Hi all.
A few more changes on CVS:
- Joseph Sarlo's GUI shadows are now controlled by -DGUISHADOW (except on MacOSX, where you're stuck with them, for now). I tried to change configure so that this is the default, but I can't test with a build. If someone who is more expert with configure and other meta-make systems could help out with a proper config option, it would be very appreciated.
- I made some aesthetic tweaks to the main window, following (in part) some feedback from people here. I'm sure you'll let me know if I got it all wrong.
- I introduced a new Tcl function in u_main.tk: accel_munge. It changes the default control-key shortcuts to command-key shortcuts that hopefully display and act correctly.
- I made further tweaks to the Mac OS X's display of shadows. Moving objects no longer screws them up.
- I slipped in the Mac OS X private name resolution flag that I discussed before. This needs to be tested, but appears to improve behavior with libraries with conflicting names (e.g. [scale] in GEM and maxlib). YMMV.
- With the deft adjustment of two #ifdefs, ~/.pdrc now works on the Mac (and other Unices). Options are introduced with a hyphen, and are one- per-line, like:
-verbose -path ~/Library/pd -path /Library/pd
As a reminder, you can grab the devel_0_36 branch with: cvs -d:pserver:anonymous@cvs.pure-data.sourceforge.net:/cvsroot/pure-data login
cvs -z3 -d:pserver:anonymous@cvs.pure-data.sourceforge.net:/cvsroot/pure- data co -r devel_0_36 pd
adam
Hi, Adam Lindsay hat gesagt: // Adam Lindsay wrote:
- I slipped in the Mac OS X private name resolution flag that I discussed
before. This needs to be tested, but appears to improve behavior with libraries with conflicting names (e.g. [scale] in GEM and maxlib). YMMV.
Currently pd-devel-cvs on linux is broken, and this changes *might* be the cause. First I had to change t_xxx makeout; to t_xxx makeout = NULL;
in src/s_loader.c. I'm not sure, if this is correct, but otherwise PD would crash as soon, as it tried to load any library.
But even now, loading libraries fails.
load_object: Symbol "iemlib1_setup" not found iemlib1: can't load library load_object: Symbol "iemlib2_setup" not found iemlib2: can't load library load_object: Symbol "iem_t3_lib_setup" not found iem_t3_lib: can't load library [...]
What might this be?
BTW: How can I check out yesterdays version: Currently I cannot run PD, because I overwrote my working package with the broken one, silly me... :(
ciao
Frank,
I'm 99.9% certain I didn't touch anything in s_loader except for a single line in the #ifdef MACOSX section. I'm away from my main computer, but you can check using the CVS web-browse pages.
The only thing linuxy I changed were the GUISHADOW macro notations.
adam
I believe Frank Barknecht said this around Thu, 28 Nov 2002:
Hi, Adam Lindsay hat gesagt: // Adam Lindsay wrote:
- I slipped in the Mac OS X private name resolution flag that I discussed
before. This needs to be tested, but appears to improve behavior with libraries with conflicting names (e.g. [scale] in GEM and maxlib). YMMV.
Currently pd-devel-cvs on linux is broken, and this changes *might* be the cause. First I had to change t_xxx makeout; to t_xxx makeout = NULL;
in src/s_loader.c. I'm not sure, if this is correct, but otherwise PD would crash as soon, as it tried to load any library.
But even now, loading libraries fails.
load_object: Symbol "iemlib1_setup" not found iemlib1: can't load library load_object: Symbol "iemlib2_setup" not found iemlib2: can't load library load_object: Symbol "iem_t3_lib_setup" not found iem_t3_lib: can't load library [...]
What might this be?
BTW: How can I check out yesterdays version: Currently I cannot run PD, because I overwrote my working package with the broken one, silly me... :(
ciao
Frank Barknecht _ ______footils.org__
PD-dev mailing list PD-dev@iem.kug.ac.at http://iem.kug.ac.at/cgi-bin/mailman/listinfo/pd-dev
-- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Adam Lindsay +44(0)1524 594 537 atl@comp.lancs.ac.uk http://www.comp.lancs.ac.uk/computing/users/atl/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Hi, Adam Lindsay hat gesagt: // Adam Lindsay wrote:
I'm 99.9% certain I didn't touch anything in s_loader except for a single line in the #ifdef MACOSX section. I'm away from my main computer, but you can check using the CVS web-browse pages.
The only thing linuxy I changed were the GUISHADOW macro notations.
Okay, now I found the mistake, and indeed it was the GUISHADOW define, in src/configure, which unintentionally changed the actual s_loader.c. Here's how it was done:
The affected lines in configure must not read:
MORECFLAGS=-DDL_OPEN -DGUISHADOW
but instead:
MORECFLAGS="-DDL_OPEN -DGUISHADOW"
In my case, this line wasn't read in by configure at all, I suppose, so DL_OPEN wasn't defined and thus I couldn't load any library, because the loader code was commented out...
Adding quotes fixes it.
Actually, I'm now going to remove -DGUISHADOW at all. ;)
ciao
On Thu, 28 Nov 2002, Frank Barknecht wrote:
BTW: How can I check out yesterdays version: Currently I cannot run PD, because I overwrote my working package with the broken one, silly me... :(
cvs update -D yesterday :)
Guenter
ciao
Frank Barknecht _ ______footils.org__
PD-dev mailing list PD-dev@iem.kug.ac.at http://iem.kug.ac.at/cgi-bin/mailman/listinfo/pd-dev
-
Adam Lindsay -
Frank Barknecht -
guenter geiger