ifzm by rty. Fixes #6043 (Wrong Chinese Category in News)

This commit is contained in:
Kovid Goyal 2010-07-01 08:46:53 -06:00
parent 3a16d7770a
commit e05dedfa0f
5 changed files with 68 additions and 4 deletions

View File

@ -7,7 +7,7 @@ class AdvancedUserRecipe1277228948(BasicNewsRecipe):
__author__ = 'rty'
__version__ = '1.0'
language = 'zh_CN'
language = 'zh'
pubisher = 'www.chinapressusa.com'
description = 'Overseas Chinese Network Newspaper in the USA'
category = 'News in Chinese, USA'

View File

@ -0,0 +1,50 @@
from calibre.web.feeds.news import BasicNewsRecipe
class AdvancedUserRecipe1277305250(BasicNewsRecipe):
title = u'infzm - China Southern Weekly'
oldest_article = 14
max_articles_per_feed = 100
feeds = [(u'\u5357\u65b9\u5468\u672b-\u70ed\u70b9\u65b0\u95fb', u'http://www.infzm.com/rss/home/rss2.0.xml'),
(u'\u5357\u65b9\u5468\u672b-\u7ecf\u6d4e\u65b0\u95fb', u'http://www.infzm.com/rss/economic.xml'),
(u'\u5357\u65b9\u5468\u672b-\u6587\u5316\u65b0\u95fb', u'http://www.infzm.com/rss/culture.xml'),
(u'\u5357\u65b9\u5468\u672b-\u751f\u6d3b\u65f6\u5c1a', u'http://www.infzm.com/rss/lifestyle.xml'),
(u'\u5357\u65b9\u5468\u672b-\u89c2\u70b9', u'http://www.infzm.com/rss/opinion.xml')
]
__author__ = 'rty'
__version__ = '1.0'
language = 'zh'
pubisher = 'http://www.infzm.com'
description = 'Chinese Weekly Tabloid'
category = 'News, China'
remove_javascript = True
use_embedded_content = False
no_stylesheets = True
#encoding = 'GB2312'
encoding = 'UTF-8'
conversion_options = {'linearize_tables':True}
masthead_url = 'http://i50.tinypic.com/2qmfb7l.jpg'
extra_css = '''
@font-face { font-family: "DroidFont", serif, sans-serif; src: url(res:///system/fonts/DroidSansFallback.ttf); }\n
body {
margin-right: 8pt;
font-family: 'DroidFont', serif;}
.detailContent {font-family: 'DroidFont', serif, sans-serif}
'''
keep_only_tags = [
dict(name='div', attrs={'id':'detailContent'}),
]
remove_tags = [
dict(name='div', attrs={'id':['detailTools', 'detailSideL', 'pageNum']}),
]
remove_tags_after = [
dict(name='div', attrs={'id':'pageNum'}),
]
def preprocess_html(self, soup):
for item in soup.findAll(color=True):
del item['font']
for item in soup.findAll(style=True):
del item['style']
return soup

View File

@ -20,7 +20,7 @@ from calibre.utils.config import config_dir
from calibre.utils.date import isoformat, now, parse_date
from calibre.utils.localization import get_lang
from calibre.utils.logging import Log
from calibre.utils.zipfile import ZipFile, safe_replace
from calibre.utils.zipfile import ZipFile
from PIL import Image as PILImage
@ -1727,7 +1727,6 @@ class ITUNES(DriverBase):
return thumb_data
thumb_path = book_path.rpartition('.')[0] + '.jpg'
format = book_path.rpartition('.')[2].lower()
if isosx:
title = book.name()
elif iswindows:

View File

@ -0,0 +1,15 @@
#!/usr/bin/env python
# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
__license__ = 'GPL v3'
__copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
from calibre.customize import Plugin
class CoverDownload(Plugin):
supported_platforms = ['windows', 'osx', 'linux']
author = 'Kovid Goyal'
type = _('Cover download')

View File

@ -8,7 +8,7 @@ import copy
from lxml import html, etree
from lxml.html.builder import HTML, HEAD, TITLE, STYLE, DIV, BODY, \
STRONG, EM, BR, SPAN, A, HR, UL, LI, H2, H3, IMG, P as PT, \
STRONG, BR, SPAN, A, HR, UL, LI, H2, H3, IMG, P as PT, \
TABLE, TD, TR
from calibre import preferred_encoding, strftime, isbytestring