On Fri, 25 Oct 2024 12:20:15 -0500
izzy Meyer <izder456@disroot.org> wrote:

> Whoops!
> 
> I forgot to remove the patches subdir.
> 
> Here's a fixed diff.
> 

Turns out that patch may not be needed. Although, it is usable, albeit
with a path change on 5.9.1. It doesn't seem to change anything with
how the program behaves. I was curious if it would fix the segfault
upon program exit issue, but it appears it does not.

Whoever put it there, looks like it was
myself, I'm unsure why it was needed for 5.8.0, but not for 5.9.x? Seems
like unistd.h is to not be needed here in the 5.9.x releases.

http://cvsweb.openbsd.org/cgi-bin/cvsweb/ports/games/minetest/patches/patch-lib_irrlichtmt_source_Irrlicht_CIrrDeviceLinux_cpp

Thoughts on this?

> --

I'm also contemplating making a default flavor, no_server, to build
without the -DBUILD_SERVER=TRUE configure flag, and have another flavor,
server, include this flag. The default would not not to build the
server just as it is upstream. I personally have no qualms with doing
this, especially considering that this is a non-default configure
option we manually enable in this port.

In the case of people who may be
running a server with this port, how can I properly inform them when
they pkg_add -u that the server isn't the default flavor? would the
flavor name be self-explanitory enough?

something like:

FLAVORS =		no_server server
FLAVORS?=		no_server

CONFIGURE_ARGS =	whatever it has now
CONFIGURE_ARGS+=	just no -DBUILD_SERVER flag

.if ${FLAVOR} == "server"
CONFIGURE_ARGS+=	-DBUILD_SERVER=TRUE
.endif

I'm unclear if this would have any benefits other than lowering build
time, slightly lowering disk space use, and giving feature flexibility
for binary users.

Just some brainstorming. I'd like to hear if others have any insight on
these things.

Thanks

Peace-

-- 
iz (they/them)

> i like to say mundane things,
> there are too many uninteresting things
> that go unnoticed.

izder456 (dot) neocities (dot) org