Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-06-23 15:30:45 -04:00
Code Issues Packages Projects Releases Wiki Activity

Added logic to find current month's magazine url and crawl through pages until end of month's articles are reached instead of taking 2 pages off the all main magazine feed

Browse Source
This commit is contained in:
truth1ness 2015-05-01 17:34:02 -04:00
parent baa5bf6a5b
commit c833ccc79e
1 changed files with 36 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

42
recipes/wired.recipe
View File

@ -5,13 +5,15 @@ www.wired.com
'''
from calibre.web.feeds.news import BasicNewsRecipe
from datetime import date
import urllib2
class WiredDailyNews(BasicNewsRecipe):
title = 'Wired Magazine, Monthly Edition'
__author__ = 'Darko Miletic, update by Zach Lapidus'
__author__ = 'Darko Miletic, update by Zach Lapidus, Michael Marotta'
description = ('Wired is a full-color monthly American magazine, published in both print '
'and online editions, that reports on how emerging technologies affect culture,'
'the economy and politics.')
'the economy and politics. Monthly version, run at start of month')
publisher = 'Conde Nast'
category = 'news, IT, computers, technology'
oldest_article = 2
@ -40,8 +42,20 @@ class WiredDailyNews(BasicNewsRecipe):
dict(attrs={'data-js':['post', 'postHeader']}),
]
def parse_wired_index_page(self, num, seen):
soup = self.index_to_soup('http://www.wired.com/category/magazine/page/%d' % num)
def get_date_url(self):
'''
get month and year, add year modifier, append to wired magazine url,
:return: url
'''
baseurl = 'http://www.wired.com/tag/magazine-'
monthurl = str('{:02d}'.format(date.today().month))
yearurl = str(date.today().year - 1992)
dateurl = baseurl + yearurl + '-' + monthurl + '/page/'
return dateurl
def parse_wired_index_page(self, currenturl, seen):
soup = self.index_to_soup(currenturl)
for a in soup.find('main').findAll('a', href=True):
url = a['href']
if url.startswith('http://www.wired.com/') and url.endswith('/'):
@ -50,12 +64,26 @@ class WiredDailyNews(BasicNewsRecipe):
date = self.tag_to_string(dateloc)
if title.lower() != 'read more' and title and url not in seen:
seen.add(url)
self.log('Found article:', title, 'in page:', num)
self.log('Found article:', title)
yield {'title':title, 'date':date, 'url':url, 'description':''}
def parse_index(self):
'''
get the current month's url, index first page to soup, find number of pages,
just keep adding to page num until soup is not none instead of scraping page for
:return:
'''
baseurl = self.get_date_url()
pagenum = 1
articles = []
seen = set()
for num in (1, 2):
articles.extend(self.parse_wired_index_page(num, seen))
morepages = True
while morepages:
try:
url = urllib2.urlopen(baseurl + str(pagenum))
currenturl = baseurl + str(pagenum)
articles.extend(self.parse_wired_index_page(currenturl, seen))
pagenum += 1
except urllib2.HTTPError, e:
morepages = False
return [('Articles', articles)]