[la7] Fix extraction (closes #23323)

Parsing `videoParams' or `videoLa7' JavaScript snippets as JSON is getting too
hard for js_to_json.  Just extract the `vid' from there and use _search_og_*
for all other data.

Remove the 2nd test: unfortunately the page are removed after a couple of
days/weeks.

Thanks to Elia Geretto for spotting and suggesting several problems with the
tests!

Closes #23323.
This commit is contained in:
Leonardo Taccari 2020-01-05 17:29:51 +01:00
parent f5863a3ea0
commit 8067a2b8b4
1 changed files with 8 additions and 19 deletions

View File

@ -3,7 +3,6 @@ from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
js_to_json,
smuggle_url, smuggle_url,
) )
@ -23,21 +22,12 @@ class LA7IE(InfoExtractor):
'id': '0_42j6wd36', 'id': '0_42j6wd36',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Inc.Cool8', 'title': 'Inc.Cool8',
'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto atletico', 'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto atletico',
'thumbnail': 're:^https?://.*', 'thumbnail': 're:^https?://.*',
'uploader_id': 'kdla7pillole@iltrovatore.it', 'uploader_id': 'kdla7pillole@iltrovatore.it',
'timestamp': 1443814869, 'timestamp': 1443814869,
'upload_date': '20151002', 'upload_date': '20151002',
}, },
}, {
# 'src' is a dictionary
'url': 'http://tg.la7.it/repliche-tgla7?id=189080',
'md5': '6b0d8888d286e39870208dfeceaf456b',
'info_dict': {
'id': '189080',
'ext': 'mp4',
'title': 'TG LA7',
},
}, { }, {
'url': 'http://www.la7.it/omnibus/rivedila7/omnibus-news-02-07-2016-189077', 'url': 'http://www.la7.it/omnibus/rivedila7/omnibus-news-02-07-2016-189077',
'only_matching': True, 'only_matching': True,
@ -48,20 +38,19 @@ class LA7IE(InfoExtractor):
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
player_data = self._parse_json( player_data = self._search_regex(
self._search_regex( [r'(?s)videoParams\s*=\s*({.+?});', r'videoLa7\(({[^;]+})\);'],
[r'(?s)videoParams\s*=\s*({.+?});', r'videoLa7\(({[^;]+})\);'], webpage, 'player data')
webpage, 'player data'), vid = self._search_regex(r'vid\s*:\s*"(.+?)",', player_data, 'vid')
video_id, transform_source=js_to_json)
return { return {
'_type': 'url_transparent', '_type': 'url_transparent',
'url': smuggle_url('kaltura:103:%s' % player_data['vid'], { 'url': smuggle_url('kaltura:103:%s' % vid, {
'service_url': 'http://nkdam.iltrovatore.it', 'service_url': 'http://nkdam.iltrovatore.it',
}), }),
'id': video_id, 'id': video_id,
'title': player_data['title'], 'title': self._og_search_title(webpage, default=None),
'description': self._og_search_description(webpage, default=None), 'description': self._og_search_description(webpage, default=None),
'thumbnail': player_data.get('poster'), 'thumbnail': self._og_search_thumbnail(webpage, default=None),
'ie_key': 'Kaltura', 'ie_key': 'Kaltura',
} }