[heise] Fix description, thumbnail and format ID

This commit is contained in:
Philipp Hagemeister 2014-11-04 23:14:16 +01:00
parent a32f253112
commit 711ede6e1b
2 changed files with 21 additions and 14 deletions

View file

@ -404,7 +404,7 @@ class InfoExtractor(object):
video_info['title'] = playlist_title video_info['title'] = playlist_title
return video_info return video_info
def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
""" """
Perform a regex search on the given string, using a single or a list of Perform a regex search on the given string, using a single or a list of
patterns returning the first matching group. patterns returning the first matching group.
@ -425,8 +425,11 @@ class InfoExtractor(object):
_name = name _name = name
if mobj: if mobj:
if group is None:
# return the first matching group # return the first matching group
return next(g for g in mobj.groups() if g is not None) return next(g for g in mobj.groups() if g is not None)
else:
return mobj.group(group)
elif default is not _NO_DEFAULT: elif default is not _NO_DEFAULT:
return default return default
elif fatal: elif fatal:
@ -436,11 +439,11 @@ class InfoExtractor(object):
'please report this issue on http://yt-dl.org/bug' % _name) 'please report this issue on http://yt-dl.org/bug' % _name)
return None return None
def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
""" """
Like _search_regex, but strips HTML tags and unescapes entities. Like _search_regex, but strips HTML tags and unescapes entities.
""" """
res = self._search_regex(pattern, string, name, default, fatal, flags) res = self._search_regex(pattern, string, name, default, fatal, flags, group)
if res: if res:
return clean_html(res).strip() return clean_html(res).strip()
else: else:
@ -534,9 +537,9 @@ class InfoExtractor(object):
display_name = name display_name = name
return self._html_search_regex( return self._html_search_regex(
r'''(?ix)<meta r'''(?ix)<meta
(?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?) (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1)
[^>]+content=["\']([^"\']+)["\']''' % re.escape(name), [^>]+content=(["\'])(?P<content>.*?)\1''' % re.escape(name),
html, display_name, fatal=fatal, **kwargs) html, display_name, fatal=fatal, group='content', **kwargs)
def _dc_search_uploader(self, html): def _dc_search_uploader(self, html):
return self._html_search_meta('dc.creator', html, 'uploader') return self._html_search_meta('dc.creator', html, 'uploader')

View file

@ -3,7 +3,7 @@ from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
get_meta_content, determine_ext,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
) )
@ -25,11 +25,11 @@ class HeiseIE(InfoExtractor):
'title': ( 'title': (
"Podcast: c't uplink 3.3 Owncloud / Tastaturen / Peilsender Smartphone" "Podcast: c't uplink 3.3 Owncloud / Tastaturen / Peilsender Smartphone"
), ),
'format_id': 'mp4_720', 'format_id': 'mp4_720p',
'timestamp': 1411812600, 'timestamp': 1411812600,
'upload_date': '20140927', 'upload_date': '20140927',
'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.', 'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.',
'thumbnail': 're:https?://.*\.jpg$', 'thumbnail': 're:^https?://.*\.jpe?g$',
} }
} }
@ -49,11 +49,12 @@ class HeiseIE(InfoExtractor):
info = { info = {
'id': video_id, 'id': video_id,
'thumbnail': self._og_search_thumbnail(webpage), 'thumbnail': self._og_search_thumbnail(webpage),
'timestamp': parse_iso8601(get_meta_content('date', webpage)), 'timestamp': parse_iso8601(
self._html_search_meta('date', webpage)),
'description': self._og_search_description(webpage), 'description': self._og_search_description(webpage),
} }
title = get_meta_content('fulltitle', webpage) title = self._html_search_meta('fulltitle', webpage)
if title: if title:
info['title'] = title info['title'] = title
else: else:
@ -64,9 +65,12 @@ class HeiseIE(InfoExtractor):
label = source_node.attrib['label'] label = source_node.attrib['label']
height = int_or_none(self._search_regex( height = int_or_none(self._search_regex(
r'^(.*?_)?([0-9]+)p$', label, 'height', default=None)) r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
video_url = source_node.attrib['file']
ext = determine_ext(video_url, '')
formats.append({ formats.append({
'url': source_node.attrib['file'], 'url': video_url,
'format_note': label, 'format_note': label,
'format_id': '%s_%s' % (ext, label),
'height': height, 'height': height,
}) })
self._sort_formats(formats) self._sort_formats(formats)