Re: Collating characters?

[Aharon Robbins]

Greetings. Re this:

> Date: Sun, 05 Dec 2004 19:33:50 -0800
> From: Art Hixson <address@hidden>
> To: address@hidden
> Subject: Collating characters?
>
> I'm just trying out gawk for the first time.  Forgive me if I'm missing 
> something.  I'm running on XP (darn it).

Native, or using cygwin?

> I'm trying to use the construction /[[.70.]]/ to treat this character 
> pair as a unit.  However, contrary to the documentation, it gives :
>
> *gawk: prg.txt:7: fatal: Invalid collation character: /[[.70.]]/*
>
> when I try just /[.70.]/ I get all the 7s and 0s in any order , i.e., 
> ORed., which is just a character list.
>
> Where have I gone astray?  (leaving out my personal life as that would 
> take too many pages)
>
> --Art

I have to admit to having no experience with collating characters.
That said, I'll convey my understanding of them.

You cannot use [. and .] to group an arbitrary pair of characters
together.  Collating characters are defined by the locale in which
you're running, and only those defined by the locale are available for
use inside [. and .].  They usually have names, defined by the
locale; the name may or may not be the actual sequence of characters,
such [.ch.].

If all you want is to match lines where 7 is immediately followed by 0,
then just use

        gawk '/70/ { ... }'

Hope this helps,

Arnold