Fix Calgary Herald

This commit is contained in:
Kovid Goyal 2012-04-17 20:58:20 +05:30
parent d9b9961c77
commit 8d3c9d93b3

View File

@ -1,220 +1,35 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
__license__ = 'GPL v3'
'''
www.canada.com
'''
import re
from calibre.web.feeds.news import BasicNewsRecipe from calibre.web.feeds.news import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import BeautifulStoneSoup
class CalgaryHerald(BasicNewsRecipe):
class CanWestPaper(BasicNewsRecipe):
# un-comment the following four lines for the Victoria Times Colonist
## title = u'Victoria Times Colonist'
## url_prefix = 'http://www.timescolonist.com'
## description = u'News from Victoria, BC'
## fp_tag = 'CAN_TC'
# un-comment the following four lines for the Vancouver Province
## title = u'Vancouver Province'
## url_prefix = 'http://www.theprovince.com'
## description = u'News from Vancouver, BC'
## fp_tag = 'CAN_VP'
# un-comment the following four lines for the Vancouver Sun
## title = u'Vancouver Sun'
## url_prefix = 'http://www.vancouversun.com'
## description = u'News from Vancouver, BC'
## fp_tag = 'CAN_VS'
# un-comment the following four lines for the Edmonton Journal
## title = u'Edmonton Journal'
## url_prefix = 'http://www.edmontonjournal.com'
## description = u'News from Edmonton, AB'
## fp_tag = 'CAN_EJ'
# un-comment the following four lines for the Calgary Herald
title = u'Calgary Herald' title = u'Calgary Herald'
url_prefix = 'http://www.calgaryherald.com' oldest_article = 3
description = u'News from Calgary, AB' max_articles_per_feed = 100
fp_tag = 'CAN_CH'
# un-comment the following four lines for the Regina Leader-Post feeds = [
## title = u'Regina Leader-Post' (u'News', u'http://rss.canada.com/get/?F233'),
## url_prefix = 'http://www.leaderpost.com' (u'Calgary', u'http://www.calgaryherald.com/scripts/sp6query.aspx?catalog=cahr&tags=keyword|calgary&output=rss?link=http%3a%2f%2fwww.calgaryherald'),
## description = u'News from Regina, SK' (u'Alberta', u'http://www.calgaryherald.com/scripts/Sp6Query.aspx?catalog=CAHR&tags=Keyword|Alberta&output=rss?link=http%3A%2F%2Fwww.calgaryherald.com%2Fnews%2Falberta%2Findex.html'),
## fp_tag = '' (u'Politics', u'http://rss.canada.com/get/?F7551'),
(u'National', u'http://rss.canada.com/get/?F7552'),
# un-comment the following four lines for the Saskatoon Star-Phoenix (u'World', u'http://rss.canada.com/get/?F7553'),
## title = u'Saskatoon Star-Phoenix' ]
## url_prefix = 'http://www.thestarphoenix.com' __author__ = 'rty'
## description = u'News from Saskatoon, SK' pubisher = 'Calgary Herald'
## fp_tag = '' description = 'Calgary, Alberta, Canada'
category = 'News, Calgary, Alberta, Canada'
# un-comment the following four lines for the Windsor Star
## title = u'Windsor Star'
## url_prefix = 'http://www.windsorstar.com'
## description = u'News from Windsor, ON'
## fp_tag = 'CAN_'
# un-comment the following four lines for the Ottawa Citizen
## title = u'Ottawa Citizen'
## url_prefix = 'http://www.ottawacitizen.com'
## description = u'News from Ottawa, ON'
## fp_tag = 'CAN_OC'
# un-comment the following four lines for the Montreal Gazette
## title = u'Montreal Gazette'
## url_prefix = 'http://www.montrealgazette.com'
## description = u'News from Montreal, QC'
## fp_tag = 'CAN_MG'
language = 'en_CA' remove_javascript = True
__author__ = 'Nick Redding' use_embedded_content = False
no_stylesheets = True no_stylesheets = True
timefmt = ' [%b %d]' language = 'en_CA'
extra_css = ''' encoding = 'utf-8'
.timestamp { font-size:xx-small; display: block; } conversion_options = {'linearize_tables':True}
#storyheader { font-size: medium; } ##masthead_url = 'http://www.calgaryherald.com/index.html'
#storyheader h1 { font-size: x-large; } keep_only_tags = [
#storyheader h2 { font-size: large; font-style: italic; } dict(name='div', attrs={'id':'storyheader'}),
.byline { font-size:xx-small; } dict(name='div', attrs={'id':'storycontent'})
#photocaption { font-size: small; font-style: italic }
#photocredit { font-size: xx-small; }'''
keep_only_tags = [dict(name='div', attrs={'id':'storyheader'}),dict(name='div', attrs={'id':'storycontent'})]
remove_tags = [{'class':'comments'},
dict(name='div', attrs={'class':'navbar'}),dict(name='div', attrs={'class':'morelinks'}),
dict(name='div', attrs={'class':'viewmore'}),dict(name='li', attrs={'class':'email'}),
dict(name='div', attrs={'class':'story_tool_hr'}),dict(name='div', attrs={'class':'clear'}),
dict(name='div', attrs={'class':'story_tool'}),dict(name='div', attrs={'class':'copyright'}),
dict(name='div', attrs={'class':'rule_grey_solid'}),
dict(name='li', attrs={'class':'print'}),dict(name='li', attrs={'class':'share'}),dict(name='ul', attrs={'class':'bullet'})]
def get_cover_url(self): ]
from datetime import timedelta, date remove_tags_after = {'class':"story_tool_hr"}
if self.fp_tag=='':
return None
cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str(date.today().day)+'/lg/'+self.fp_tag+'.jpg'
br = BasicNewsRecipe.get_browser()
daysback=1
try:
br.open(cover)
except:
while daysback<7:
cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str((date.today() - timedelta(days=daysback)).day)+'/lg/'+self.fp_tag+'.jpg'
br = BasicNewsRecipe.get_browser()
try:
br.open(cover)
except:
daysback = daysback+1
continue
break
if daysback==7:
self.log("\nCover unavailable")
cover = None
return cover
def fixChars(self,string):
# Replace lsquo (\x91)
fixed = re.sub("\x91","",string)
# Replace rsquo (\x92)
fixed = re.sub("\x92","",fixed)
# Replace ldquo (\x93)
fixed = re.sub("\x93","“",fixed)
# Replace rdquo (\x94)
fixed = re.sub("\x94","”",fixed)
# Replace ndash (\x96)
fixed = re.sub("\x96","",fixed)
# Replace mdash (\x97)
fixed = re.sub("\x97","—",fixed)
fixed = re.sub("&#x2019;","",fixed)
return fixed
def massageNCXText(self, description):
# Kindle TOC descriptions won't render certain characters
if description:
massaged = unicode(BeautifulStoneSoup(description, convertEntities=BeautifulStoneSoup.HTML_ENTITIES))
# Replace '&' with '&'
massaged = re.sub("&","&", massaged)
return self.fixChars(massaged)
else:
return description
def populate_article_metadata(self, article, soup, first):
if first:
picdiv = soup.find('body').find('img')
if picdiv is not None:
self.add_toc_thumbnail(article,re.sub(r'links\\link\d+\\','',picdiv['src']))
xtitle = article.text_summary.strip()
if len(xtitle) == 0:
desc = soup.find('meta',attrs={'property':'og:description'})
if desc is not None:
article.summary = article.text_summary = desc['content']
def strip_anchors(self,soup):
paras = soup.findAll(True)
for para in paras:
aTags = para.findAll('a')
for a in aTags:
if a.img is None:
a.replaceWith(a.renderContents().decode('cp1252','replace'))
return soup
def preprocess_html(self, soup):
return self.strip_anchors(soup)
def parse_index(self):
soup = self.index_to_soup(self.url_prefix+'/news/todays-paper/index.html')
articles = {}
key = 'News'
ans = ['News']
# Find each instance of class="sectiontitle", class="featurecontent"
for divtag in soup.findAll('div',attrs={'class' : ["section_title02","featurecontent"]}):
#self.log(" div class = %s" % divtag['class'])
if divtag['class'].startswith('section_title'):
# div contains section title
if not divtag.h3:
continue
key = self.tag_to_string(divtag.h3,False)
ans.append(key)
self.log("Section name %s" % key)
continue
# div contains article data
h1tag = divtag.find('h1')
if not h1tag:
continue
atag = h1tag.find('a',href=True)
if not atag:
continue
url = self.url_prefix+'/news/todays-paper/'+atag['href']
#self.log("Section %s" % key)
#self.log("url %s" % url)
title = self.tag_to_string(atag,False)
#self.log("title %s" % title)
pubdate = ''
description = ''
ptag = divtag.find('p');
if ptag:
description = self.tag_to_string(ptag,False)
#self.log("description %s" % description)
author = ''
autag = divtag.find('h4')
if autag:
author = self.tag_to_string(autag,False)
#self.log("author %s" % author)
if not articles.has_key(key):
articles[key] = []
articles[key].append(dict(title=title,url=url,date=pubdate,description=description,author=author,content=''))
ans = [(key, articles[key]) for key in ans if articles.has_key(key)]
return ans