bug-gnu-utils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Rv: Re: Patch to fix /[A-Z]/ and internationalization bug


From: Aharon Robbins
Subject: Re: Rv: Re: Patch to fix /[A-Z]/ and internationalization bug
Date: Sun, 05 Nov 2006 23:18:40 +0200

Mr. Trenholme,

> Date: Fri, 03 Nov 2006 10:59:00 -0600 (CST)
> From: Sam Trenholme <address@hidden>
> Subject: Rv: Re: Patch to fix /[A-Z]/ and internationalization bug
> To: address@hidden
>
> I'm letting other people on bug-gawk see my response
> to Eric's message.  Basically: Posix says nowhere that
> regular expression ranges *must* break when using a
> non-POSIX locale, and the "K" in Awk thinks breaking
> ranges is foolishness.

This is a hard problem to solve, and there's no good solution, other
than to move to [[:upper:]] and [[:lower:]] which are guaranteed to do
the right thing no matter what the locale.

It is not just GNU programs that have to deal with POSIX, but all the
commercial Unix vendors as well.

You are free to join the POSIX committee and try to get the standard
changed; membership is open to anyone with an interest and a willingness
to contribute.

In addition, you can use LC_ALL=C in your environment to get the behavior
you want.

If that's not acceptable, you can use [[:upper:]] and [[:lower:]].

If that's not acceptable, you can patch gawk and use your private version.

If that's not acceptable, you can use a different version of awk, there
are others out there, and the authors of the two freely-available versions
are both good friends of mine.

Free Software is all about giving the end user choice and freedom. So
if you wish to promote your comp.lang.awk boycott on gawk, you have my
blessing.

Thank you for your interest and proposed patches,

Arnold Robbins




reply via email to

[Prev in Thread] Current Thread [Next in Thread]