[PATCH 2/5] subsurface-configure.pri: enable the libusb detection stage

Lubomir I. Ivanov neolit123 at gmail.com
Wed Oct 9 15:06:48 UTC 2013


On 10 October 2013 01:00, Dirk Hohndel <dirk at hohndel.org> wrote:
> On Thu, 2013-10-10 at 00:54 +0300, Lubomir I. Ivanov wrote:
>> On 10 October 2013 00:26, Thiago Macieira <thiago at macieira.org> wrote:
>> > On quarta-feira, 9 de outubro de 2013 23:30:00, Lubomir I. Ivanov wrote:
>> >> but if tell libdc to *use* libusb and then i produce a
>> >> libdivecomputer.a should libusb be part of that archive?
>> >> my guess is no...
>> >
>> > No.
>> >
>> > When static linking to a given library, we have to link to its dependencies
>> > directly.
>> >
>> > But I see the problem. Go ahead with your original patch, but remove the lines
>> > that link to libusb that were added in
>> > cfbbf1bffec5e917aa4987f33825f324e2f03976.
>> >
>>
>> ok, i see.
>> it works if i do a: git revert cfbbf1bffec
>
> Hmm - as discussed many times - I /prefer/ to link statically against
> libdivecomputer. But if that causes us too much grief on Windows, then
> so be it.
>

well technically, it still links statically for me as i simply pass a
static library via pkg-config and i don't need to have a DLL next to
the .exe.
perhaps the section: "Find libdivecomputer" can be improved later.

>> seems that we can't have both patches, but your original line the way to go:
>> link_pkgconfig: packagesExist(libusb-1.0): PKGCONFIG += libusb-1.0
>
> With all the back and forth... can you send a cleaned up series? :-)
>

ok, will do.

lubomir
--


More information about the subsurface mailing list