mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-11-03 19:17:02 -05:00
139 lines
5.0 KiB
Python
139 lines
5.0 KiB
Python
#!/usr/bin/env python2
|
|
|
|
__license__ = 'GPL v3'
|
|
'''
|
|
Hacker News
|
|
'''
|
|
from calibre.web.feeds.news import BasicNewsRecipe
|
|
from calibre.ptempfile import PersistentTemporaryFile
|
|
from urlparse import urlparse
|
|
import re
|
|
|
|
|
|
class HNWithCommentsLink(BasicNewsRecipe):
|
|
title = 'HN With Comments Link'
|
|
__author__ = 'Tom Scholl & David Kerschner'
|
|
description = u'Hacker News, run by Y Combinator. Anything that good hackers would find interesting, with a focus on programming and startups.'
|
|
publisher = 'Y Combinator'
|
|
category = 'news, programming, it, technology'
|
|
masthead_url = 'http://img585.imageshack.us/img585/5011/hnle.png'
|
|
cover_url = 'http://img585.imageshack.us/img585/5011/hnle.png'
|
|
delay = 1
|
|
max_articles_per_feed = 30
|
|
use_embedded_content = False
|
|
no_stylesheets = True
|
|
encoding = 'utf-8'
|
|
language = 'en'
|
|
requires_version = (0, 8, 16)
|
|
|
|
feeds = [
|
|
(u'Hacker News', 'http://news.ycombinator.com/rss')
|
|
]
|
|
|
|
temp_files = []
|
|
articles_are_obfuscated = True
|
|
|
|
def get_readable_content(self, url):
|
|
self.log('get_readable_content(' + url + ')')
|
|
br = self.get_browser()
|
|
f = br.open(url)
|
|
html = f.read()
|
|
f.close()
|
|
|
|
return self.extract_readable_article(html, url)
|
|
|
|
def get_hn_content(self, url):
|
|
self.log('get_hn_content(' + url + ')')
|
|
soup = self.index_to_soup(url)
|
|
main = soup.find('tr').findNextSiblings('tr', limit=2)[1].td
|
|
|
|
title = self.tag_to_string(main.find('td', 'title'))
|
|
link = main.find('td', 'title').find('a')['href']
|
|
if link.startswith('item?'):
|
|
link = 'http://news.ycombinator.com/' + link
|
|
readable_link = link.rpartition('http://')[2].rpartition('https://')[2]
|
|
subtext = self.tag_to_string(main.find('td', 'subtext'))
|
|
|
|
title_content_td = main.find('td', 'title').findParent(
|
|
'tr').findNextSiblings('tr', limit=3)[2].findAll('td', limit=2)[1]
|
|
title_content = u''
|
|
if not title_content_td.find('form'):
|
|
title_content_td.name = 'div'
|
|
title_content = title_content_td.prettify()
|
|
|
|
comments = u''
|
|
for td in main.findAll('td', 'default'):
|
|
comhead = td.find('span', 'comhead')
|
|
if comhead:
|
|
com_title = u'<h4>' + \
|
|
self.tag_to_string(comhead).replace(
|
|
' | link', '') + u'</h4>'
|
|
comhead.parent.extract()
|
|
br = td.find('br')
|
|
if br:
|
|
br.extract()
|
|
reply = td.find('a', attrs={'href': re.compile('^reply?')})
|
|
if reply:
|
|
reply.parent.extract()
|
|
td.name = 'div'
|
|
indent_width = (int(td.parent.find('td').img['width']) * 2) / 3
|
|
td['style'] = 'padding-left: ' + str(indent_width) + 'px'
|
|
comments = comments + com_title + td.prettify()
|
|
|
|
body = u'<h3>' + title + u'</h3><p><a href="' + link + u'">' + readable_link + \
|
|
u'</a><br/><strong>' + subtext + u'</strong></p>' + title_content + u'<br/>'
|
|
body = body + comments
|
|
return u'<html><title>' + title + u'</title><body>' + body + '</body></html>'
|
|
|
|
def parse_feeds(self):
|
|
a = super(HNWithCommentsLink, self).parse_feeds()
|
|
self.hn_articles = a[0].articles
|
|
return a
|
|
|
|
def get_obfuscated_article(self, url):
|
|
if url.startswith('http://news.ycombinator.com'):
|
|
content = self.get_hn_content(url)
|
|
else:
|
|
# TODO: use content-type header instead of url
|
|
is_image = False
|
|
for ext in ['.jpg', '.png', '.svg', '.gif', '.jpeg', '.tiff', '.bmp', ]:
|
|
if url.endswith(ext):
|
|
is_image = True
|
|
break
|
|
|
|
if is_image:
|
|
self.log('using image_content (' + url + ')')
|
|
content = u'<html><body><img src="' + url + u'"></body></html>'
|
|
else:
|
|
content = self.get_readable_content(url)
|
|
|
|
article = 0
|
|
for a in self.hn_articles:
|
|
if a.url == url:
|
|
article = a
|
|
|
|
content = re.sub(r'</body>\s*</html>\s*$', '', content) + \
|
|
article.summary + '</body></html>'
|
|
|
|
self.temp_files.append(PersistentTemporaryFile('_fa.html'))
|
|
self.temp_files[-1].write(content)
|
|
self.temp_files[-1].close()
|
|
return self.temp_files[-1].name
|
|
|
|
def is_link_wanted(self, url, tag):
|
|
if url.endswith('.pdf'):
|
|
return False
|
|
return True
|
|
|
|
def prettyify_url(self, url):
|
|
return urlparse(url).hostname
|
|
|
|
def populate_article_metadata(self, article, soup, first):
|
|
article.text_summary = self.prettyify_url(article.url)
|
|
article.summary = article.text_summary
|
|
|
|
# def parse_index(self):
|
|
# feeds = []
|
|
# feeds.append((u'Hacker News',[{'title': 'Testing', 'url': 'http://news.ycombinator.com/item?id=2935944'}]))
|
|
# return feeds
|