yt-dlc/youtube_dl/extractor/adobetv.py

120 lines
4.2 KiB
Python
Raw Normal View History

2014-12-22 17:05:47 +01:00
from __future__ import unicode_literals
import re
2014-12-22 17:05:47 +01:00
from .common import InfoExtractor
from ..utils import (
parse_duration,
unified_strdate,
str_to_int,
int_or_none,
float_or_none,
ISO639Utils,
2014-12-22 17:05:47 +01:00
)
class AdobeTVIE(InfoExtractor):
_VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?watch/(?P<show_urlname>[^/]+)/(?P<id>[^/]+)'
2014-12-22 17:05:47 +01:00
_TEST = {
'url': 'http://tv.adobe.com/watch/the-complete-picture-with-julieanne-kost/quick-tip-how-to-draw-a-circle-around-an-object-in-photoshop/',
'md5': '9bc5727bcdd55251f35ad311ca74fa1e',
'info_dict': {
'id': '10981',
2014-12-22 17:05:47 +01:00
'ext': 'mp4',
'title': 'Quick Tip - How to Draw a Circle Around an Object in Photoshop',
'description': 'md5:99ec318dc909d7ba2a1f2b038f7d2311',
'thumbnail': 're:https?://.*\.jpg$',
'upload_date': '20110914',
'duration': 60,
'view_count': int,
},
}
def _real_extract(self, url):
language, show_urlname, urlname = re.match(self._VALID_URL, url).groups()
if not language:
language = 'en'
2014-12-22 17:05:47 +01:00
video_data = self._download_json(
'http://tv.adobe.com/api/v4/episode/get/?language=%s&show_urlname=%s&urlname=%s&disclosure=standard' % (language, show_urlname, urlname),
urlname)['data'][0]
2014-12-22 17:05:47 +01:00
formats = [{
'url': source['url'],
'format_id': source.get('quality_level') or source['url'].split('-')[-1].split('.')[0] or None,
'width': int_or_none(source.get('width')),
'height': int_or_none(source.get('height')),
'tbr': int_or_none(source.get('video_data_rate')),
} for source in video_data['videos']]
2014-12-22 17:05:47 +01:00
self._sort_formats(formats)
return {
'id': str(video_data['id']),
'title': video_data['title'],
'description': video_data.get('description'),
'thumbnail': video_data.get('thumbnail'),
'upload_date': unified_strdate(video_data.get('start_date')),
'duration': parse_duration(video_data.get('duration')),
'view_count': str_to_int(video_data.get('playcount')),
2014-12-22 17:05:47 +01:00
'formats': formats,
}
class AdobeTVVideoIE(InfoExtractor):
_VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
_TEST = {
# From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
'url': 'https://video.tv.adobe.com/v/2456/',
'md5': '43662b577c018ad707a63766462b1e87',
'info_dict': {
'id': '2456',
'ext': 'mp4',
'title': 'New experience with Acrobat DC',
'description': 'New experience with Acrobat DC',
'duration': 248.667,
},
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
player_params = self._parse_json(self._search_regex(
r'var\s+bridge\s*=\s*([^;]+);', webpage, 'player parameters'),
video_id)
formats = [{
'url': source['src'],
'width': source.get('width'),
'height': source.get('height'),
'tbr': source.get('bitrate'),
} for source in player_params['sources']]
# For both metadata and downloaded files the duration varies among
# formats. I just pick the max one
duration = max(filter(None, [
float_or_none(source.get('duration'), scale=1000)
for source in player_params['sources']]))
subtitles = {}
for translation in player_params.get('translations', []):
lang_id = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
if lang_id not in subtitles:
subtitles[lang_id] = []
subtitles[lang_id].append({
'url': translation['vttPath'],
'ext': 'vtt',
})
return {
'id': video_id,
'formats': formats,
'title': player_params['title'],
'description': self._og_search_description(webpage),
'duration': duration,
'subtitles': subtitles,
}