[ndtv] Fix title extraction and modernize

pull/3073/head
Sergey M․ 2014-06-10 19:37:38 +07:00
parent 9706f3f802
commit 3141feb73b
1 changed files with 28 additions and 18 deletions

View File

@ -1,22 +1,28 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import month_by_name
from ..utils import (
month_by_name,
int_or_none,
)
class NDTVIE(InfoExtractor):
_VALID_URL = r'^https?://(?:www\.)?ndtv\.com/video/player/[^/]*/[^/]*/(?P<id>[a-z0-9]+)'
_TEST = {
u"url": u"http://www.ndtv.com/video/player/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal/300710",
u"file": u"300710.mp4",
u"md5": u"39f992dbe5fb531c395d8bbedb1e5e88",
u"info_dict": {
u"title": u"NDTV exclusive: Don't need character certificate from Rahul Gandhi, says Arvind Kejriwal",
u"description": u"In an exclusive interview to NDTV, Aam Aadmi Party's Arvind Kejriwal says it makes no difference to him that Rahul Gandhi said the Congress needs to learn from his party.",
u"upload_date": u"20131208",
u"duration": 1327,
u"thumbnail": u"http://i.ndtvimg.com/video/images/vod/medium/2013-12/big_300710_1386518307.jpg",
'url': 'http://www.ndtv.com/video/player/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal/300710',
'md5': '39f992dbe5fb531c395d8bbedb1e5e88',
'info_dict': {
'id': '300710',
'ext': 'mp4',
'title': "NDTV exclusive: Don't need character certificate from Rahul Gandhi, says Arvind Kejriwal",
'description': 'md5:ab2d4b4a6056c5cb4caa6d729deabf02',
'upload_date': '20131208',
'duration': 1327,
'thumbnail': 'http://i.ndtvimg.com/video/images/vod/medium/2013-12/big_300710_1386518307.jpg',
},
}
@ -27,13 +33,12 @@ class NDTVIE(InfoExtractor):
webpage = self._download_webpage(url, video_id)
filename = self._search_regex(
r"__filename='([^']+)'", webpage, u'video filename')
video_url = (u'http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' %
r"__filename='([^']+)'", webpage, 'video filename')
video_url = ('http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' %
filename)
duration_str = filename = self._search_regex(
r"__duration='([^']+)'", webpage, u'duration', fatal=False)
duration = None if duration_str is None else int(duration_str)
duration = int_or_none(self._search_regex(
r"__duration='([^']+)'", webpage, 'duration', fatal=False))
date_m = re.search(r'''(?x)
<p\s+class="vod_dateline">\s*
@ -41,7 +46,7 @@ class NDTVIE(InfoExtractor):
(?P<monthname>[A-Za-z]+)\s+(?P<day>[0-9]+),\s*(?P<year>[0-9]+)
''', webpage)
upload_date = None
assert date_m
if date_m is not None:
month = month_by_name(date_m.group('monthname'))
if month is not None:
@ -49,14 +54,19 @@ class NDTVIE(InfoExtractor):
date_m.group('year'), month, int(date_m.group('day')))
description = self._og_search_description(webpage)
READ_MORE = u' (Read more)'
READ_MORE = ' (Read more)'
if description.endswith(READ_MORE):
description = description[:-len(READ_MORE)]
title = self._og_search_title(webpage)
TITLE_SUFFIX = ' - NDTV'
if title.endswith(TITLE_SUFFIX):
title = title[:-len(TITLE_SUFFIX)]
return {
'id': video_id,
'url': video_url,
'title': self._og_search_title(webpage),
'title': title,
'description': description,
'thumbnail': self._og_search_thumbnail(webpage),
'duration': duration,