lpc4/lib/
lpc4/lib/doc/efun/
lpc4/lib/doc/lfun/
lpc4/lib/doc/operators/
lpc4/lib/doc/simul_efuns/
lpc4/lib/doc/types/
lpc4/lib/etc/
lpc4/lib/include/
lpc4/lib/include/arpa/
lpc4/lib/obj/d/
lpc4/lib/save/
lpc4/lib/secure/
lpc4/lib/std/
lpc4/lib/std/living/
A lot of calls to xalloc() should be replaced with alloca().

Only the first two arguments to efuns have their types checked, and only
at run time. Not at compile time. The data structure should be enhanced.

What happens to user when an error occurs in create() ?

The people demand more runtime configuration possibilites.

The compiler should use my dynamic_buffer routines instead of the mem_block
thingies that are there now.

the documentation needs updating

does the loop -> once opimization work correctly?

Doc on dbase efuns is wrong.

parent->opt_info is used, but not yet computed in optimze

string refs seems to go overboard when a program fails to compile
  (Or at least my debug routines says so)

Is it possible to have a local variable with the same name as a function?

the nodes in las.c should probably hold svalues all the time.

/* more optimizations */
/*
	(a+a) => a<<1
	a*x+a*y => (x+y)*a
*/

valid_override isn't used yet