[zype] improve extraction

- extract subtitles(closes #21258)
- support URLs with alternative keys/tokens(#21258)
- extract more metadata
This commit is contained in:
Remita Amine 2020-01-15 14:02:57 +01:00
parent 90ea83c64d
commit a9866c0366
3 changed files with 101 additions and 62 deletions

View file

@ -2098,6 +2098,9 @@ class GenericIE(InfoExtractor):
'ext': 'mp4', 'ext': 'mp4',
'title': 'Smoky Barbecue Favorites', 'title': 'Smoky Barbecue Favorites',
'thumbnail': r're:^https?://.*\.jpe?g', 'thumbnail': r're:^https?://.*\.jpe?g',
'description': 'md5:5ff01e76316bd8d46508af26dc86023b',
'upload_date': '20170909',
'timestamp': 1504915200,
}, },
'add_ie': [ZypeIE.ie_key()], 'add_ie': [ZypeIE.ie_key()],
'params': { 'params': {

View file

@ -1,21 +1,12 @@
from __future__ import unicode_literals from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import (
dict_get,
float_or_none,
int_or_none,
unified_timestamp,
update_url_query,
url_or_none,
)
class TruNewsIE(InfoExtractor): class TruNewsIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?trunews\.com/stream/(?P<id>[^/?#&]+)' _VALID_URL = r'https?://(?:www\.)?trunews\.com/stream/(?P<id>[^/?#&]+)'
_TEST = { _TEST = {
'url': 'https://www.trunews.com/stream/will-democrats-stage-a-circus-during-president-trump-s-state-of-the-union-speech', 'url': 'https://www.trunews.com/stream/will-democrats-stage-a-circus-during-president-trump-s-state-of-the-union-speech',
'md5': 'a19c024c3906ff954fac9b96ce66bb08',
'info_dict': { 'info_dict': {
'id': '5c5a21e65d3c196e1c0020cc', 'id': '5c5a21e65d3c196e1c0020cc',
'display_id': 'will-democrats-stage-a-circus-during-president-trump-s-state-of-the-union-speech', 'display_id': 'will-democrats-stage-a-circus-during-president-trump-s-state-of-the-union-speech',
@ -28,48 +19,16 @@ class TruNewsIE(InfoExtractor):
}, },
'add_ie': ['Zype'], 'add_ie': ['Zype'],
} }
_ZYPE_TEMPL = 'https://player.zype.com/embed/%s.js?api_key=X5XnahkjCwJrT_l5zUqypnaLEObotyvtUKJWWlONxDoHVjP8vqxlArLV8llxMbyt'
def _real_extract(self, url): def _real_extract(self, url):
display_id = self._match_id(url) display_id = self._match_id(url)
video = self._download_json( zype_id = self._download_json(
'https://api.zype.com/videos', display_id, query={ 'https://api.zype.com/videos', display_id, query={
'app_key': 'PUVKp9WgGUb3-JUw6EqafLx8tFVP6VKZTWbUOR-HOm__g4fNDt1bCsm_LgYf_k9H', 'app_key': 'PUVKp9WgGUb3-JUw6EqafLx8tFVP6VKZTWbUOR-HOm__g4fNDt1bCsm_LgYf_k9H',
'per_page': 1, 'per_page': 1,
'active': 'true', 'active': 'true',
'friendly_title': display_id, 'friendly_title': display_id,
})['response'][0] })['response'][0]['_id']
return self.url_result(self._ZYPE_TEMPL % zype_id, 'Zype', zype_id)
zype_id = video['_id']
thumbnails = []
thumbnails_list = video.get('thumbnails')
if isinstance(thumbnails_list, list):
for thumbnail in thumbnails_list:
if not isinstance(thumbnail, dict):
continue
thumbnail_url = url_or_none(thumbnail.get('url'))
if not thumbnail_url:
continue
thumbnails.append({
'url': thumbnail_url,
'width': int_or_none(thumbnail.get('width')),
'height': int_or_none(thumbnail.get('height')),
})
return {
'_type': 'url_transparent',
'url': update_url_query(
'https://player.zype.com/embed/%s.js' % zype_id,
{'api_key': 'X5XnahkjCwJrT_l5zUqypnaLEObotyvtUKJWWlONxDoHVjP8vqxlArLV8llxMbyt'}),
'ie_key': 'Zype',
'id': zype_id,
'display_id': display_id,
'title': video.get('title'),
'description': dict_get(video, ('description', 'ott_description', 'short_description')),
'duration': int_or_none(video.get('duration')),
'timestamp': unified_timestamp(video.get('published_at')),
'average_rating': float_or_none(video.get('rating')),
'view_count': int_or_none(video.get('request_count')),
'thumbnails': thumbnails,
}

View file

@ -4,10 +4,20 @@ from __future__ import unicode_literals
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_HTTPError
from ..utils import (
dict_get,
ExtractorError,
int_or_none,
js_to_json,
parse_iso8601,
)
class ZypeIE(InfoExtractor): class ZypeIE(InfoExtractor):
_VALID_URL = r'https?://player\.zype\.com/embed/(?P<id>[\da-fA-F]+)\.js\?.*?api_key=[^&]+' _ID_RE = r'[\da-fA-F]+'
_COMMON_RE = r'//player\.zype\.com/embed/%s\.(?:js|json|html)\?.*?(?:access_token|(?:ap[ip]|player)_key)='
_VALID_URL = r'https?:%s[^&]+' % (_COMMON_RE % ('(?P<id>%s)' % _ID_RE))
_TEST = { _TEST = {
'url': 'https://player.zype.com/embed/5b400b834b32992a310622b9.js?api_key=jZ9GUhRmxcPvX7M3SlfejB6Hle9jyHTdk2jVxG7wOHPLODgncEKVdPYBhuz9iWXQ&autoplay=false&controls=true&da=false', 'url': 'https://player.zype.com/embed/5b400b834b32992a310622b9.js?api_key=jZ9GUhRmxcPvX7M3SlfejB6Hle9jyHTdk2jVxG7wOHPLODgncEKVdPYBhuz9iWXQ&autoplay=false&controls=true&da=false',
'md5': 'eaee31d474c76a955bdaba02a505c595', 'md5': 'eaee31d474c76a955bdaba02a505c595',
@ -16,6 +26,9 @@ class ZypeIE(InfoExtractor):
'ext': 'mp4', 'ext': 'mp4',
'title': 'Smoky Barbecue Favorites', 'title': 'Smoky Barbecue Favorites',
'thumbnail': r're:^https?://.*\.jpe?g', 'thumbnail': r're:^https?://.*\.jpe?g',
'description': 'md5:5ff01e76316bd8d46508af26dc86023b',
'timestamp': 1504915200,
'upload_date': '20170909',
}, },
} }
@ -24,34 +37,98 @@ class ZypeIE(InfoExtractor):
return [ return [
mobj.group('url') mobj.group('url')
for mobj in re.finditer( for mobj in re.finditer(
r'<script[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//player\.zype\.com/embed/[\da-fA-F]+\.js\?.*?api_key=.+?)\1', r'<script[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?%s.+?)\1' % (ZypeIE._COMMON_RE % ZypeIE._ID_RE),
webpage)] webpage)]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) try:
response = self._download_json(re.sub(
r'\.(?:js|html)\?', '.json?', url), video_id)['response']
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 401, 403):
raise ExtractorError(self._parse_json(
e.cause.read().decode(), video_id)['message'], expected=True)
raise
title = self._search_regex( body = response['body']
r'video_title\s*[:=]\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage, video = response['video']
'title', group='value') title = video['title']
m3u8_url = self._search_regex(
r'(["\'])(?P<url>(?:(?!\1).)+\.m3u8(?:(?!\1).)*)\1', webpage,
'm3u8 url', group='url')
if isinstance(body, dict):
formats = []
for output in body.get('outputs', []):
output_url = output.get('url')
if not output_url:
continue
name = output.get('name')
if name == 'm3u8':
formats = self._extract_m3u8_formats( formats = self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', output_url, video_id, 'mp4',
m3u8_id='hls') 'm3u8_native', m3u8_id='hls', fatal=False)
else:
f = {
'format_id': name,
'tbr': int_or_none(output.get('bitrate')),
'url': output_url,
}
if name in ('m4a', 'mp3'):
f['vcodec'] = 'none'
else:
f.update({
'height': int_or_none(output.get('height')),
'width': int_or_none(output.get('width')),
})
formats.append(f)
text_tracks = body.get('subtitles') or []
else:
m3u8_url = self._search_regex(
r'(["\'])(?P<url>(?:(?!\1).)+\.m3u8(?:(?!\1).)*)\1',
body, 'm3u8 url', group='url')
formats = self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
text_tracks = self._search_regex(
r'textTracks\s*:\s*(\[[^]]+\])',
body, 'text tracks', default=None)
if text_tracks:
text_tracks = self._parse_json(
text_tracks, video_id, js_to_json, False)
self._sort_formats(formats) self._sort_formats(formats)
thumbnail = self._search_regex( subtitles = {}
r'poster\s*[:=]\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage, 'thumbnail', if text_tracks:
default=False, group='url') for text_track in text_tracks:
tt_url = dict_get(text_track, ('file', 'src'))
if not tt_url:
continue
subtitles.setdefault(text_track.get('label') or 'English', []).append({
'url': tt_url,
})
thumbnails = []
for thumbnail in video.get('thumbnails', []):
thumbnail_url = thumbnail.get('url')
if not thumbnail_url:
continue
thumbnails.append({
'url': thumbnail_url,
'width': int_or_none(thumbnail.get('width')),
'height': int_or_none(thumbnail.get('height')),
})
return { return {
'id': video_id, 'id': video_id,
'display_id': video.get('friendly_title'),
'title': title, 'title': title,
'thumbnail': thumbnail, 'thumbnails': thumbnails,
'description': dict_get(video, ('description', 'ott_description', 'short_description')),
'timestamp': parse_iso8601(video.get('published_at')),
'duration': int_or_none(video.get('duration')),
'view_count': int_or_none(video.get('request_count')),
'average_rating': int_or_none(video.get('rating')),
'season_number': int_or_none(video.get('season')),
'episode_number': int_or_none(video.get('episode')),
'formats': formats, 'formats': formats,
'subtitles': subtitles,
} }