Recipe for "Robb Wolf"

[thoraxe]

Trying to play with Calibre instead of fighting with the browser on the Kindle, just for giggles.

Starting to go through my various blogs, and started with http://www.robbwolf.com

Here's the recipe so far:

PHP Code:

from calibre.web.feeds.recipes import BasicNewsRecipe

class RobbWolf(BasicNewsRecipe):
    title          = u'Robb Wolf - Paleo Solution'
    __author__  = 'Erik M Jacobs'
    oldest_article = 7
    max_articles_per_feed = 100
    no_stylesheets = True
    use_embedded_content = False
    feeds          = [(u'Robb Wolf - Paleo Solution', u'http://feeds.feedburner.com/RobbWolfThePaleoSolution?format=xml')]
    keep_only_tags = dict(id='content')
    remove_tags_after = [dict(name='div', attrs={'class':['endpost']})]
    remove_tags = [dict(name='div', attrs={'align':['center']}),
                   dict(name='div', attrs={'class':['postinfo']})] 


Main issue I'm having is that the h2 is a link and falls inside of the content, which seems to confuse Calibre. I end up with a single page on the Kindle with just the article title, and then the real article begins on the next page.

Is it possible to use regexp in the keep/remove/etc tags lines?

This is a standard Wordpress blog, but only the abstracts are presented. I tried messing around with the recipe for Mish's Global Economic Analysis but end up basically only getting the abstracts and no real articles.

Any suggestions here?

[Starson17]

Originally Posted by thoraxe

Is it possible to use regexp in the keep/remove/etc tags lines?

Yes.

Here's how I used it in the Skeptic recipe to remove div tags with an id that started with "follow":

    remove_tags = [dict(name='div', attrs={'class':['Introduction','divider']}),
                  dict(name='div', attrs={'id':['feature', 'podcast']}),
                  dict(name='div', attrs={'id':re.compile(r'follow.*', re.DOTALL|re.IGNORECASE)}), 
                  dict(name='hr'),
                  ]