Re: [PATCHv2 15/15] gdb/tui: fair split of delta after a resize

[Andrew Burgess <aburgess@redhat.com>]

Tom Tromey <tom@tromey.com> writes:

>>>>>> "Andrew" == Andrew Burgess via Gdb-patches <gdb-patches@sourceware.org> writes:
>
> Andrew> Currently, in master gdb, when a tui window is changed in size, the
> Andrew> screen delta is mostly just added to the next available window.  We
> Andrew> do take care to respect the min/max size, but in most cases, these
> Andrew> limits are just "the terminal size", and so, we end up placing the
> Andrew> whole delta on the next window.
>
> This seems like a change that corresponds to the same change in the
> patch to change how layouts are applied.  That is, the resize-one-window
> code and the layout-from-scratch code ought to agree algorithmically.
> So maybe this should have been merged into that one -- though I'm not
> really too picky about that kind of thing.

Originally, I did try to do this, but in the end I decided not too as
the amount of change seemed much bigger than needed.

The problem is that we end up needing to add a mechanism to "lock" a
window size - we have a layout with windows A, B, and C.  The user sets
the size of window B.

With the code in my patch we set the size of B as the user requested,
and then resize A and C around B.

If we try to use apply, then, after changing the size of B, we need some
way to tell ::apply that it must not change the size of B again.

This is totally doable, at one point I had a version of this patch that
did just this, but it was a big code change.

I'd much rather push this smaller change, which allows me to get
additional tests into place, then, if we want, we can try to refactor
things later, and merge the two code paths together.

Would this be OK?

Thanks,
Andrew