calibre/recipes/wprost.recipe
Kovid Goyal 567040ee1e Perform PEP8 compliance checks on the entire codebase
Some bits of PEP 8 are turned off via setup.cfg
2016-07-29 21:25:17 +05:30

101 lines
4.1 KiB
Python

#!/usr/bin/env python2
__license__ = 'GPL v3'
__copyright__ = '''2010, matek09, matek09@gmail.com
Modified 2011, Mariusz Wolek <mariusz_dot_wolek @ gmail dot com>
Modified 2012, Artur Stachecki <artur.stachecki@gmail.com>'''
from calibre.web.feeds.news import BasicNewsRecipe
import re
class Wprost(BasicNewsRecipe):
EDITION = 0
FIND_LAST_FULL_ISSUE = True
EXCLUDE_LOCKED = True
ICO_BLOCKED = 'http://www.wprost.pl/G/layout2/ico_blocked.png'
title = u'Wprost'
__author__ = 'matek09'
description = u'Popularny tygodnik ogólnopolski - Wprost. Najlepszy wśród polskich tygodników - opiniotwórczy - społeczno-informacyjny - społeczno-kulturalny.' # noqa
encoding = 'ISO-8859-2'
no_stylesheets = True
language = 'pl'
remove_javascript = True
recursions = 0
remove_tags_before = dict(dict(name='div', attrs={'id': 'print-layer'}))
remove_tags_after = dict(dict(name='div', attrs={'id': 'print-layer'}))
'''
keep_only_tags =[]
keep_only_tags.append(dict(name = 'table', attrs = {'id' : 'title-table'}))
keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'div-header'}))
keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'div-content'}))
keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'def element-autor'}))
'''
preprocess_regexps = [(re.compile(r'style="display: none;"'), lambda match: ''),
(re.compile(r'display: block;'), lambda match: ''),
(re.compile(r'\<td\>\<tr\>\<\/table\>'), lambda match: ''),
(re.compile(r'\<table .*?\>'), lambda match: ''),
(re.compile(r'\<tr>'), lambda match: ''),
(re.compile(r'\<td .*?\>'), lambda match: ''),
(re.compile(r'\<div id="footer"\>.*?\</footer\>'), lambda match: '')]
remove_tags = []
remove_tags.append(dict(name='div', attrs={'class': 'def element-date'}))
remove_tags.append(dict(name='div', attrs={'class': 'def silver'}))
remove_tags.append(
dict(name='div', attrs={'id': 'content-main-column-right'}))
extra_css = '''.div-header {font-size: x-small; font-weight: bold}'''
# h2 {font-size: x-large; font-weight: bold}
def is_blocked(self, a):
if a.findNextSibling('img') is None:
return False
else:
return True
def find_last_issue(self):
soup = self.index_to_soup('http://www.wprost.pl/archiwum/')
a = 0
if self.FIND_LAST_FULL_ISSUE:
ico_blocked = soup.findAll('img', attrs={'src': self.ICO_BLOCKED})
a = ico_blocked[-1].findNext(
'a', attrs={'title': re.compile(r'Spis *', re.IGNORECASE | re.DOTALL)})
else:
a = soup.find('a', attrs={'title': re.compile(
r'Spis *', re.IGNORECASE | re.DOTALL)})
self.EDITION = a['href'].replace('/tygodnik/?I=', '')
self.EDITION_SHORT = a['href'].replace('/tygodnik/?I=15', '')
self.cover_url = a.img['src']
def parse_index(self):
self.find_last_issue()
soup = self.index_to_soup(
'http://www.wprost.pl/tygodnik/?I=' + self.EDITION)
feeds = []
headers = soup.findAll(
attrs={'class': 'block-header block-header-left mtop20 mbottom20'})
articles_list = soup.findAll(attrs={'class': 'standard-box'})
for i in range(len(headers)):
articles = self.find_articles(articles_list[i])
if len(articles) > 0:
section = headers[i].find('a').string
feeds.append((section, articles))
return feeds
def find_articles(self, main_block):
articles = []
for a in main_block.findAll('a'):
if a.name in "td":
break
if self.EXCLUDE_LOCKED and self.is_blocked(a):
continue
articles.append({
'title': self.tag_to_string(a),
'url': 'http://www.wprost.pl' + a['href'],
'date': '',
'description': ''
})
return articles