volkskrant.recipe broken

m.tarenskeen · 10-10-2010, 03:41 PM

Hi,

Recently De Volkskrant (NL) changed their website. I think that is why I cannot read news downloaded with calibre offline on my e-reader anymore.

The titles and headlines are fetched, but instead of the corresponding articles I can just see an url pointing to the article that I want to see. My simple e-reader does not have WiFi.

I am using calibre 0.7.23.
Maybe one of the "recipe gurus" here can take a look at volkskrant.recipe and see if it can be fixed ?

MT

TonytheBookworm · 10-10-2010, 06:40 PM

Quote:

Originally Posted by m.tarenskeen

Hi,

Recently De Volkskrant (NL) changed their website. I think that is why I cannot read news downloaded with calibre offline on my e-reader anymore.

The titles and headlines are fetched, but instead of the corresponding articles I can just see an url pointing to the article that I want to see. My simple e-reader does not have WiFi.

I am using calibre 0.7.23.
Maybe one of the "recipe gurus" here can take a look at volkskrant.recipe and see if it can be fixed ?

MT

the very last two rss feeds appear to be broken on the site itself. They are feedburner links (cough cough) but anyway i commented them out. If valid rss links are discovered later let me know and I will gladly update it.

Here is a working version - Modified to utilize obfuscation to get the print version of the articles and removed the keep_only_tags

Spoiler:

Code:

#!/usr/bin/env python
# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
from __future__ import with_statement

__license__   = 'GPL v3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'

'''
 Modified by Tony Stegall
 on 10/10/10 to include function to grab print version of articles
'''

from calibre.web.feeds.news import BasicNewsRecipe
'''
added by Tony Stegall
'''
#######################################################
from calibre.ptempfile import PersistentTemporaryFile
#######################################################

class AdvancedUserRecipe1249039563(BasicNewsRecipe):
    title          = u'De Volkskrant'
    __author__     = 'acidzebra'
    oldest_article = 7
    max_articles_per_feed = 100
    no_stylesheets = True
    language = 'nl'

    
    extra_css      = '''
                        body{font-family:Arial,Helvetica,sans-serif; font-size:small;}
                        h1{font-size:large;}
                     '''
    '''
      Change Log:
        Date:       10/10/10  - Modified code to include obfuscated to get the print version
        Author:   Tony Stegall
    '''
   ####################################################################################################### 
    temp_files = []
    articles_are_obfuscated = True

    def get_obfuscated_article(self, url):
        br = self.get_browser()
        print 'THE CURRENT URL IS: ', url
        br.open(url)
        
        try:
         response = br.follow_link(url_regex='.*?(2010)(\\/)(article)(\\/)(print)(\\/)', nr = 0)
         html = response.read()
        except:
         response = br.open(url)
         html = response.read()
         
        self.temp_files.append(PersistentTemporaryFile('_fa.html'))
        self.temp_files[-1].write(html)
        self.temp_files[-1].close()
        return self.temp_files[-1].name

   ###############################################################################################################
   
    feeds          = [
                      (u'Laatste Nieuws', u'http://volkskrant.nl/rss/laatstenieuws.rss'),
                      (u'Binnenlands nieuws', u'http://volkskrant.nl/rss/nederland.rss'), 
                      (u'Buitenlands nieuws', u'http://volkskrant.nl/rss/internationaal.rss'), 
                      (u'Economisch nieuws', u'http://volkskrant.nl/rss/economie.rss'), 
                      (u'Sportnieuws', u'http://volkskrant.nl/rss/sport.rss'), 
                      (u'Kunstnieuws', u'http://volkskrant.nl/rss/kunst.rss'), 
                      '''
                        both of these rss feeds link back to the main volksrant.nl url a.k.a Broken
                        If someone happens to know the correct paths then they can put them in here
                      '''
                      #(u'Wetenschapsnieuws', u'http://feeds.feedburner.com/DeVolkskrantWetenschap'), 
                      #(u'Technologienieuws', u'http://feeds.feedburner.com/vkmedia')
                      ]

''' 
example for formating
'''
# original url: http://www.volkskrant.nl/vk/nl/2668/Buitenland/article/detail/1031493/2010/10/10/Noord-Korea-ziet-nieuwe-leider.dhtml 
# print url :   http://www.volkskrant.nl/vk/nl/2668/2010/article/print/detail/1031493/Noord-Korea-ziet-nieuwe-leider.dhtml

m.tarenskeen · 10-11-2010, 04:23 AM

Thanks!

m.tarenskeen · 10-12-2010, 05:58 AM

[QUOTE=TonytheBookworm;1156268]
try:
response = br.follow_link(url_regex='.*?(2010)(\\/)(article)(\\/)(print)(\\/)', nr = 0)
html = response.read()
except:
response = br.open(url)
html = response.read()

Looks like this will only work in 2010 and will be outdated after 3 months already ?
What about something like

try:
for yy in range(2010,2020):
response = br.follow_link(url_regex='.*?(%d)(\\/)(article(\\/)(print)(\\/)', nr = 0) % yy

(BTW: How do I keep my indentation intact in this online message editor ? If I "preview Post" it disappears. Not exactly what I want when posting Python code.)

Starson17 · 10-12-2010, 07:54 AM

Quote:

Originally Posted by m.tarenskeen

(BTW: How do I keep my indentation intact in this online message editor ? If I "preview Post" it disappears. Not exactly what I want when posting Python code.)

Highlight code and press the code button (hash/pound/number sign)

TonytheBookworm · 10-12-2010, 06:00 PM

Quote:

Looks like this will only work in 2010 and will be outdated after 3 months already ?
What about something like

try:
for yy in range(2010,2020):
response = br.follow_link(url_regex='.*?(%d)(\\/)(article(\\/)(print)(\\/)', nr = 0) % yy

(BTW: How do I keep my indentation intact in this online message editor ? If I "preview Post" it disappears. Not exactly what I want when posting Python code.)

yeah i thought about that when writing the code; however i figured who is to say it will be in that format in 2011 so i just kept it the way it was. I figure when 2011 rolls around they will change the site anyway and force us to rewrite the recipe anyway

but good thinking on the date range thing.

m.tarenskeen · 10-14-2010, 07:29 PM

Quote:

Originally Posted by TonytheBookworm

yeah i thought about that when writing the code; however i figured who is to say it will be in that format in 2011 so i just kept it the way it was. I figure when 2011 rolls around they will change the site anyway and force us to rewrite the recipe anyway

but good thinking on the date range thing.

...but I have to do some more thinking to make the trick really work. My first experiments did not give the wished result.

But I did find and update/replace the broken links for "Technologie nieuws" en "Wetenschap". Replaced them with "Media" and "Gezondheid & Wetenschap".

My updated version of volkskrant.recipe attached.

BTW: I still think the "only-2010 bug" should be fixed before it can be included in the calibre distribution.

BTW2: The volkskrant recipe that comes with the current version of Calibre not only is broken, but also the filename is wrong: "volkskant.recipe" should be "volkskrant.recipe"

kovidgoyal · 10-14-2010, 07:40 PM

I fixed the year problem in the calibre version of the recipe. You can see it in the calibre source code, or wait for next release.

m.tarenskeen · 12-31-2010, 06:27 AM

Quote:

Originally Posted by kovidgoyal

I fixed the year problem in the calibre version of the recipe. You can see it in the calibre source code, or wait for next release.

I have a feeling the recipe will not work correctly in the beginning of 2011: It will then only fetch articles from 2011, but it probably still has to be able to fetch older articles from december 2010 also. I think the current recipe will fail on that point.

I have thought of a fix for this issue. I will test it as soon as 2011 has begun, and will report what I find and what I did to fix the problem. Stay tuned!

m.tarenskeen · 01-01-2011, 11:18 AM

Quote:

Originally Posted by m.tarenskeen

I have a feeling the recipe will not work correctly in the beginning of 2011: It will then only fetch articles from 2011, but it probably still has to be able to fetch older articles from december 2010 also. I think the current recipe will fail on that point.

I have thought of a fix for this issue. I will test it as soon as 2011 has begun, and will report what I find and what I did to fix the problem. Stay tuned!

I have done some testing and my modified version gives better results than the currently distributed recipe. This is my version:

Code:

# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
from __future__ import with_statement

__license__   = 'GPL v3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'

'''
 Modified by Tony Stegall
 on 10/10/10 to include function to grab print version of articles
'''

from datetime import date
from calibre.web.feeds.news import BasicNewsRecipe
'''
added by Tony Stegall
'''
#######################################################
from calibre.ptempfile import PersistentTemporaryFile
#######################################################

class AdvancedUserRecipe1249039563(BasicNewsRecipe):
    title          = u'De Volkskrant'
    __author__     = 'acidzebra'
    oldest_article = 7
    max_articles_per_feed = 100
    no_stylesheets = True
    language = 'nl'

    extra_css      = '''
                        body{font-family:Arial,Helvetica,sans-serif;font-size:small;}
                        h1{font-size:large;}
                     '''
    '''
      Change Log:
        Date:       10/10/10  - Modified code to include obfuscated to get the print version
        Author:   Tony Stegall
        
        Date:       01/01/11  - Modified for better results around December/January.
        Author:   Martin Tarenskeen
    '''
   #######################################################################################################
    temp_files = []
    articles_are_obfuscated = True

    def get_obfuscated_article(self, url):
        br = self.get_browser()
        print 'THE CURRENT URL IS: ', url
        br.open(url)
        year = date.today().year

        try:
            response = br.follow_link(url_regex='.*?(%d)(\\/)(article)(\\/)(print)(\\/)'%year, nr = 0)
            html = response.read()
        except:
            year = year-1
            try:
                response = br.follow_link(url_regex='.*?(%d)(\\/)(article)(\\/)(print)(\\/)'%year, nr = 0)
                html = response.read()
            except:
                response = br.open(url)
                html = response.read()


        self.temp_files.append(PersistentTemporaryFile('_fa.html'))
        self.temp_files[-1].write(html)
        self.temp_files[-1].close()
        return self.temp_files[-1].name

   ###############################################################################################################

    '''
      Change Log:
       Date: 10/15/2010
       Feeds updated by Martin Tarenskeen
    '''

    feeds          = [
                      (u'Laatste Nieuws', u'http://www.volkskrant.nl/rss/laatstenieuws.rss'),
                      (u'Binnenland', u'http://www.volkskrant.nl/rss/nederland.rss'),
                      (u'Buitenland', u'http://www.volkskrant.nl/rss/internationaal.rss'),
                      (u'Economie', u'http://www.volkskrant.nl/rss/economie.rss'),
                      (u'Sport', u'http://www.volkskrant.nl/rss/sport.rss'),
                      (u'Cultuur', u'http://www.volkskrant.nl/rss/kunst.rss'),
                      (u'Gezondheid & Wetenschap', u'http://www.volkskrant.nl/rss/wetenschap.rss'),
                      (u'Internet & Media', u'http://www.volkskrant.nl/rss/media.rss') ]

10-10-2010, 03:41 PM	#1
m.tarenskeen Member Posts: 14 Karma: 10 Join Date: Oct 2010 Device: BeBook One	volkskrant.recipe broken Hi, Recently De Volkskrant (NL) changed their website. I think that is why I cannot read news downloaded with calibre offline on my e-reader anymore. The titles and headlines are fetched, but instead of the corresponding articles I can just see an url pointing to the article that I want to see. My simple e-reader does not have WiFi. I am using calibre 0.7.23. Maybe one of the "recipe gurus" here can take a look at volkskrant.recipe and see if it can be fixed ? MT

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
New York Times recipe broken?	gianfri	Calibre	1	03-20-2010 09:52 AM
Recipe for The Week broken?	gianfri	Calibre	3	03-19-2010 08:05 PM
Recipe Volkskrant paid version	prodsaaw	Calibre	0	02-18-2010 04:00 PM
Engadget Recipe Broken	pars_andy	Calibre	1	12-01-2009 10:39 PM
Economist Recipe - broken?	dieterpops	Calibre	1	02-20-2009 09:14 PM

10-11-2010, 04:23 AM	#3
m.tarenskeen Member Posts: 14 Karma: 10 Join Date: Oct 2010 Device: BeBook One	Thanks!

10-12-2010, 05:58 AM	#4
m.tarenskeen Member Posts: 14 Karma: 10 Join Date: Oct 2010 Device: BeBook One	[QUOTE=TonytheBookworm;1156268] try: response = br.follow_link(url_regex='.?(2010)(\\/)(article)(\\/)(print)(\\/)', nr = 0) html = response.read() except: response = br.open(url) html = response.read() Looks like this will only work in 2010 and will be outdated after 3 months already ? What about something like try: for yy in range(2010,2020): response = br.follow_link(url_regex='.?(%d)(\\/)(article(\\/)(print)(\\/)', nr = 0) % yy (BTW: How do I keep my indentation intact in this online message editor ? If I "preview Post" it disappears. Not exactly what I want when posting Python code.)

10-14-2010, 07:40 PM	#8
kovidgoyal creator of calibre Posts: 43,858 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	I fixed the year problem in the calibre version of the recipe. You can see it in the calibre source code, or wait for next release.

Advert

Advert