From mboxrd@z Thu Jan  1 00:00:00 1970
MIME-Version: 1.0
In-Reply-To: <51a0fc4b4bb26e2894cc078f681b2bd2@swcp.com>
References: <2bca08f7e5b22db500c54deb4b58bc0c@swcp.com>
	<b3ff81f04ee52820237c23ead451c137@kw.quanstro.net>
	<51a0fc4b4bb26e2894cc078f681b2bd2@swcp.com>
From: Stanley Lieber <stanley.lieber@gmail.com>
Date: Mon, 14 Jun 2010 10:16:24 -0500
Message-ID: <AANLkTimYWnwt01Ch530xDUH3Ixni2_AC8lgW43yRXkTQ@mail.gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [9fans] naming conventions
Topicbox-Message-UUID: 332ed36e-ead6-11e9-9d60-3106f5b1d025

On Mon, Jun 14, 2010 at 8:47 AM, EBo <ebo@sandien.com> wrote:
>
>> your shell is broken. =C2=A0[0-9] are perfectly valid anywhere in a unix=
 file
>> name.
>
> does bash completion work for you on *anything*? =C2=A0Or does it get con=
fused
> thinking that you are specifying a previous command (!### in the history)=
?
> Yes, a file/script named [0-9]* is valid, but still breaks things. =C2=A0=
Maybe
> there is some trick to get around it that I do not knwo about, but the bi=
g
> issues where those that the editor raised.
>
> =C2=A0EBo --

For what it's worth, ksh on OpenBSD handles completion on 9* files just fin=
e.

-sl