[appletrailers] extract info from an alternative source if available(closes #8422)(closes #8422)

This commit is contained in:
Remita Amine 2016-06-23 15:49:42 +01:00
parent 6e3c2047f8
commit 494172d2e5

View File

@ -7,6 +7,8 @@ from .common import InfoExtractor
from ..compat import compat_urlparse
from ..utils import (
int_or_none,
parse_duration,
unified_strdate,
)
@ -16,7 +18,8 @@ class AppleTrailersIE(InfoExtractor):
_TESTS = [{
'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
'info_dict': {
'id': 'manofsteel',
'id': '5111',
'title': 'Man of Steel',
},
'playlist': [
{
@ -70,6 +73,15 @@ class AppleTrailersIE(InfoExtractor):
'id': '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,
}, {
'url': 'http://trailers.apple.com/ca/metropole/autrui/',
'only_matching': True,
@ -85,6 +97,45 @@ class AppleTrailersIE(InfoExtractor):
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),
'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('runtime'),
'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'))
playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
def fix_html(s):