jdormit/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[crooksandliars] Improve embed extractor and remove article extractor

Browse Source
This commit is contained in:
Sergey M․ 2015-04-11 20:03:12 +06:00
parent af14ded75e
commit 7a91d1fc43
2 changed files with 39 additions and 50 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
youtube_dl/extractor/__init__.py
View File

@ -90,7 +90,7 @@ from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
from .condenast import CondeNastIE
from .cracked import CrackedIE
from .criterion import CriterionIE
from .crooksandliars import CrooksAndLiarsIE, CrooksAndLiarsArticleIE
from .crooksandliars import CrooksAndLiarsIE
from .crunchyroll import (
CrunchyrollIE,
CrunchyrollShowPlaylistIE

87
youtube_dl/extractor/crooksandliars.py
View File

@ -1,71 +1,60 @@
from __future__ import unicode_literals
import json
from .common import InfoExtractor
from ..utils import (
mimetype2ext,
int_or_none,
qualities,
)
class CrooksAndLiarsIE(InfoExtractor):
_VALID_URL = r'(?:https?:)?//embed.crooksandliars.com/embed/(?P<id>[A-Za-z0-9]+)(?:$|[?#])'
_VALID_URL = r'https?://embed\.crooksandliars\.com/(?:embed|v)/(?P<id>[A-Za-z0-9]+)'
_TESTS = [{
'url': 'https://embed.crooksandliars.com/embed/8RUoRhRi',
'info_dict': {
'id': 'https://embed.crooksandliars.com/embed/8RUoRhRi',
'id': '8RUoRhRi',
'ext': 'mp4',
'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
'description': "Fox News, Fox & Friends Weekend, April 4, 2015. Read more... http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists",
'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
'thumbnail': 're:^https?://.*\.jpg',
'timestamp': 1428207000,
'thumbnail': 'https://crooksandliars.com/files/mediaposters/2015/04/31235.jpg?ts=1428207050',
'uploader': "Heather",
'upload_date': '20150405',
'uploader': 'Heather',
'duration': 236,
}
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
manifest = json.loads(self._html_search_regex(r'var manifest = ({.*?})\n', webpage, 'manifest JSON'))
formats = []
for item in manifest['flavors']:
if not item['mime'].startswith('video/'): # XXX: or item['exclude']?
continue
formats.append({
'format_id': item['type'],
'ext': mimetype2ext(item['mime']),
'url': item['url'],
})
# XXX: manifest['url']?
return {
'url': url,
'id': video_id,
'uploader': manifest['author'],
'title': manifest['title'],
'description': manifest['description'],
'thumbnail': self._proto_relative_url(manifest['poster']),
'duration': manifest['duration'],
'timestamp': int(manifest['created']),
'formats': formats,
}
class CrooksAndLiarsArticleIE(InfoExtractor):
_VALID_URL = r'(?:https?:)?//crooksandliars.com/\d+/\d+/(?P<id>[a-z\-]+)(?:/|$)'
_TESTS = [{
'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
}, {
'url': 'http://embed.crooksandliars.com/v/MTE3MjUtMzQ2MzA',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
player_url = self._proto_relative_url(self._html_search_regex(r'<iframe src="(//embed.crooksandliars.com/.*)"', webpage, 'embedded player'))
webpage = self._download_webpage(
'http://embed.crooksandliars.com/embed/%s' % video_id, video_id)
manifest = self._parse_json(
self._search_regex(
r'var\s+manifest\s*=\s*({.+?})\n', webpage, 'manifest JSON'),
video_id)
quality = qualities(('webm_low', 'mp4_low', 'webm_high', 'mp4_high'))
formats = [{
'url': item['url'],
'format_id': item['type'],
'quality': quality(item['type']),
} for item in manifest['flavors'] if item['mime'].startswith('video/')]
self._sort_formats(formats)
return {
'_type': 'url',
'url': player_url
'url': url,
'id': video_id,
'title': manifest['title'],
'description': manifest.get('description'),
'thumbnail': self._proto_relative_url(manifest.get('poster')),
'timestamp': int_or_none(manifest.get('created')),
'uploader': manifest.get('author'),
'duration': int_or_none(manifest.get('duration')),
'formats': formats,
}