Re: [RFC/PATCH 00/48] Libifying git apply

Christian Couder [off-list ref] writes:

One point I'd especially welcome feedback about is the fact that there
are many boolean options that are using OPT_BOOL(...), so they use an
int. And there are a few others that are using OPT_BIT(...), so they
use just a bit. I wonder if it is worth it to try to be consistent,
and maybe also to try to save some memory.

Related to this, some of the variables for these options have not been
moved into the "apply_state" structure, because they are not global to
the file, but maybe for consistency they should be.

These might be worthy clean-ups but that is only if they are done
after we make sure conversion proper is done faithfully to the
original, i.e. without introducing unnecessary bugs.  I'd advise
against doing them before the libification is done.

On Wed, Mar 9, 2016 at 7:14 PM, Junio C Hamano [off-list ref] wrote:

Christian Couder [off-list ref] writes:

quoted

One point I'd especially welcome feedback about is the fact that there
are many boolean options that are using OPT_BOOL(...), so they use an
int. And there are a few others that are using OPT_BIT(...), so they
use just a bit. I wonder if it is worth it to try to be consistent,
and maybe also to try to save some memory.

Related to this, some of the variables for these options have not been
moved into the "apply_state" structure, because they are not global to
the file, but maybe for consistency they should be.

These might be worthy clean-ups but that is only if they are done
after we make sure conversion proper is done faithfully to the
original, i.e. without introducing unnecessary bugs.  I'd advise
against doing them before the libification is done.

Ok, I will try to avoid those kind of clean-ups.