PSEUDOCODE & OTHE

[chet.gottfried@xxxxxxxx]

XY-> T. Baehr:

XY-> Chet writes about adding support for (at least) the epsilon char. and the
 -> possibility of using @CNV to get from XPL code into something we can list
 -> here. Thanks! I'll see what I can do. I'll also see if I can add wildcar
 -> to the program. Anybody have problems with using the colon to surround
 -> function calls? Would any other character be better?

Why do you want to put colons around would-be function calls?
Consider a function call: two capital letters. You can either
test the pair against a string or allow an error. For example,

-------------------------------------------------------------

{lb-goof.pm, to see about function calls}{sx13,0}{sx14,0} {gl-first}


Above, two counters set to zero. Below, the labels first, second,
and third are merely repeats of what would be one line in the
actual program: putting function calls in place.


{lb-first}BC pfunc bcXC {if{er}}{sx13,{pv13}+1}{gl-oops}{ei}{sx14,{pv14}+1}{gl-second}

{lb-second}BC pfunc zzXC {if{er}}{sx13,{pv13}+1}{gl-oops}{ei}{sx14,{pv14}+1}{gl-third}

{lb-third}BC pfunc ccXC {if{er}}{sx13,{pv13}+1}{gl-oops}
{ei}{sx14,{pv14}+1}{gl-end}

{lb-end}BC mistakes = {pv13}, correct = {pv14}{ex}

{lb-oops}{gl-third}

The whole point is that one can attempt bogus function calls that
won't print, but as long as you have error control, you can keep
moving the XPL forward. For each bogus, instead of deleting, you
leave the text alone. (Of course, function calls in the program
but not yet implimented *will* print, so a string comparison to a
list of existing function calls might be better if more tedious.)

--Chet
---
 ? SLMR 2.1a ? Art + write + dtp = chet.gottfried@xxxxxxxx