[Gambas-user] BSD gambas compile

Bruce Steers bsteers4 at gmail.com
Thu Dec 31 17:00:24 CET 2020

On Wed, 30 Dec 2020 at 19:20, Benoît Minisini <g4mba5 at gmail.com> wrote:

> Le 30/12/2020 à 20:10, Bruce Steers a écrit :
> > aah , now you tell me lol.
> > okay point noted.
> >
> > On the plus side though i have BSD compiling up to date :)
> > (well GhostBSD at least)
> >
> > I've encountered 3 issues with 3 fixes applied after getting the
> > dependencies right...
> >
> > 1. trunk_version.h does not get created in ./main/
> > fix: copy the trunk_version.h file from another source folder
> No idea why.

Me neither,  where is the command that does it? I can investigate.

> > 2. sys/sysmacros.h not found for gbx_c_file.c
> > fix: commented out the define line on line 35.
> > (is there a way sysmacros.h can be ditched? maybe import the code?)
> You can use an "#ifdef OS_BSD" to not include the file on BSD.

Excellent , #ifndif even ;)

> > 3. BSD rmdir command does not like the --ignore-non-empty flag in the
> > makefiles  (it does not use ANY -- prefixes)
> > Fix: remove flag
> Is there a flag that does the same thing as in Linux?
Does not look like it, it has 2 args -pv
BSD info....

The following option is available:

     *-p*	     Each *directory* argument is	treated	as a pathname of which all
	     components	will be	removed, if they are empty, starting with the
	     last most component.  (See	rm(1)
for fully	non-discriminant re-
	     cursive removal.)

     *-v*	     Be	verbose, listing each directory	as it is removed.

I've made a branch for testing (cannot find BSD docker image on gitlab)

Applied the #ifndef OS_BSD fix  (works a treat)

Removed the -ignore-fail-on-non-empty flag (can what this does be done
another way?)

Only problems from master are missing rmdir --ignore-fail-on-non-empty flag
and trunk_version.h creation needs sorting out.

Wishing well
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gambas-basic.org/pipermail/user/attachments/20201231/65631296/attachment.htm>

More information about the User mailing list