| 1 |
# -*- coding: utf-8 -*- |
| 2 |
from __future__ import unicode_literals |
| 3 |
|
| 4 |
__license__ = 'GPL v3' |
| 5 |
__copyright__ = '2011, Piotr Kontek, [email protected] 2013-2018, Tomasz Długosz, [email protected]' |
| 6 |
|
| 7 |
import re |
| 8 |
import time |
| 9 |
|
| 10 |
from lxml import html |
| 11 |
|
| 12 |
from calibre.web.feeds.news import BasicNewsRecipe |
| 13 |
|
| 14 |
|
| 15 |
class GN(BasicNewsRecipe): |
| 16 |
__author__ = 'Piotr Kontek, Tomasz Długosz' |
| 17 |
title = 'Gość Niedzielny - pełny numer' |
| 18 |
publisher = 'Wydawnictwo Kurii Metropolitalnej w Katowicach' |
| 19 |
description = 'Ogólnopolski tygodnik katolicki - pełny numer sprzed 4 tygodni' |
| 20 |
encoding = 'utf-8' |
| 21 |
no_stylesheets = True |
| 22 |
language = 'pl' |
| 23 |
remove_javascript = True |
| 24 |
ignore_duplicate_articles = {'url'} |
| 25 |
masthead_url = 'http://gosc.pl/static/images/base/logotypes/100x68/gosc-niedzielny.png' |
| 26 |
|
| 27 |
def find_last_issue(self): |
| 28 |
raw = self.index_to_soup('http://gosc.pl/wyszukaj/wydania/3.Gosc-Niedzielny/', raw=True) |
| 29 |
doc = html.fromstring(raw) |
| 30 |
page = doc.xpath('//div[@class="search-result release-result"]/div[1]/div[1]/a/@href') |
| 31 |
|
| 32 |
if time.strftime('%w') in ['3', '4']: |
| 33 |
return page[5] |
| 34 |
else: |
| 35 |
return page[4] |
| 36 |
|
| 37 |
def parse_index(self): |
| 38 |
self.last_issue = self.find_last_issue() |
| 39 |
soup = self.index_to_soup('http://gosc.pl' + self.last_issue) |
| 40 |
self.cover_url = 'http://www.gosc.pl' + soup.find('div', attrs={'class': 'fl-w100 release-wp'}).findAll('a')[-4].contents[0]['src'] |
| 41 |
feeds = [] |
| 42 |
# editorial: |
| 43 |
a = soup.find('div', attrs={'class': 'release-wp-b'}) |
| 44 |
art = a.find('a') |
| 45 |
articles = [ |
| 46 |
{ |
| 47 |
'title': self.tag_to_string(art), |
| 48 |
'url': 'http://www.gosc.pl' + art['href'], |
| 49 |
'description': self.tag_to_string(a.find('p', attrs={'class': 'b lead'})), |
| 50 |
} |
| 51 |
] |
| 52 |
feeds.append(('Na dobry początek', articles)) |
| 53 |
# columns: |
| 54 |
for addr in soup.findAll('a', attrs={'href': re.compile(r'kategoria')}): |
| 55 |
if not addr.span: |
| 56 |
main_block = self.index_to_soup('http://www.gosc.pl' + addr['href']) |
| 57 |
articles = list(self.find_articles(main_block)) |
| 58 |
if len(articles) > 0: |
| 59 |
section = addr.contents[0] |
| 60 |
feeds.append((section, articles)) |
| 61 |
# not assigned content: |
| 62 |
page = 1 |
| 63 |
not_assigned = [] |
| 64 |
while True: |
| 65 |
soup = self.index_to_soup('http://gosc.pl' + self.last_issue.replace('przeglad', 'wszystko') + '/' + str(page)) |
| 66 |
articles = list(self.find_articles(soup)) |
| 67 |
not_assigned.extend(articles) |
| 68 |
page += 1 |
| 69 |
pages = soup.find('span', attrs={'class': 'pgr_nrs'}) |
| 70 |
if str(page) not in [self.tag_to_string(x) for x in pages.findAll('a')]: |
| 71 |
break |
| 72 |
|
| 73 |
feeds.append(('Nieprzypisane', not_assigned)) |
| 74 |
return feeds |
| 75 |
|
| 76 |
def find_articles(self, main_block): |
| 77 |
for a in main_block.findAll('div', attrs={'class': ['infoBox']}): |
| 78 |
art = a.find('a') |
| 79 |
yield { |
| 80 |
'title': self.tag_to_string(art), |
| 81 |
'url': 'http://www.gosc.pl' + art['href'].split('.')[0], |
| 82 |
'date': self.tag_to_string(a.find('b', attrs={'class': 'time'})).replace('DODANE', ' '), |
| 83 |
'description': self.tag_to_string(a.find('div', attrs={'class': 'txt'})), |
| 84 |
} |
| 85 |
|
| 86 |
def append_page(self, soup, appendtag): |
| 87 |
chpage = appendtag.find(attrs={'class': 'pgr_nrs'}) |
| 88 |
if chpage: |
| 89 |
for page in chpage.findAll('a'): |
| 90 |
soup2 = self.index_to_soup('http://gosc.pl' + page['href']) |
| 91 |
pagetext = soup2.find(attrs={'class': 'intextAd'}) |
| 92 |
pos = len(appendtag.contents) |
| 93 |
appendtag.insert(pos, pagetext) |
| 94 |
|
| 95 |
def preprocess_html(self, soup): |
| 96 |
self.append_page(soup, soup.body) |
| 97 |
r = soup.find(attrs={'class': 'lightbox'}) |
| 98 |
if r: |
| 99 |
r.contents[0]['src'] = r['href'] |
| 100 |
return soup |
| 101 |
|
| 102 |
def postprocess_html(self, soup, first_fetch): |
| 103 |
for r in soup.findAll(attrs={'class': 'pgr'}): |
| 104 |
r.extract() |
| 105 |
for r in soup.findAll(attrs={'class': ['di_dr', 'doc_image']}): |
| 106 |
del r['style'] |
| 107 |
''' |
| 108 |
for r in soup.findAll(attrs={'class': 'cm-i-a'}): |
| 109 |
r.replaceWith('<div style="clear:both"></div>' + |
| 110 |
r.prettify() + '<div style="clear:both"></div>') |
| 111 |
''' |
| 112 |
return soup |
| 113 |
|
| 114 |
keep_only_tags = [dict(name='div', attrs={'class': ['cf txt ', 'cf txt att-audio']})] |
| 115 |
|
| 116 |
remove_tags = [ |
| 117 |
dict(name='p', attrs={'class': 'l l-2 doc-source'}), |
| 118 |
dict(name='span', attrs={'class': 'wykop'}), |
| 119 |
dict( |
| 120 |
name='div', attrs={'class': ['doc_actions', 'cf', 'fr1_cl', 'txt__social-icons', 'txt__tags', 'fb-like fb_iframe_widget', 'jp-audio', 'jp-jplayer']} |
| 121 |
), |
| 122 |
dict(name='div', attrs={'id': 'vote'}), |
| 123 |
dict(name='link'), |
| 124 |
dict(name='a', attrs={'class': 'img_enlarge'}), |
| 125 |
] |
| 126 |
|
| 127 |
extra_css = ''' |
| 128 |
h1 {font-size:150%} |
| 129 |
p.limiter {font-size:150%; font-weight: bold} |
| 130 |
p.gn_subtitle {font-weight: bold} |
| 131 |
span.cm-i-a {text-transform:uppercase;font-size:50%} |
| 132 |
span.cm-i-p {font-style:italic; font-size:70%;text-align:right} |
| 133 |
span.gn_brb {color: red; font-weight: bold} |
| 134 |
div.txt__lead {font-weight: bold; font-size:150%} |
| 135 |
''' |