Stop using DO_NADA when not needed

[bapt]

Summary

There is no reason to run 'true' for empty targets

[mat]

Mk/bsd.port.mk

2785–2792

Maybe change that and put the ignorelist-verbose: taget before the first .if test, and remove the .else.

2871–2872

Maybe put the .if defined(NO_CHECKSUM) around the call to checksum.sh and remove this.

2877–2878

Maybe move this to the test: definition at the end.

2885–2886

Mmm, NO_DESCRIBE is not currently in use, and I don't feel like it is a good idea anyway, many remove it.

2896–2897

Maybe just remove this, if it is created, it goes in the targets pipeline, if not, it does not.

2902–2903

GNOME_OPTION_MSG was removed ages ago, maybe just remove all this.

3289–3290

Why not just remove the .elif ? If it does not exist, it just won't be added to the targets pipeline.

3397–3399

Nothing currently uses install-mtree, time to go ?

3668–3670

maybe change that to create a dependency below so that the target is not created if it is not needed, like

.if target(pre-distclean)
distclean: pre-distclean
.endif

[bapt]

Address mat comments

[bapt]

Mk/bsd.port.mk

2896–2897

This one must exist to be hooked on

3397–3399

Can't :) added a comment :)

3668–3670

I think we should just nuke pre-distclean

[bapt]

Fix bad rebase

[bapt]

do-test

[bapt]

rebase

[mat]

Mk/bsd.port.mk

3668–3670

It was my first reaction, yes, I'll have a look at if it was ever used.

[bapt]

Mk/bsd.port.mk

3668–3670

I think we should just keep this removal for another patchset :)

[mat]

Mk/bsd.port.mk

2896–2897

Mmmm, it does not need to be hooked in, it's in _SANITY_SEQ, so, I don't understand.

3668–3670

So, I asked git log -Spre-distclean -p, it was added in 1996 and never, ever used, so, yeah, remove it :-)

[bapt]

Remove more unused targets

[mat]

Looks good to me, it might need an exp-run, I'll leave the other portmgr members look at it here.