Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix #1022163 (Updated recipe for Adventure Gamers)

Browse Source
This commit is contained in:
Kovid Goyal 2012-07-08 09:26:27 +05:30
parent 436965bd25
commit 2a551d882b
1 changed files with 31 additions and 28 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

59
recipes/adventuregamers.recipe
View File

@ -1,5 +1,5 @@
__license__ = 'GPL v3' __license__ = 'GPL v3'
__copyright__ = '2009-2010, Darko Miletic <darko.miletic at gmail.com>' __copyright__ = '2009-2012, Darko Miletic <darko.miletic at gmail.com>'
''' '''
www.adventuregamers.com www.adventuregamers.com
''' '''
@ -14,24 +14,24 @@ class AdventureGamers(BasicNewsRecipe):
publisher = 'Adventure Gamers' publisher = 'Adventure Gamers'
category = 'news, games, adventure, technology' category = 'news, games, adventure, technology'
oldest_article = 10 oldest_article = 10
delay = 10 #delay = 10
max_articles_per_feed = 100 max_articles_per_feed = 100
no_stylesheets = True no_stylesheets = True
encoding = 'cp1252' encoding = 'utf8'
remove_javascript = True remove_javascript = True
use_embedded_content = False use_embedded_content = False
INDEX = u'http://www.adventuregamers.com' INDEX = u'http://www.adventuregamers.com'
extra_css = """ extra_css = """
.pageheader_type{font-size: x-large; font-weight: bold; color: #828D74} .pageheader_type{font-size: x-large; font-weight: bold; color: #828D74}
.pageheader_title{font-size: xx-large; color: #394128} .pageheader_title,.page_title{font-size: xx-large; color: #394128}
.pageheader_byline{font-size: small; font-weight: bold; color: #394128} .pageheader_byline{font-size: small; font-weight: bold; color: #394128}
.score_bg {display: inline; width: 100%; margin-bottom: 2em} .score_bg {display: inline; width: 100%; margin-bottom: 2em}
.score_column_1{ padding-left: 10px; font-size: small; width: 50%} .score_column_1{ padding-left: 10px; font-size: small; width: 50%}
.score_column_2{ padding-left: 10px; font-size: small; width: 50%} .score_column_2{ padding-left: 10px; font-size: small; width: 50%}
.score_column_3{ padding-left: 10px; font-size: small; width: 50%} .score_column_3{ padding-left: 10px; font-size: small; width: 50%}
.score_header{font-size: large; color: #50544A} .score_header{font-size: large; color: #50544A}
.bodytext{display: block} img{margin-bottom: 1em;}
body{font-family: Helvetica,Arial,sans-serif} body{font-family: 'Open Sans',Helvetica,Arial,sans-serif}
""" """
conversion_options = { conversion_options = {
@ -41,35 +41,38 @@ class AdventureGamers(BasicNewsRecipe):
, 'language' : language , 'language' : language
} }
keep_only_tags = [ keep_only_tags = [dict(name='div', attrs={'class':'cleft_inn'})]
dict(name='div', attrs={'class':'content_middle'})
]
remove_tags = [ remove_tags = [
dict(name=['object','link','embed','form']) dict(name=['object','link','embed','form','iframe','meta'])
,dict(name='div', attrs={'class':['related-stories','article_leadout','prev','next','both']}) ,dict(name='a', attrs={'href':'http://www.adventuregamers.com/about/scoring'})
,dict(name='a', attrs={'href':'http://www.adventuregamers.com/about/policies'})
] ]
remove_tags_after = [dict(name='div', attrs={'class':'bodytext'})]
remove_tags_after = [dict(name='div', attrs={'class':'toolbar_fat'})]
remove_attributes = ['width','height'] remove_attributes = ['width','height']
feeds = [(u'Articles', u'http://feeds2.feedburner.com/AdventureGamers')] feeds = [(u'Articles', u'http://www.adventuregamers.com/rss/')]
def get_article_url(self, article): def get_article_url(self, article):
return article.get('guid', None) url = BasicNewsRecipe.get_article_url(self, article)
if '/videos/' in url or '/hypeometer/' in url:
return None
return url
def append_page(self, soup, appendtag, position): def append_page(self, soup, appendtag, position):
pager = soup.find('div',attrs={'class':'toolbar_fat_next'}) pager = soup.find('div', attrs={'class':'pagination_big'})
if pager: if pager:
nexturl = self.INDEX + pager.a['href'] nextpage = soup.find('a', attrs={'class':'next-page'})
soup2 = self.index_to_soup(nexturl) if nextpage:
texttag = soup2.find('div', attrs={'class':'bodytext'}) nexturl = nextpage['href']
for it in texttag.findAll(style=True): soup2 = self.index_to_soup(nexturl)
del it['style'] texttag = soup2.find('div', attrs={'class':'bodytext'})
newpos = len(texttag.contents) for it in texttag.findAll(style=True):
self.append_page(soup2,texttag,newpos) del it['style']
texttag.extract() newpos = len(texttag.contents)
appendtag.insert(position,texttag) self.append_page(soup2,texttag,newpos)
texttag.extract()
pager.extract()
appendtag.insert(position,texttag)
def preprocess_html(self, soup): def preprocess_html(self, soup):
@ -78,7 +81,7 @@ class AdventureGamers(BasicNewsRecipe):
for item in soup.findAll('div', attrs={'class':'floatright'}): for item in soup.findAll('div', attrs={'class':'floatright'}):
item.extract() item.extract()
self.append_page(soup, soup.body, 3) self.append_page(soup, soup.body, 3)
pager = soup.find('div',attrs={'class':'toolbar_fat'}) pager = soup.find('div',attrs={'class':'pagination_big'})
if pager: if pager:
pager.extract() pager.extract()
return self.adeify_images(soup) return self.adeify_images(soup)