Parsing html tags to ruby characters

rubyonrails-talk
1

Hi,

I have a string containing some ruby code and html tags in-between. For example,

str = "require 'my_class.rb'<br>require 'your_class.rb'<br>&nbsp;&nbsp;:key=&gt;'hello'"

I want these html tags('<br>', '&nbsp;', '&gt;', '&lt;', '<p>', '<font>' etc...) to be replaced by the equivalent ruby characters("\n", " ", ">", "<" etc...).

These html tags can change dynamically according to the inputs.

Is there any way to parse these html tags to equivalent ruby characters?

Thanks in advance...

2

Thanks Ryan. But I can't guess what are all the tags i will be getting. Because those are dynamic. Any possible tag can come. So if I have to use the 'gsub' method, I will have to write for each and every html tag. Then that will be big.

So I am looking for any other easier way to implement this(something like html parser kind of).

3

You never specified what you wanted the

and tags replaced with either.

4

Sorry. That's my mistake. The final thing i want from the string is a runnable ruby code. So <p> and <font> tags can be removed from the string without any replacement.

Now I think, the only way to implement this is to use the 'gsub' method for each and every possible tag.

5

Sorry. That's my mistake. The final thing i want from the string is a runnable ruby code. So <p> and <font> tags can be removed from the string without any replacement.

Now I think, the only way to implement this is to use the 'gsub'
method for each and every possible tag.

Well assuming the only tag with special meaning is <br> Then you can
just convert entities to their respective characters (there are tables
of these), <br> to "\n" and then just replace every other tag with ''.
No need for one regexp per tag for that!

Fred

6

But "&gt;" and "&lt;" need to be replaced with ">" and "<" respectively. Because I will having some ruby hash code in the string.

Also I need to find out all the html tags in that string. Is there any way to find that?

7

But "&gt;" and "&lt;" need to be replaced with ">" and "<"
respectively. Because I will having some ruby hash code in the string.

I'm not seeing the problem :slight_smile: Replace entities and then look for
everything between < and >. Change it to a newline if it's a br, or
just replace it with blank and add it to your list of html tags. Fred

8

Thanks for your replies. I have done as I wanted. The following the code for that.

    markup = markup.gsub('<br>', "\n")     markup = markup.gsub(/[\<]([\/])*([A-Za-z0-9])*[\>]/, '')     markup = markup.gsub('&gt;', ">")     markup = markup.gsub('&lt;', "<")     markup = markup.gsub('&nbsp;', " ")     markup = markup.gsub('&amp;', "&")

It's working fine now. But I am not sure whether I have covered all the tags and characters or not.

9

depends what you are trying todo. there are far more html entities that that. (a partial list is here http://www.w3schools.com/tags/ref_entities.asp) and of course there are the unicode style ones (http://theorem.ca/~mvcorks/code/charsets/auto.html)

Fred

10

I have a very, very strong suspicion that the need is only to translate character enconding (e.g., &amp=>'&').

It might be worth considering iterating over an array of hashes rather than repeating the same code with different parameters:

[{:regex=>/\<br\>/, :decoded=>"\n"}, {:regex=>/[\<]([\/])*([A-Za-z0-9])*[\>]/, :decoded=>''}, {:regex=>/&gt;/, :decoded=>'>'} ... ].each do |decoding_hash|   markup.gsub!(decoding_hash[:regex], decoding_hash[:decoded]) end

The advantage is in keeping the code DRY and making the intentions of the block a bit clearer.