Recipe for ng.pl

[markoz]

Hi all,

I'm complete newbie user of calibre and has zero experience with programming.
I was trying to make my first custom recipe for a print version of national-geographic.pl articles.

I compared urls both non print and print version, and for me seems that I need to replace 'artykuly/pokaz' with 'drukuj-artykul'.
Examples:
http://www.national-geographic.pl/ar...ojny-z-zaraza/
http://www.national-geographic.pl/dr...ojny-z-zaraza/

I created the part after "def print_version" according to instruction I found - but it does not work - fetching basic, non print version of an article.

Can anyone help me please?

class al(BasicNewsRecipe):
author = 'mk'
description = 'National Geographic'
title = u'National Geographic'
oldest_article = 7
max_articles_per_feed = 5


feeds = [(u'National Geographic', u'http://www.national-geographic.pl/rss/')]

def print_version(self, url):
return url.replace('artykuly/pokaz', 'drukuj-artykul')

[bubak]

I assume your recipe has the correct indentation, what you've written would not run in Python. The feed links have a '0C' instead of '/', so your function must be

def print_version(self, url):
    return url.replace('artykuly0Cpokaz', 'drukuj-artykul')

[markoz]

Originally Posted by bubak

I assume your recipe has the correct indentation, what you've written would not run in Python. The feed links have a '0C' instead of '/', so your function must be

def print_version(self, url):
    return url.replace('artykuly0Cpokaz', 'drukuj-artykul')

Hi, thanks that's working!

Marcin

[markoz]

Originally Posted by markoz

Hi, thanks that's working!

Marcin

Where can I find table for replacing '_' with '0C' etc.?

[bubak]

If you mean the encoding in this source, this is not the standard char to hex conversion, that would be 0x2f for '/', so you have to look e.g. in the trace output of ebook-convert to see what urls are involved.