Subject: [ecasound] re: qtecasound not compiling
From: Vadim Tkachenko (vt@freehold.crocodile.org)
Date: Sun Jan 02 2000 - 06:28:24 EET
Kai,
Guess I got a little bit closer to understanding why qtecasound is SO
picky, meaning it doesn't want to compile if even all the components are
present.
Well, first of all, are you sure that you want main() in libqt for
ecasound to compile? I haven't actually *run* the resulting qtecasound
binary successfully ("Unable to open envelope
file/etc/ecasound/generic_oscillators"), but I'll get to that,
nevertheless, I presume that if I got that far, I didn't need main() in
libqt. Please correct me if I'm wrong.
Second, the --enable-qt/--disable-qt clause in the configure.in is not
correct. The proper (implies: convenient, as everyone does it) usage is
--enable-Z switches it on, --disable-Z switches it off. The way you did
it is both --enable-qt and --disable-qt switch it OFF. Moreover, even
though configure says that qtecasound *is* normally enabled, it doesn't
give away any warning, neither it terminates, if the conditions for
qtecasound compilation aren't met.
And finally, it's not just missing main() (I don't even know if it is
there in the library), but something else is wrong with the libraries
setup, and configure doesn't properly register the location of libqt
(even though I specify --with-qt-libraries), so if I override the
$qt_support to yes, and then 'make LIBS="-L/where/qt/is"', that does
work.
I guess I could come up with the patch that does the intelligent
recognition of qt libraries and fixes this problem, and since I intend
to try *ecasound one day I'll probably do that, but again (as quite a
while ago) I ask you to reconsider and put the code into the CVS
repository, which could boost the development speed in order of
magnitude.
Just so it doesn't seem to be too complicated, here's HOWTO:
http://www.working-dogs.com/docs/cvs.htma
--vt
This archive was generated by hypermail 2a24 : Tue Jan 04 2000 - 09:45:21 EET