http://bugs.openembedded.org/show_bug.cgi?id=867
------- Additional Comments From [EMAIL PROTECTED] 2006-04-17 05:51 -------
(In reply to comment #2)
> It looks like you can get rid of the hal+glibc runtime dep build putting
pmount-hal in it's own package.
> Something like:
>
> PACKAGES += "${PN}_hal"
> RDEPENDS_${PN}_hal += "${PN}"
> FILES_${PN}-hal = "${bindir}/pmount-hal"
If you look at the patch, I did some pretty nasty stuff to remove the glib
dependency (like creating a local intl.h with #define gettext(X) X). Took me a
few minutes to recall why I didn't use the solution above, but I remember now: I
didn't want to have to compile glib just to use pmount =P
The fact that my patch is not needed to build pmount probably means it shouldn't
be used, but build time dependencies is still an issue. I'm building a distro
for a machine with no screen, hence it seems silly that I have to compile glib
to just use pmount/pumount.
Classpath had a similar problem where it depended on gtk and I ended up
splitting the package into classpath-gtk and classpath-minimal. The split is
more natural with classpath though (you can add a classpath-qt). In pmount's
case, creating a pmount-minimal build doesn't sound like the correct solution
(USE flags anyone? ;)
>
> That would mean no hal+glib dependency in the pmount package and people who
want pmount-hal can
> just install pmount-hal. Does that sound OK to you?
>
>
--
Configure bugmail: http://bugs.openembedded.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
_______________________________________________
Oe mailing list
[email protected]
https://www.handhelds.org/mailman/listinfo/oe
