Re: [hackers] [sbase] [PATCH 3/3] ed: Fix substitutions with non-determinate patterns

From: Evan Gates <evan.gates_AT_gmail.com>
Date: Mon, 25 Jul 2016 17:41:44 -0700

On Sun, Jul 24, 2016 at 8:17 PM, Wolfgang Corcoran-Mathe
<wcm_AT_sigwinch.xyz> wrote:
> Sorry, this last patch is simpleminded.
>
> Given the following text and substitution, what should happen?
>
> 'glorious'
>
> s/G*/P/g
>
> sbase sed(1) gives us 'PgPlPoPrPiPoPuPsP' (matching every null
> string), while GNU ed(1) gives an error. The above patch does
> not throw an error, but rather performs the first substitution
> and stops.
>
> Is there a case in which a global substitution for a regex that
> matches the null string is useful? Should we produce output like
> 'PgPlPoPrPiPoPuPsP' (which is almost certainly unwanted garbage)
> or assume that such an expression is a mistake?
>
> Regards,
>
> --
> wcm
>

The sed behavior is based on other sed implementations. I think I
based it on GNU sed and busybox sed test cases that make sure to
increment the position in the string by one after each 0 length match.
Whereas GNU has problems with consistency, I think sbase should not.
Whether that means changing ed or sed, I think they should give the
same result. That being said I still think sed is in desperate need of
a (probably complete) rewrite in order to handle null bytes (i.e.
don't use libc string functions).

-emg
Received on Tue Jul 26 2016 - 02:41:44 CEST