SCOUG-Programming Mailing List Archives

<< Previous Message <<

Date:	Mon, 16 Feb 2004 15:46:12 PST8
From:	"Lynn H. Maxson" <lmaxson@pacbell.net >
Reply-To:	scoug-programming@scoug.com
To:	< "scoug-programming@scoug.com" > scoug-programming@scoug.com >
Subject:	SCOUG-Programming: PL/I most important statement (was: call(TZ) ? - PL/I)

Content Type: text/plain

"...Are you confusing compilation, interpretation, and
translation?..."

Believe me I wasn't confused before we started this thread.
Nowadays I'm not sure. When someone says they are going to
write a compiler for APL in Fortran it means they are going to
process APL source like the APL interpreter, but instead of
interpretive output produce compiled output. Obviously
translation has to occur as it occurs in every compiler in order
to produce machine language executables. Even in something
like the APL interpreter translation occurs.

You are correct in that many reasons exist for a project to
fail. When I initially heard the announcement I said it would
fail. I had reasons for believing that, some of which I shared
with you. It could very well in fact have failed for other
reasons, but it would have eventually failed for the reasons I
offered.

The point is that APL is more than a language. It's also a
computing environment. It has its own idiosyncrasies which
contribute to its uniqueness. REXX is similar. You cannot
write an APL interpreter in APL nor REXX in REXX. You can
translate C into PL/I, but only a limited subset of PL/I into C.
There are certain C expressions that don't have an identical
form in PL/I. For example, you can invoke a procedure in C as
a single statement without return code. You can't in PL/I. I
won't argue why you would want to do so, but the fact is
that you can.

LEX and YACC stand on their own as does LPEX. If people
want to have presentations on them in the Programming SIG,
we will arrange it. LPEX and probably any other so-called
"smart" editor offers an "initial" user interface to the
Developer's Assistant. It offers an example of a few of the
things that should be supported.

As SL/I is based on PL/I syntax and a superset of it the initial
implementation will be in PL/I in order to reuse most of the
source later. SL/I is written in SL/I. That includes syntax
analysis. LEX and YACC provide a method, not an
implementation. The same applies to any LALR2 syntax
analyser.

Frankly I would as soon put SL/I and the Developer's Assistant
on the Programming SIG backburner as it is so far out of sync
with mainstream open source. There are more things we
could do for the OS/2 community in the near term to keep it
viable for the long term when other things may or may not
occur. I have heard that the OS/2 version of PMMail will be
offered as open source on SourceForge. I think that would
offer us a more immediately worthwhile project to support for
the OS/2 community.

=====================================================

To unsubscribe from this list, send an email message
to "steward@scoug.com". In the body of the message,
put the command "unsubscribe scoug-programming".

For problems, contact the list owner at
"rollin@scoug.com".

=====================================================

Return to [ 16 | February | 2004 ]

SCOUG, Warp Expo West, and Warpfest are trademarks of the Southern California OS/2 User Group. OS/2, Workplace Shell, and IBM are registered trademarks of International Business Machines Corporation. All other trademarks remain the property of their respective owners.