I was notified of some odd errors compiling Mac::Carbon for Panther, along
the lines of:
In file included from
/System/Library/Frameworks/CoreServices.framework/Frameworks/CarbonCore.frame
work/Headers/CarbonCore.h:113,
from
/System/Library/Frameworks/CoreServices.framework/Headers/CoreServices.h:21,
from
/System/Library/Frameworks/Carbon.framework/Headers/Carbon.h:20,
from /Developer/Headers/FlatCarbon/Events.h:1,
from ../Carbon.h:47,
from AppleEvents.xs:50:
/System/Library/Frameworks/CoreServices.framework/Frameworks/CarbonCore.frame
work/Headers/Debugging.h:285:2: #else without #if
/System/Library/Frameworks/CoreServices.framework/Frameworks/CarbonCore.frame
work/Headers/Debugging.h:287:2: #endif without #if
/System/Library/Frameworks/CoreServices.framework/Frameworks/CarbonCore.frame
work/Headers/Debugging.h:301:1: missing binary operator before token "enum"
AppleEvents.c:2578:1: unterminated #if
*** Error code 1
Weird. Turns out DEBUG in Debugging.h is conflicting with DEBUG in perl.
This is not a new thing, really; in Mac::Carbon and MacPerl we do an #undef
Move before #include'ing the Mac headers; so I added an #undef DEBUG (to
Carbon.h and AppleEvents/PerlAEUtils.h) and it seemed happy.
I just thought I would mention it for anyone who cares or has thoughts.
Also, in my machine here, one of the Speech tests fails because I count the
# of voices on disk and the number loaded, and the Good News voice is not
being loaded (do perl -Mblib -x Speech/eg/DumpVoices.plx), though it is on
disk (ls /System/Library/Speech/Voices/).
Also, another test or few in Mac::Notification fails, apparently -- in part
-- because the Notification Manager behavior changed in Panther (now it is a
separate process, so the old method of checking if the front process has
changed now fails miserably).
Also, in the Notification test, LaunchApplication() appears to be returning
bad values in Panther; after successfully launching an app, I get an -50
error code, and nothing in the LaunchParamBlock. The app is actually
launching, so I suspect it *shouldn't* be returning an error. That's what I
get for checking error values. ;-)
Oh, and for the curious, no, the load times of Mac::Carbon with Panther and
perl 5.8.1 and gcc 3.3 are not magically faster.
--
Chris Nandor [EMAIL PROTECTED] http://pudge.net/
Open Source Development Network [EMAIL PROTECTED] http://osdn.com/
