youtube-dl/youtube_dl/extractor/appletrailers.py

284 lines
10 KiB
Python
Raw Permalink Normal View History

2014-01-07 09:04:48 +00:00
from __future__ import unicode_literals
2013-08-28 00:18:44 +00:00
import re
import json
2013-08-28 00:18:44 +00:00
from .common import InfoExtractor
from ..compat import compat_urlparse
2013-08-28 00:18:44 +00:00
from ..utils import (
int_or_none,
parse_duration,
unified_strdate,
2013-08-28 00:18:44 +00:00
)
class AppleTrailersIE(InfoExtractor):
IE_NAME = 'appletrailers'
_VALID_URL = r'https?://(?:www\.|movie)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)'
2015-02-22 13:58:39 +00:00
_TESTS = [{
2015-10-05 14:21:53 +00:00
'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
2015-02-17 23:49:10 +00:00
'info_dict': {
'id': '5111',
'title': 'Man of Steel',
2015-02-17 23:49:10 +00:00
},
2015-10-05 14:21:53 +00:00
'playlist': [
2013-08-28 00:18:44 +00:00
{
2015-10-05 14:21:53 +00:00
'md5': 'd97a8e575432dbcb81b7c3acb741f8a8',
'info_dict': {
'id': 'manofsteel-trailer4',
'ext': 'mov',
'duration': 111,
'title': 'Trailer 4',
'upload_date': '20130523',
'uploader_id': 'wb',
2013-08-28 00:18:44 +00:00
},
},
{
2015-10-05 14:21:53 +00:00
'md5': 'b8017b7131b721fb4e8d6f49e1df908c',
'info_dict': {
'id': 'manofsteel-trailer3',
'ext': 'mov',
'duration': 182,
'title': 'Trailer 3',
'upload_date': '20130417',
'uploader_id': 'wb',
2013-08-28 00:18:44 +00:00
},
},
{
2015-10-05 14:21:53 +00:00
'md5': 'd0f1e1150989b9924679b441f3404d48',
'info_dict': {
'id': 'manofsteel-trailer',
'ext': 'mov',
'duration': 148,
'title': 'Trailer',
'upload_date': '20121212',
'uploader_id': 'wb',
2013-08-28 00:18:44 +00:00
},
},
{
2015-10-05 14:21:53 +00:00
'md5': '5fe08795b943eb2e757fa95cb6def1cb',
'info_dict': {
'id': 'manofsteel-teaser',
'ext': 'mov',
'duration': 93,
'title': 'Teaser',
'upload_date': '20120721',
'uploader_id': 'wb',
2013-08-28 00:18:44 +00:00
},
2014-03-27 20:10:51 +00:00
},
2013-08-28 00:18:44 +00:00
]
}, {
'url': 'http://trailers.apple.com/trailers/magnolia/blackthorn/',
'info_dict': {
2017-04-27 08:04:21 +00:00
'id': '4489',
'title': 'Blackthorn',
},
'playlist_mincount': 2,
'expected_warnings': ['Unable to download JSON metadata'],
}, {
# json data only available from http://trailers.apple.com/trailers/feeds/data/15881.json
'url': 'http://trailers.apple.com/trailers/fox/kungfupanda3/',
'info_dict': {
'id': '15881',
'title': 'Kung Fu Panda 3',
},
'playlist_mincount': 4,
2015-02-22 13:58:39 +00:00
}, {
'url': 'http://trailers.apple.com/ca/metropole/autrui/',
'only_matching': True,
}, {
'url': 'http://movietrailers.apple.com/trailers/focus_features/kuboandthetwostrings/',
'only_matching': True,
2015-02-22 13:58:39 +00:00
}]
2013-08-28 00:18:44 +00:00
_JSON_RE = r'iTunes.playURL\((.*?)\);'
2013-08-28 00:18:44 +00:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
movie = mobj.group('movie')
uploader_id = mobj.group('company')
webpage = self._download_webpage(url, movie)
film_id = self._search_regex(r"FilmId\s*=\s*'(\d+)'", webpage, 'film id')
film_data = self._download_json(
'http://trailers.apple.com/trailers/feeds/data/%s.json' % film_id,
film_id, fatal=False)
if film_data:
entries = []
for clip in film_data.get('clips', []):
clip_title = clip['title']
formats = []
for version, version_data in clip.get('versions', {}).items():
for size, size_data in version_data.get('sizes', {}).items():
src = size_data.get('src')
if not src:
continue
formats.append({
'format_id': '%s-%s' % (version, size),
2017-10-09 16:50:53 +00:00
'url': re.sub(r'_(\d+p\.mov)', r'_h\1', src),
'width': int_or_none(size_data.get('width')),
'height': int_or_none(size_data.get('height')),
'language': version[:2],
})
self._sort_formats(formats)
entries.append({
'id': movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', clip_title).lower(),
'formats': formats,
'title': clip_title,
'thumbnail': clip.get('screen') or clip.get('thumb'),
'duration': parse_duration(clip.get('runtime') or clip.get('faded')),
'upload_date': unified_strdate(clip.get('posted')),
'uploader_id': uploader_id,
})
page_data = film_data.get('page', {})
return self.playlist_result(entries, film_id, page_data.get('movie_title'))
2014-03-27 20:10:51 +00:00
playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
2014-11-23 19:41:03 +00:00
2013-12-10 20:03:53 +00:00
def fix_html(s):
2014-03-27 20:10:51 +00:00
s = re.sub(r'(?s)<script[^<]*?>.*?</script>', '', s)
s = re.sub(r'<img ([^<]*?)/?>', r'<img \1/>', s)
2013-12-10 20:03:53 +00:00
# The ' in the onClick attributes are not escaped, it couldn't be parsed
# like: http://trailers.apple.com/trailers/wb/gravity/
2014-11-23 19:41:03 +00:00
2013-12-10 20:03:53 +00:00
def _clean_json(m):
2014-03-27 20:10:51 +00:00
return 'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
2013-12-10 20:03:53 +00:00
s = re.sub(self._JSON_RE, _clean_json, s)
2014-11-26 11:41:24 +00:00
s = '<html>%s</html>' % s
2013-12-10 20:03:53 +00:00
return s
doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
2013-08-28 00:18:44 +00:00
playlist = []
for li in doc.findall('./div/ul/li'):
on_click = li.find('.//a').attrib['onClick']
trailer_info_json = self._search_regex(self._JSON_RE,
2014-11-23 20:39:15 +00:00
on_click, 'trailer info')
trailer_info = json.loads(trailer_info_json)
first_url = trailer_info.get('url')
if not first_url:
continue
title = trailer_info['title']
2013-08-28 00:18:44 +00:00
video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
thumbnail = li.find('.//img').attrib['src']
upload_date = trailer_info['posted'].replace('-', '')
2013-08-28 00:18:44 +00:00
runtime = trailer_info['runtime']
m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
2013-08-28 00:18:44 +00:00
duration = None
if m:
duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
2014-03-27 20:10:51 +00:00
settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')
2013-08-28 00:18:44 +00:00
formats = []
for format in settings['metadata']['sizes']:
# The src is a file pointing to the real video file
2017-10-09 16:50:53 +00:00
format_url = re.sub(r'_(\d*p\.mov)', r'_h\1', format['src'])
formats.append({
'url': format_url,
'format': format['type'],
'width': int_or_none(format['width']),
'height': int_or_none(format['height']),
})
self._sort_formats(formats)
2013-08-28 00:18:44 +00:00
playlist.append({
2013-08-28 00:18:44 +00:00
'_type': 'video',
'id': video_id,
'formats': formats,
'title': title,
'duration': duration,
'thumbnail': thumbnail,
'upload_date': upload_date,
'uploader_id': uploader_id,
'http_headers': {
'User-Agent': 'QuickTime compatible (youtube-dl)',
},
})
2013-08-28 00:18:44 +00:00
return {
'_type': 'playlist',
'id': movie,
'entries': playlist,
}
class AppleTrailersSectionIE(InfoExtractor):
IE_NAME = 'appletrailers:section'
_SECTIONS = {
'justadded': {
'feed_path': 'just_added',
'title': 'Just Added',
},
'exclusive': {
'feed_path': 'exclusive',
'title': 'Exclusive',
},
'justhd': {
'feed_path': 'just_hd',
'title': 'Just HD',
},
'mostpopular': {
'feed_path': 'most_pop',
'title': 'Most Popular',
},
'moviestudios': {
'feed_path': 'studios',
'title': 'Movie Studios',
},
}
_VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/#section=(?P<id>%s)' % '|'.join(_SECTIONS)
_TESTS = [{
'url': 'http://trailers.apple.com/#section=justadded',
'info_dict': {
'title': 'Just Added',
'id': 'justadded',
},
'playlist_mincount': 80,
}, {
'url': 'http://trailers.apple.com/#section=exclusive',
'info_dict': {
'title': 'Exclusive',
'id': 'exclusive',
},
'playlist_mincount': 80,
}, {
'url': 'http://trailers.apple.com/#section=justhd',
'info_dict': {
'title': 'Just HD',
'id': 'justhd',
},
'playlist_mincount': 80,
}, {
'url': 'http://trailers.apple.com/#section=mostpopular',
'info_dict': {
'title': 'Most Popular',
'id': 'mostpopular',
},
2017-04-27 08:04:21 +00:00
'playlist_mincount': 30,
}, {
'url': 'http://trailers.apple.com/#section=moviestudios',
'info_dict': {
'title': 'Movie Studios',
'id': 'moviestudios',
},
'playlist_mincount': 80,
}]
def _real_extract(self, url):
section = self._match_id(url)
section_data = self._download_json(
'http://trailers.apple.com/trailers/home/feeds/%s.json' % self._SECTIONS[section]['feed_path'],
section)
entries = [
self.url_result('http://trailers.apple.com' + e['location'])
for e in section_data]
return self.playlist_result(entries, section, self._SECTIONS[section]['title'])