i tried to comment out the MACOS_10_3_OR_EARLIER conditional
and i got numerous errors, which leads me to think that maybe im missing a project definition.
uncommenting it resulted in a successfull build, but trying to build my own project now results in linking errors. after a cleanup the errors return.
im using the 10.5 base sdk for compiling juce
trying to compile in 32/64bit universal mode , gave me even more errors…
yes checked out the last svn revision yesterday
(639 from the 12th)
the code seems correct when i look at it,
and the wierdness gets worse after i comment that line - get an error uncomment it, and i get a successful build. but then it doesnt link in my own project. another cleanup in the juce project after that pseudo successful compilations just brings the same 4 errors back.
the error happens when compiling juce_socket.cpp …
ill try going back some revisions and see where it broke
small updates:
i updated to revision 641 and i still get the errors.
i downgraded to revision 632 and i get no errors. so i guess im going to pin it down soon
638 compiles no problem…
Ah, well I did remove an old Carbon.h include, but it seemed to be fine without it. Funny thing is that you’re getting an error in a CoreServices header, and it really shouldn’t be loading any CoreServices stuff…
Ok im no expert on OS stuff but i did some digging and here is what i got:
i went to "juce_PlatformDefs.h"
and set: #define MACOS_10_3_OR_EARLIER 1
to 0
compiled 641 and it worx
so i wanted to see what defines my 10.5 as 10.3
and reached: “AvailabilityMacros.h” in the mac sdk
where it said:
[code]/*
If min OS not specified, assume 10.1
Note: gcc driver may set ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED based on MACOSX_DEPLOYMENT_TARGET environment variable
*/[/code]
in the xcode project , on the juce target, there is a user definition :
MACOSX_DEPLOYMENT_TARGET_ppc [10.3]
and im wondering if that has anything to do with it.
as i understand cocoa is just fully functional from 10.4 or 10.5, and power pcs do work with leopard and tiger so why is this definition there? should i have changed it?