From: eden <eden.li@...>
Date: 2006-10-06T18:20:16+09:00
Subject: Re: Stripping unwanted html

Hey, that's RoR's strip_tags method :)

I ran into the same issue so went down your path and instead of
fretting about no docs, just hacked my way through using IRB.

Here's a modified version of what I came up with, maybe you'll find it
useful?

  def strip_tags_except(html, exceptions = [])
    if html.index("<")
      text = ""
      tokenizer = HTML::Tokenizer.new(html)
      while token = tokenizer.next
        case node = HTML::Node.parse(nil, 0, 0, token, false)
        when HTML::Tag
          text << node.to_s if exceptions.include?(node.name)
        when HTML::Text
          text << node.to_s
        end
      end
      text
    else
      html
    end
  end

The one I had also stripped attributes and closed up dangling tags if
it found any.  Have a look at RoR's strip_links for more examples of
HTML::Node/HTML::Tokenizer usage.

Wild Al wrote:
> Hi everyone:
>
> I'm trying to strip html with the exception of a few html tags.
>
> I have found the following code:
>
>   def strip_tags(html)
>     if html.index("<")
>       text = ""
>       tokenizer = HTML::Tokenizer.new(html)
>
>       while token = tokenizer.next
>         node = HTML::Node.parse(nil, 0, 0, token, false)
>         # result is only the content of any Text nodes
>         text << node.to_s if node.class == HTML::Text
>       end
>       # strip any comments, and if they have a newline at the end (ie.
> line with
>       # only a comment) strip that too
>       text.gsub(/<!--(.*?)-->[\n]?/m, "")
>     else
>       html # already plain text
>     end
> end
>
> I'm trying to understand what is going on in this code but cannot find
> documenation for HTML::Tokenizer or HTML::Node.parse.  Does anyone know
> the use of the parameters in the parse method?
>
> In the while loop, how do you access the html tag.  If I could access
> the html tags, I could then decide if I wanted to keep the tag or not.
>
> Thanks for reading,
> Wild Al
> 
> -- 
> Posted via http://www.ruby-forum.com/.