Can't get print_version to do anything

[MikeBlyth]

I'm brand new to Calibre and haven't used Python for years. I am following the tutorial for custom news feeds. I notice that BBC printable pages are now of the form ...something?print=true, so I have added to the basic recipe

def print_version(self, url):
    return url.append(u'?print=true')

But nothing seems to happen. In fact, no matter what print_version returns, even garbage, the recipe still runs just the same. Using ebook-convert for debugging, the URLs fetched are unchanged. If I put a divide by zero line in there, still no error, so it seems the function is not even being called. I must be doing something fundamentally wrong!

Full recipe:

class AdvancedUserRecipe1321386207(BasicNewsRecipe):
    title          = u'BBC'
    oldest_article = 3
    max_articles_per_feed = 2
    auto_cleanup = True
    no_stylesheets = True
    feeds          = [(u'BBC Front Page', u'http://newsrss.bbc.co.uk/rss/newsonline_world_edition/front_page/rss.xml')]

def print_version(self, url):
    return url.append(u'?print=true')

[kovidgoyal]

When you call ebook-convert, call it with the full path to the edited recipe file.

[MikeBlyth]

Yes, I'm using

ebook-convert c:\temp\test.recipe .epub --test -vv --debug-pipeline debug

[kovidgoyal]

You need a indent before def it has to be in line with the rest of the contents of the class

[MikeBlyth]

Aha, so that's what the tutorial meant about indentation ... I missed that--have been using Ruby too long. Thanks for the quick answer!

[MikeBlyth]

Although the print_version method works fine now, the result is still not the print version of the BBC story. Perhaps it's something funny about BBC. Using the command line
wget http://www.bbc.co.uk/news/world-midd...260?print=true
gives the right page, printable form. However, if I modify print_version to

def print_version(self, url):
    return 'http://www.bbc.co.uk/news/world-middle-east-15762260?print=true'

the non-print, full version is used in the resulting epub (I'm just hard-coding the whole URL as a way to debug; in practice, the expression would be "return url + u'?print=true' "). I know it seems unbelievable and possibly I'm still doing something wrong, but I'm only having the problem with BBC.

[Starson17]

Originally Posted by MikeBlyth

Although the print_version method works fine now, the result is still not the print version of the BBC story.

You probably should post your recipe, just in case there's something that someone can spot.

[MikeBlyth]

Here is the recipe:

class AdvancedUserRecipe1321386207(BasicNewsRecipe):
    title          = u'BBC'
    oldest_article = 3
    max_articles_per_feed = 2
    auto_cleanup = True
    no_stylesheets = True
    feeds          = [(u'BBC Front Page', u'http://newsrss.bbc.co.uk/rss/newsonline_world_edition/front_page/rss.xml')]

    def print_version(self, url):
        return url + u'?print=true'

Maybe it will work for someone else?

[kovidgoyal]

The problem is probably that the RSS feed provides links to an intermediate page which is automatically redirected (this is done by websites for tracking purposes). So your print=true is being discarded during the redirect. To get around it you would need to implement the get_article_url method in your recipe to resolve the correct article url.

[MikeBlyth]

Thanks.

[Starson17]

your print=true is being discarded during the redirect. To get around it you would need to implement the get_article_url method in your recipe to resolve the correct article url.

To do this, add this to your recipe.

    def get_article_url(self, article):
        print article
        return None

The recipe won't produce anything, but you can check the printed results to see if the link you need is in there. You can check to see what link you need by checking where the print version link points to.

[kiavash]

Did you find out why? I am having the same issue.

Edit:

Mine got solved this way:

PHP Code:

    def print_version(self, url):
        print url
        if url.find('/Journal/article.asp?HH_ID=') >= 0:
            return self.browser.open_novisit(url).geturl().replace('/Journal/article.asp?HH_ID=', '/Journal/Print.asp?Id=')
        elif  url.find('/News/article.asp?HH_ID=') >= 0:
            return self.browser.open_novisit(url).geturl().replace('/News/article.asp?HH_ID=', '/Journal/Print.asp?Id=')
        elif  url.find('/Resources/TechLib.asp?HH_ID=') >= 0:
            return self.browser.open_novisit(url).geturl().replace('/Resources/TechLib.asp?HH_ID=', '/Resources/PrintRessource.asp?Id=') 


[roedi06]

I had the exact same problem the last days, the solution was very simple after all

I just entered the feed I provided to the script into the browser, in you case:
http://newsrss.bbc.co.uk/rss/newsonl...t_page/rss.xml

It will then redirect you to:
http://feeds.bbci.co.uk/news/rss.xml?edition=int

Put this link into you script instead of the first xml-link. Now use the

def print_version(self, url):
    return url + '?print=true')

to correct the link. This will probably fix your problem. Have a look into this link if you don't get it.