Re: getting wide character related matching info out of regex?

[Paolo Bonzini]

On 09/15/2010 08:25 AM, Aharon Robbins wrote:
> Hi all.
>
> A discussion came up recently that in order for gawk's match() function
> to set RSTART and RLENGTH in terms of characters and not bytes

May I first of all ask for a pointer about this discussion?  I 
understand that this is to make sure that substr() complies with POSIX:

"Return the at most n-character substring of s that begins at position 
m, numbering from 1. If n is missing, the length of the substring will 
be limited by the length of the string s."

Honestly, it seems a gratuitous change to me, going against the existing 
practice of pretty much every awk implementation around.  Are you sure 
you want to go through this burden??

Paolo