From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from joyce.cs.su.OZ.AU ([129.78.8.208]) by hawkwind.utcs.toronto.edu with SMTP id <2766>; Mon, 22 Feb 1993 22:37:00 -0500
Received: from moria.cs.su.OZ.AU (for hawkwind.utcs.toronto.edu) with MHSnet;
	Tue, 23 Feb 1993 14:36:37 +1100
From:	David Hogan <dhog@cs.su.oz.au>
Date:	Mon, 22 Feb 1993 22:26:28 -0500
To:	sam-fans@hawkwind.utcs.toronto.edu
In-Reply-To: <93Feb21.235516est.2758@hawkwind.utcs.toronto.edu>
Message-ID: <199302231426.26622.out.babus@cs.su.oz.au>
X-Face: "mmQCpqDtPafky^DZk|$l[`q{gw{Au4c>b/k]-H]fvn[nY@JvvEM^nP-ja^O5\bw!4~x\OH
 RKu~gL$=J$<Y~p_Onj0dnc2uBSa0^4:,&{JNJ^AA)%mR<~2_BIee|[u_XN}-:RBV|w=g![Y5_PZJr*
 wX:]4fB83j7/kcWRc*D"q:$?Qhq{SB#2OtL:&=NVSH|;LH9v7n~x'J/u*?Q_\:zUYN)Okvgyq-mi

> From:	rob@research.att.com

> 	y/^ +/ x/ +/ c/ /

> that's using ^ again; i thought the exercise was to avoid it.
> at least, that is the exercise i've been working on.

I thought the point was to avoid using it in sub-commands.  Although
the command we are trying to find might have to be the sub-command of
another x command, I suppose.

> when i made my suggestion
> 	x s2/ +/ /g
> i implicitly assumed, as i think did the original poster, that
> each line began with white space.
>[...]

Hmmm, I missed that when I first read the post.
Anyway, here's another way to do it:

	x/[^ ].*/ s/ +/ /g

(Dang! That ^ is unavoidable :-)