From: "David A. Black" <dblack@...>
Date: 2005-09-29T00:01:47+09:00
Subject: Re: In search of a Regexp character class for "weird command characters"

Hi --

On Wed, 28 Sep 2005 Nuralanur@aol.com wrote:

> Hello,
>
> I have a regexp search problem.
> I have written a text-correction program  in Ruby which
> reads in a text file and marks every word that's not
> in a  dictionary array red in an  RTF output file
> (still a plain text file,  from Ruby's viewpoint) and saves that
> file (i.e., it is still a text, with  some commands specific
> to Rich Text Format).
> For instance, in a text, I  have a citation "(Fox , 1970)."
> Now, "(Fox" is not a correct English word, so  it should
> be red and bold, the comma is all right, so it stays black,  and
> "1970)." is not a correct English word, either, so it
> should be red  and bold, also.
> In RTF, you can achieve this by replacing
>
> a="(Fox , 1970)."
>
>   by
>
> b=" \cf1\b (Fox \cf0\b0 , \cf1\b 1970). \cf0\b0 ".
>
> Now, if you say
>
> p b
>
> Ruby will give the following output
>
>> " \0061\010 (Fox \0060\010 1970). \0060\0100 "
>
> However, I would like to remove all the characters of the form '\' +  number
> from the RTF file in a next step.
> Is there a character class for  Regexps (like \w,\S etc.) that achieves
> this?

Not a predefined one, but you can do:

   b.gsub!(/[006-010]/,"")

which leaves you with:

   " 1 (Fox 00 , 1 1970). 00 "

if you're sure that's what you want.


David

-- 
David A. Black
dblack@wobblini.net