCU-NEWS',
'upload_date': '20150205',
}, },
}, },
{ {
@ -245,10 +247,12 @@ class NBCNewsIE(ThePlatformIE):
'info_dict': { 'info_dict': {
'id': '529953347624', 'id': '529953347624',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Volkswagen U.S. Chief: We \'Totally Screwed Up\'', 'title': 'Volkswagen U.S. Chief:\xa0 We Have Totally Screwed Up',
'description': 'md5:d22d1281a24f22ea0880741bb4dd6301', 'description': 'md5:c8be487b2d80ff0594c005add88d8351',
'upload_date': '20150922',
'timestamp': 1442917800,
'uploader': 'NBCU-NEWS',
}, },
'expected_warnings': ['http-6000 is not available']
}, },
{ {
'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788', 'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
@ -260,6 +264,22 @@ class NBCNewsIE(ThePlatformIE):
'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1', 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
'upload_date': '20160420', 'upload_date': '20160420',
'timestamp': 1461152093, 'timestamp': 1461152093,
'uploader': 'NBCU-NEWS',
},
},
{
'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
'info_dict': {
'id': '314487875924',
'ext': 'mp4',
'title': 'The chaotic GOP immigration vote',
'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
'thumbnail': 're:^https?://.*\.jpg$',
'timestamp': 1406937606,
'upload_date': '20140802',
'uploader': 'NBCU-NEWS',
'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
}, },
}, },
{ {
@ -290,105 +310,28 @@ class NBCNewsIE(ThePlatformIE):
} }
else: else:
# "feature" and "nightly-news" pages use theplatform.com # "feature" and "nightly-news" pages use theplatform.com
display_id = mobj.group('display_id') video_id = mobj.group('mpx_id')
webpage = self._download_webpage(url, display_id) if not video_id.isdigit():
info = None webpage = self._download_webpage(url, video_id)
bootstrap_json = self._search_regex( info = None
[r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$', bootstrap_json = self._search_regex(
r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'], [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$',
webpage, 'bootstrap json', default=None) r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'],
bootstrap = self._parse_json( webpage, 'bootstrap json', default=None)
bootstrap_json, display_id, transform_source=unescapeHTML) bootstrap = self._parse_json(
if 'results' in bootstrap: bootstrap_json, video_id, transform_source=unescapeHTML)
info = bootstrap['results'][0]['video'] if 'results' in bootstrap:
elif 'video' in bootstrap: info = bootstrap['results'][0]['video']
info = bootstrap['video'] elif 'video' in bootstrap:
else: info = bootstrap['video']
info = bootstrap
video_id = info['mpxId']
title = info['title']
subtitles = {}
caption_links = info.get('captionLinks')
if caption_links:
for (sub_key, sub_ext) in (('smpte-tt', 'ttml'), ('web-vtt', 'vtt'), ('srt', 'srt')):
sub_url = caption_links.get(sub_key)
if sub_url:
subtitles.setdefault('en', []).append({
'url': sub_url,
'ext': sub_ext,
})
formats = []
for video_asset in info['videoAssets']:
video_url = video_asset.get('publicUrl')
if not video_url:
continue
container = video_asset.get('format')
asset_type = video_asset.get('assetType') or ''
if container == 'ISM' or asset_type == 'FireTV-Once':
continue
elif asset_type == 'OnceURL':
tp_formats, tp_subtitles = self._extract_theplatform_smil(
video_url, video_id)
formats.extend(tp_formats)
subtitles = self._merge_subtitles(subtitles, tp_subtitles)
else: else:
tbr = int_or_none(video_asset.get('bitRate') or video_asset.get('bitrate'), 1000) info = bootstrap
format_id = 'http%s' % ('-%d' % tbr if tbr else '') video_id = info['mpxId']
video_url = update_url_query(
video_url, {'format': 'redirect'})
# resolve the url so that we can check availability and detect the correct extension
head = self._request_webpage(
HEADRequest(video_url), video_id,
'Checking %s url' % format_id,
'%s is not available' % format_id,
fatal=False)
if head:
video_url = head.geturl()
formats.append({
'format_id': format_id,
'url': video_url,
'width': int_or_none(video_asset.get('width')),
'height': int_or_none(video_asset.get('height')),
'tbr': tbr,
'container': video_asset.get('format'),
})
self._sort_formats(formats)
return { return {
'_type': 'url_transparent',
'id': video_id, 'id': video_id,
'title': title, # http://feed.theplatform.com/f/2E2eJC/nbcnews also works
'description': info.get('description'), 'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byId=%s' % video_id,
'thumbnail': info.get('thumbnail'), 'ie_key': 'ThePlatformFeed',
'duration': int_or_none(info.get('duration')),
'timestamp': parse_iso8601(info.get('pubDate') or info.get('pub_date')),
'formats': formats,
'subtitles': subtitles,
} }
class MSNBCIE(InfoExtractor):
# https URLs redirect to corresponding http ones
_VALID_URL = r'https?://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
_TEST = {
'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
'info_dict': {
'id': 'n_hayes_Aimm_140801_272214',
'ext': 'mp4',
'title': 'The chaotic GOP immigration vote',
'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
'thumbnail': 're:^https?://.*\.jpg$',
'timestamp': 1406937606,
'upload_date': '20140802',
'uploader': 'NBCU-NEWS',
'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
},
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
embed_url = self._html_search_meta('embedURL', webpage)
return self.url_result(embed_url)