calibre/recipes/pc_lab.recipe

71 lines
2.4 KiB
Python

#!/usr/bin/env python
from calibre.web.feeds.recipes import BasicNewsRecipe
class PCLab(BasicNewsRecipe):
cover_url = 'http://pclab.pl/img/logo.png'
title = u"PC Lab"
__author__ = 'ravcio - rlelusz[at]gmail.com'
description = u"Articles from PC Lab website"
language = 'pl'
oldest_article = 30.0
max_articles_per_feed = 100
recursions = 0
encoding = 'iso-8859-2'
no_stylesheets = True
remove_javascript = True
use_embedded_content = False
keep_only_tags = [
dict(name='div', attrs={'class':['substance']})
]
remove_tags = [
dict(name='div', attrs={'class':['chapters']})
,dict(name='div', attrs={'id':['script_bxad_slot_display_list_bxad_slot']})
]
remove_tags_after = [
dict(name='div', attrs={'class':['navigation']})
]
#links to RSS feeds
feeds = [ ('PCLab', u'http://pclab.pl/xml/artykuly.xml') ]
#load second and subsequent page content
# in: soup - full page with 'next' button
# out: appendtag - tag to which new page is to be added
def append_page(self, soup, appendtag):
# find the 'Next' button
pager = soup.find('div', attrs={'class':'next'})
if pager:
#search for 'a' element with link to next page (exit if not found)
a = pager.find('a')
if a:
nexturl = a['href']
soup2 = self.index_to_soup('http://pclab.pl/' + nexturl)
pagetext_substance = soup2.find('div', attrs={'class':'substance'})
pagetext = pagetext_substance.find('div', attrs={'class':'data'})
pagetext.extract()
pos = len(appendtag.contents)
appendtag.insert(pos, pagetext)
pos = len(appendtag.contents)
self.append_page(soup2, appendtag)
def preprocess_html(self, soup):
# soup.body contains no title and no navigator, they are in soup
self.append_page(soup, soup.body)
# finally remove some tags
tags = soup.findAll('div',attrs={'class':['tags', 'index', 'script_bxad_slot_display_list_bxad_slot', 'index first', 'zumi', 'navigation']})
[tag.extract() for tag in tags]
return soup