mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-11-27 02:42:30 +00:00
86e5f3ed2e
Using https://github.com/asottile/pyupgrade 1. `__future__` imports and `coding: utf-8` were removed 2. Files were rewritten with `pyupgrade --py36-plus --keep-percent-format` 3. f-strings were cherry-picked from `pyupgrade --py36-plus` Extractors are left untouched (except removing header) to avoid unnecessary merge conflicts
98 lines
3.2 KiB
Python
98 lines
3.2 KiB
Python
import re
|
|
|
|
from .common import InfoExtractor
|
|
from ..utils import (
|
|
determine_ext,
|
|
float_or_none,
|
|
int_or_none,
|
|
parse_age_limit,
|
|
qualities,
|
|
random_birthday,
|
|
unified_timestamp,
|
|
urljoin,
|
|
)
|
|
|
|
|
|
class VideoPressIE(InfoExtractor):
|
|
_ID_REGEX = r'[\da-zA-Z]{8}'
|
|
_PATH_REGEX = r'video(?:\.word)?press\.com/embed/'
|
|
_VALID_URL = r'https?://%s(?P<id>%s)' % (_PATH_REGEX, _ID_REGEX)
|
|
_TESTS = [{
|
|
'url': 'https://videopress.com/embed/kUJmAcSf',
|
|
'md5': '706956a6c875873d51010921310e4bc6',
|
|
'info_dict': {
|
|
'id': 'kUJmAcSf',
|
|
'ext': 'mp4',
|
|
'title': 'VideoPress Demo',
|
|
'thumbnail': r're:^https?://.*\.jpg',
|
|
'duration': 634.6,
|
|
'timestamp': 1434983935,
|
|
'upload_date': '20150622',
|
|
'age_limit': 0,
|
|
},
|
|
}, {
|
|
# 17+, requires birth_* params
|
|
'url': 'https://videopress.com/embed/iH3gstfZ',
|
|
'only_matching': True,
|
|
}, {
|
|
'url': 'https://video.wordpress.com/embed/kUJmAcSf',
|
|
'only_matching': True,
|
|
}]
|
|
|
|
@staticmethod
|
|
def _extract_urls(webpage):
|
|
return re.findall(
|
|
r'<iframe[^>]+src=["\']((?:https?://)?%s%s)' % (VideoPressIE._PATH_REGEX, VideoPressIE._ID_REGEX),
|
|
webpage)
|
|
|
|
def _real_extract(self, url):
|
|
video_id = self._match_id(url)
|
|
|
|
query = random_birthday('birth_year', 'birth_month', 'birth_day')
|
|
query['fields'] = 'description,duration,file_url_base,files,height,original,poster,rating,title,upload_date,width'
|
|
video = self._download_json(
|
|
'https://public-api.wordpress.com/rest/v1.1/videos/%s' % video_id,
|
|
video_id, query=query)
|
|
|
|
title = video['title']
|
|
|
|
file_url_base = video.get('file_url_base') or {}
|
|
base_url = file_url_base.get('https') or file_url_base.get('http')
|
|
|
|
QUALITIES = ('std', 'dvd', 'hd')
|
|
quality = qualities(QUALITIES)
|
|
|
|
formats = []
|
|
for format_id, f in (video.get('files') or {}).items():
|
|
if not isinstance(f, dict):
|
|
continue
|
|
for ext, path in f.items():
|
|
if ext in ('mp4', 'ogg'):
|
|
formats.append({
|
|
'url': urljoin(base_url, path),
|
|
'format_id': '%s-%s' % (format_id, ext),
|
|
'ext': determine_ext(path, ext),
|
|
'quality': quality(format_id),
|
|
})
|
|
original_url = video.get('original')
|
|
if original_url:
|
|
formats.append({
|
|
'url': original_url,
|
|
'format_id': 'original',
|
|
'quality': len(QUALITIES),
|
|
'width': int_or_none(video.get('width')),
|
|
'height': int_or_none(video.get('height')),
|
|
})
|
|
self._sort_formats(formats)
|
|
|
|
return {
|
|
'id': video_id,
|
|
'title': title,
|
|
'description': video.get('description'),
|
|
'thumbnail': video.get('poster'),
|
|
'duration': float_or_none(video.get('duration'), 1000),
|
|
'timestamp': unified_timestamp(video.get('upload_date')),
|
|
'age_limit': parse_age_limit(video.get('rating')),
|
|
'formats': formats,
|
|
}
|