[comp.lang.apl] Pascal Source for APL Interpreter--Want Info

crichmon@digi.lonestar.org (Charles Richmond) (10/03/90)

I have a book entitled _The Byte Book of Pascal_ that was published
about 12 years ago.  It contains an APL interpreter written entirely
in Pascal (for the CDC6600, I think).  It is somewhere in the neighbor-
hood of 50 pages long.  I typed it all into my Atari ST and now cannot
get it to work.

I suspect that there were errors or omissions in the published listing.
If anyone knows of this program, maybe they can tell me where to get
a list of the fixes for the code in this book.  (Maybe some old issue
of Byte magazine has a list of errors?)

Even better, if some of the authors of this code are net readers . . .
I would appreciate hearing from you very much.  I could tell you what
changes I made to the code for the Atari ST.

(I would write to Byte magazine, but I have had little luck with this
approach in the past on other subjects . . . any Byte magazine people
care to comment on that?)

Thanks much for your help in advance!


   Charles Richmond              crichmon@digi.lonestar.org

sandee@sun16.scri.fsu.edu (Daan Sandee) (10/03/90)

In article <1057@digi.lonestar.org> crichmon@digi.lonestar.org (Charles Richmond) writes:
>I have a book entitled _The Byte Book of Pascal_ that was published
>about 12 years ago.  It contains an APL interpreter written entirely
>in Pascal (for the CDC6600, I think).  It is somewhere in the neighbor-
>hood of 50 pages long.  I typed it all into my Atari ST and now cannot
>get it to work.
>
>   Charles Richmond              crichmon@digi.lonestar.org

Well, there's Pascal and Pascal ... and 1970's style CDC Pascal and current
Atari is about as far apart as you can get. Although Pascal is designed as
a machine-independent language, anybody can write a machine-dependent program.
If it was published 12 years ago, it probably indeed was CDC 6000 Pascal.
FYI, the 6000 had 6-bit characters packed 10 to a 60-bit word. Character set
not ASCII but CDC's own. So how about mixing integers and characters through
abuse of chr() and ord()? And packed data types? And records with conditional
clauses of different data type (resulting in different memory allocations)?
And 'set' data type limited to 60 elements? Plenty of opportunity for bad
programming there.