[ndtv] Add support for sub-sites

This commit is contained in:
rawcoder 2017-10-22 07:02:20 +05:30 committed by Sergey M
parent 36e2d3ca43
commit 55c727a547

View File

@ -1,45 +1,106 @@
# coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_unquote_plus
)
from ..utils import ( from ..utils import (
int_or_none, parse_duration,
remove_end, remove_end,
unified_strdate, unified_strdate,
urljoin
) )
class NDTVIE(InfoExtractor): class NDTVIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?ndtv\.com/video/(?:[^/]+/)+[^/?^&]+-(?P<id>\d+)' _VALID_URL = r'https?://(?:[^/]+\.)?ndtv\.com/(?:[^/]+/)*videos?/?(?:[^/]+/)*[^/?^&]+-(?P<id>\d+)'
_TEST = { _TESTS = [
'url': 'http://www.ndtv.com/video/news/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal-300710', {
'md5': '39f992dbe5fb531c395d8bbedb1e5e88', 'url': 'https://khabar.ndtv.com/video/show/prime-time/prime-time-ill-system-and-poor-education-468818',
'md5': '78efcf3880ef3fd9b83d405ca94a38eb',
'info_dict': { 'info_dict': {
'id': '300710', 'id': '468818',
'ext': 'mp4', 'ext': 'mp4',
'title': "NDTV exclusive: Don't need character certificate from Rahul Gandhi, says Arvind Kejriwal", 'title': "प्राइम टाइम: सिस्टम बीमार, स्कूल बदहाल",
'description': 'md5:ab2d4b4a6056c5cb4caa6d729deabf02', 'description': 'md5:f410512f1b49672e5695dea16ef2731d',
'upload_date': '20131208', 'upload_date': '20170928',
'duration': 1327, 'duration': 2218,
'thumbnail': r're:https?://.*\.jpg', 'thumbnail': r're:https?://.*\.jpg',
},
} }
},
{
# __filename is url
'url': 'http://movies.ndtv.com/videos/cracker-free-diwali-wishes-from-karan-johar-kriti-sanon-other-stars-470304',
'md5': 'f1d709352305b44443515ac56b45aa46',
'info_dict': {
'id': '470304',
'ext': 'mp4',
'title': "Cracker-Free Diwali Wishes From Karan Johar, Kriti Sanon & Other Stars",
'description': 'md5:f115bba1adf2f6433fa7c1ade5feb465',
'upload_date': '20171019',
'duration': 137,
'thumbnail': r're:https?://.*\.jpg',
}
},
{
'url': 'https://www.ndtv.com/video/news/news/delhi-s-air-quality-status-report-after-diwali-is-very-poor-470372',
'only_matching': True
},
{
'url': 'https://auto.ndtv.com/videos/the-cnb-daily-october-13-2017-469935',
'only_matching': True
},
{
'url': 'https://sports.ndtv.com/cricket/videos/2nd-t20i-rock-thrown-at-australia-cricket-team-bus-after-win-over-india-469764',
'only_matching': True
},
{
'url': 'http://gadgets.ndtv.com/videos/uncharted-the-lost-legacy-review-465568',
'only_matching': True
},
{
'url': 'http://profit.ndtv.com/videos/news/video-indian-economy-on-very-solid-track-international-monetary-fund-chief-470040',
'only_matching': True
},
{
'url': 'http://food.ndtv.com/video-basil-seeds-coconut-porridge-419083',
'only_matching': True
},
{
'url': 'https://doctor.ndtv.com/videos/top-health-stories-of-the-week-467396',
'only_matching': True
},
{
'url': 'https://swirlster.ndtv.com/video/how-to-make-friends-at-work-469324',
'only_matching': True
}
]
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) webpage = self._download_webpage(url, video_id)
title = remove_end(self._og_search_title(webpage), ' - NDTV') # '__title' does not contain extra words such as sub-site name, "Video" etc.
title = compat_urllib_parse_unquote_plus(
self._search_regex(r"__title\s*=\s*'([^']+)'", webpage, 'title', default=None) or
self._og_search_title(webpage))
filename = self._search_regex( filename = self._search_regex(
r"__filename='([^']+)'", webpage, 'video filename') r"(?:__)?filename\s*[:=]\s*'([^']+)'", webpage, 'video filename')
video_url = 'http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' % filename # in "movies" sub-site pages, filename is URL
video_url = urljoin('https://ndtvod.bc-ssl.cdn.bitgravity.com/23372/ndtv/', filename.lstrip('/'))
duration = int_or_none(self._search_regex( # "doctor" sub-site has MM:SS format
r"__duration='([^']+)'", webpage, 'duration', fatal=False)) duration = parse_duration(self._search_regex(
r"(?:__)?duration\s*[:=]\s*'([^']+)'", webpage, 'duration', fatal=False))
# "sports", "doctor", "swirlster" sub-sites don't have 'publish-date'
upload_date = unified_strdate(self._html_search_meta( upload_date = unified_strdate(self._html_search_meta(
'publish-date', webpage, 'upload date', fatal=False)) 'publish-date', webpage, 'upload date', default=None) or self._html_search_meta(
'uploadDate', webpage, 'upload date', default=None) or self._search_regex(
r'datePublished"\s*:\s*"([^"]+)"', webpage, 'upload date', fatal=False))
description = remove_end(self._og_search_description(webpage), ' (Read more)') description = remove_end(self._og_search_description(webpage), ' (Read more)')