[yt-dlp.git] / youtube_dl / extractor / appletrailers.py

from __future__ import unicode_literals

import re
import json

from .common import InfoExtractor
from ..compat import compat_urlparse
from ..utils import (
    int_or_none,
)


class AppleTrailersIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
    _TEST = {
        "url": "http://trailers.apple.com/trailers/wb/manofsteel/",
        "playlist": [
            {
                "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
                "info_dict": {
                    "id": "manofsteel-trailer4",
                    "ext": "mov",
                    "duration": 111,
                    "title": "Trailer 4",
                    "upload_date": "20130523",
                    "uploader_id": "wb",
                },
            },
            {
                "md5": "b8017b7131b721fb4e8d6f49e1df908c",
                "info_dict": {
                    "id": "manofsteel-trailer3",
                    "ext": "mov",
                    "duration": 182,
                    "title": "Trailer 3",
                    "upload_date": "20130417",
                    "uploader_id": "wb",
                },
            },
            {
                "md5": "d0f1e1150989b9924679b441f3404d48",
                "info_dict": {
                    "id": "manofsteel-trailer",
                    "ext": "mov",
                    "duration": 148,
                    "title": "Trailer",
                    "upload_date": "20121212",
                    "uploader_id": "wb",
                },
            },
            {
                "md5": "5fe08795b943eb2e757fa95cb6def1cb",
                "info_dict": {
                    "id": "manofsteel-teaser",
                    "ext": "mov",
                    "duration": 93,
                    "title": "Teaser",
                    "upload_date": "20120721",
                    "uploader_id": "wb",
                },
            },
        ]
    }

    _JSON_RE = r'iTunes.playURL\((.*?)\);'

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        movie = mobj.group('movie')
        uploader_id = mobj.group('company')

        playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')

        def fix_html(s):
            s = re.sub(r'(?s)<script[^<]*?>.*?</script>', '', s)
            s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s)
            # The ' in the onClick attributes are not escaped, it couldn't be parsed
            # like: http://trailers.apple.com/trailers/wb/gravity/

            def _clean_json(m):
                return 'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
            s = re.sub(self._JSON_RE, _clean_json, s)
            s = '<html>%s</html>' % s
            return s
        doc = self._download_xml(playlist_url, movie, transform_source=fix_html)

        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,
                                                   on_click, 'trailer info')
            trailer_info = json.loads(trailer_info_json)
            title = trailer_info['title']
            video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
            thumbnail = li.find('.//img').attrib['src']
            upload_date = trailer_info['posted'].replace('-', '')

            runtime = trailer_info['runtime']
            m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
            duration = None
            if m:
                duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))

            first_url = trailer_info['url']
            trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
            settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
            settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')

            formats = []
            for format in settings['metadata']['sizes']:
                # The src is a file pointing to the real video file
                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)

            playlist.append({
                '_type': 'video',
                'id': video_id,
                'title': title,
                'formats': formats,
                'title': title,
                'duration': duration,
                'thumbnail': thumbnail,
                'upload_date': upload_date,
                'uploader_id': uploader_id,
                'http_headers': {
                    'User-Agent': 'QuickTime compatible (youtube-dl)',
                },
            })

        return {
            '_type': 'playlist',
            'id': movie,
            'entries': playlist,
        }
Commit	Line	Data
3798eadc PH	1	from __future__ import unicode_literals
3798eadc PH	2
44586389	3	import re
84353056	4	import json
44586389 PH	5
44586389 PH	6	from .common import InfoExtractor
1cc79574	7	from ..compat import compat_urlparse
44586389	8	from ..utils import (
9572013d	9	int_or_none,
44586389 PH	10	)
	11
	12
	13	class AppleTrailersIE(InfoExtractor):
c0ade33e	14	_VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
44586389	15	_TEST = {
3798eadc PH	16	"url": "http://trailers.apple.com/trailers/wb/manofsteel/",
3798eadc PH	17	"playlist": [
44586389	18	{
3798eadc PH	19	"md5": "d97a8e575432dbcb81b7c3acb741f8a8",
3798eadc PH	20	"info_dict": {
28acf550 JMF	21	"id": "manofsteel-trailer4",
28acf550 JMF	22	"ext": "mov",
3798eadc PH	23	"duration": 111,
	24	"title": "Trailer 4",
	25	"upload_date": "20130523",
	26	"uploader_id": "wb",
44586389 PH	27	},
	28	},
	29	{
3798eadc PH	30	"md5": "b8017b7131b721fb4e8d6f49e1df908c",
3798eadc PH	31	"info_dict": {
28acf550 JMF	32	"id": "manofsteel-trailer3",
28acf550 JMF	33	"ext": "mov",
3798eadc PH	34	"duration": 182,
	35	"title": "Trailer 3",
	36	"upload_date": "20130417",
	37	"uploader_id": "wb",
44586389 PH	38	},
	39	},
	40	{
3798eadc PH	41	"md5": "d0f1e1150989b9924679b441f3404d48",
3798eadc PH	42	"info_dict": {
28acf550 JMF	43	"id": "manofsteel-trailer",
28acf550 JMF	44	"ext": "mov",
3798eadc PH	45	"duration": 148,
	46	"title": "Trailer",
	47	"upload_date": "20121212",
	48	"uploader_id": "wb",
44586389 PH	49	},
	50	},
	51	{
3798eadc PH	52	"md5": "5fe08795b943eb2e757fa95cb6def1cb",
3798eadc PH	53	"info_dict": {
28acf550 JMF	54	"id": "manofsteel-teaser",
28acf550 JMF	55	"ext": "mov",
3798eadc PH	56	"duration": 93,
	57	"title": "Teaser",
	58	"upload_date": "20120721",
	59	"uploader_id": "wb",
44586389	60	},
28acf550	61	},
44586389 PH	62	]
	63	}
	64
84353056 JMF	65	_JSON_RE = r'iTunes.playURL\((.*?)\);'
84353056 JMF	66
44586389 PH	67	def _real_extract(self, url):
	68	mobj = re.match(self._VALID_URL, url)
	69	movie = mobj.group('movie')
	70	uploader_id = mobj.group('company')
	71
28acf550	72	playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
5f6a1245	73
18258362	74	def fix_html(s):
28acf550	75	s = re.sub(r'(?s)<script[^<]?>.?</script>', '', s)
18258362 JMF	76	s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s)
	77	# The ' in the onClick attributes are not escaped, it couldn't be parsed
	78	# like: http://trailers.apple.com/trailers/wb/gravity/
5f6a1245	79
18258362	80	def _clean_json(m):
28acf550	81	return 'iTunes.playURL(%s);' % m.group(1).replace('\'', ''')
18258362	82	s = re.sub(self._JSON_RE, _clean_json, s)
e91cdcae	83	s = '<html>%s</html>' % s
18258362 JMF	84	return s
18258362 JMF	85	doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
44586389	86
44586389 PH	87	playlist = []
44586389 PH	88	for li in doc.findall('./div/ul/li'):
84353056 JMF	89	on_click = li.find('.//a').attrib['onClick']
84353056 JMF	90	trailer_info_json = self._search_regex(self._JSON_RE,
9e1a5b84	91	on_click, 'trailer info')
84353056 JMF	92	trailer_info = json.loads(trailer_info_json)
84353056 JMF	93	title = trailer_info['title']
44586389 PH	94	video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
44586389 PH	95	thumbnail = li.find('.//img').attrib['src']
84353056	96	upload_date = trailer_info['posted'].replace('-', '')
44586389	97
84353056 JMF	98	runtime = trailer_info['runtime']
84353056 JMF	99	m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
44586389 PH	100	duration = None
	101	if m:
	102	duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
	103
84353056	104	first_url = trailer_info['url']
bb4aa62c	105	trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
84353056	106	settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
28acf550	107	settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')
44586389	108
84353056 JMF	109	formats = []
	110	for format in settings['metadata']['sizes']:
	111	# The src is a file pointing to the real video file
	112	format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
	113	formats.append({
	114	'url': format_url,
84353056	115	'format': format['type'],
9572013d PH	116	'width': int_or_none(format['width']),
9572013d PH	117	'height': int_or_none(format['height']),
84353056	118	})
7b8af563 PH	119
7b8af563 PH	120	self._sort_formats(formats)
44586389	121
fb7abb31	122	playlist.append({
44586389 PH	123	'_type': 'video',
	124	'id': video_id,
	125	'title': title,
	126	'formats': formats,
	127	'title': title,
	128	'duration': duration,
	129	'thumbnail': thumbnail,
	130	'upload_date': upload_date,
	131	'uploader_id': uploader_id,
e1554a40 JMF	132	'http_headers': {
	133	'User-Agent': 'QuickTime compatible (youtube-dl)',
	134	},
fb7abb31	135	})
44586389 PH	136
	137	return {
	138	'_type': 'playlist',
	139	'id': movie,
	140	'entries': playlist,
	141	}