Alex Smith, 26 March 2008

syslibc.c is an example of a C program designed to be used with the
INTERCAL external call code. It only attempts to emulate the base 2
algorithms, but in a base 2 program it should be a valid drop-in
replacement for the real system library.  (Don't forget to use the -eE
options when linking this to an INTERCAL program; -e to allow C
programs to be linked in, and -E to suppress the 'real' syslib.)

The library is designed foremost to ensure that all errors are checked
(as they would be in the INTERCAL, although you get the error messages
produced by C code), and secondarily for speed (it is substantially
faster than the INTERCAL version of the system library, but that isn't
particularly surprising).

Note that some INTERCAL programs rely on undocumented features of the
system library; I have implemented the commonly-used (1001), but some
of the example programs (for instance primes.i) NEXT to other
undocumented internal lines of syslib.i, and such programs will error
out on the C version of the syslib.

There are some situations in which the C program will cause errors,
but the INTERCAL version wouldn't; remember that any attempt to alter
or view an INTERCAL variable from C code causes an error if that
variable doesn't exist in the INTERCAL program. Therefore, in order to
use a particular routine, you should mention all the variables that
are its inputs and outputs somewhere in your INTERCAL program; in
particular, if you're using overflow-checked routines, make sure you
mention .4 at least once in your program.