bug-bash
[Top][All Lists]
Advanced

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

Re: Regression in pattern substitution with compat42


From: Martin D Kealey
Subject: Re: Regression in pattern substitution with compat42
Date: Fri, 10 Feb 2023 11:32:54 +1300 (NZDT)
User-agent: Alpine 2.21 (DEB 202 2017-01-01)

On 8 Feb 2023 at 18:50Z Tom Briden wrote:
> Bash Version: 5.2
> Patch Level: 15
> Release Status: release
>
> As of version 5.2-beta, replacing a single backslash with a double backslash 
> is no longer possible when using BASH_COMPAT=4.2.

On 9 Feb 2023 at 15:57Z Chet Ramey wrote:
> shopt -u patsub_replacement

Surely that should be automatic with compat51 and earlier, since it's a
breaking change to previous behaviour?

Sometimes mis-features like this make their way into public releases of
Bash, and it seems to me that there needs to be a way of saying "sorry, that
release was a mistake, don't use it", rather than every script from now to
eternity having to cope with potentially being run under this broken version
of bash.

In other projects this is done by having two release streams, "LTS" for long
term support, and "standard" with a relatively short end of life, typically
6-24 months. (And yes sometimes a new short term version is released that
has no changes from the previous one except a new expiry date.)

Is there any chance that we could get a scheme like this for Bash, where
standard releases come with a "don't use after" date, so that script writers
don't have to cope with broken versions potentially remaining forever?

-Martin

PS: the current "beta testing" arrangement seems to me to have too few
participants by at least two orders of magnitude. We need to encourage
"bleeding edge" Linux distros and similar to offer pre-built alpha and beta
releases of Bash.

PPS: personally I would go as far as emitting a warning message after a
cut-off date, otherwise people will find some excuse to ignore the
installation instructions.

PPPS: I'd be happy to work up a patch to implement this latter suggestion.



reply via email to

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