Re: Remove HISTSIZE env var altogether? (was: Re: [PATCH] Tweak the handling of $HISTSIZE edge cases [PR gdb/16999])

[Patrick Palka <patrick@parcs.ath.cx>]

On Fri, May 22, 2015 at 6:00 AM, Pedro Alves <palves@redhat.com> wrote:
> Changing title to call for attention.  Maybe we should ask
> on gdb@.  Background here:
>
>  https://sourceware.org/ml/gdb-patches/2015-05/msg00349.html
>  https://sourceware.org/ml/gdb-patches/2015-05/msg00563.html
>
>> What do you think about removing HISTSIZE/GDBHISTSIZE support
>> altogether?  It is awfully redundant (we can already automatically set
>> the history size via .gdbinit or via -ex "set history size foo") and
>> thus not really useful.  Even if we go along with replacing HISTSIZE
>> with GDBHISTSIZE I just can't see much use for it.
>
> What about GDBHISTFILE?  I think that the rationale for the existence
> of one should apply to both.  (with the HISTSIZE vs GDBHISTSIZE distinction
> being a separate matter.)

GDBHISTFILE is less useless than HISTSIZE I think.  I can imagine
unique use cases for GDBHISTFILE (e.g. to have separate per-project
history files) whereas for HISTSIZE, not so much.  So I don't think
their usefulness can be conflated.

>
> I'm really not sure.  Trying to play devil's advocate:
>
> #1 - An env var can be set once, for all users.  But that can be
>    done with --with-system-gdbinit=FILE as well.
>
> #2 - Along with GDBHISTFILE, it survives -nx.  Does it really matter?
>    I don't know.
>
> #3 - Seems friendly to allow at least GDBHISTFILE be an env var so it
>    can easily be toggled per host.  Though that can be done through
>    Python inside .gdbinit nowadays.  Though^2, Python isn't always
>    available.
>
> OTOH, I'm getting more convinced that we should at least
> rename HISTSIZE -> GDBHISTSIZE.  The cost of keeping that
> doesn't seem to be much.

I just don't see any utility in this environment variable.  I imagine
most users stumble upon this feature by realizing that their global
HISTSIZE variable is being read by GDB.  Once we rename it to
GDBHISTSIZE we will no longer have this coincidence and the variable
will be forever ignored.

>
> Thanks,
> Pedro Alves
>