handle SIGFPE and allow the registered signal handler to do recovery
rather than bail
progs.h, pr_exec.c, pr_load.c:
if pr_faultchecks is 1 (0 is default), handle division by 0 gracefully
by loading the maximum representable number into the answer
Closes: #58
the rest:
kill the SIGFPE stuff
methods directly or setup the progs struct for a normal function call to
call a progs method rather than recursively calling PR_ExecuteProgram. This
will reduce method call overhead (both cpu and stack usage), fix the loss
of tracing when calling a method and makes it possible to break out of the
progs engine simply if threaded progs are ever implemented.
(if not empty) to be saved in the inputline history.
Clear the input (but put in history) of say or say team when pressing escape.
Don't say or say_team if the text is empty.
the SDL target files for Despair since he's working on the currently). Also
make progs segfault handling much more robust (and general: sigill etc too:)
functions out of bi_inputline.c. now somebody can be insane enough to
implement a qc controled console for the server ;) (more importantly, one
step closer to being able to have the server link with libQFcsqc (hah!))
call obj_find_message with super instead of class when doing a super
lookup
client_menu.[ch]:
use the InputLine object instead of directly using the api
InputLine.r:
builint functions are = #0, not = 0 :P
function.[ch]:
provide copy_params
method.[ch]
provide copy_keywordargs
expr.c:
call inc_users for the parm temp if it's a bind expression
revers a copy of the selector when building the selector name so the
selector doesn't get mangled
qc-parse.y:
catch erronous func = const inintializers (should be func = #const)
This gives us a bunch more flexibility. plugins no longer have the "lib"
prefix or the version suffix, they're now installed in
$fs_sharepath/QFplugins,, builds should take much less time (in general,
only one of pic or non-pic versions are build), bins and libs can have
individual CFLAGS
