SCOUG-Programming Mailing List Archives

<< Previous Message << >> Next Message >>

Date:	Mon, 16 Feb 2004 14:17:40 PST8
From:	Peter Skye <pskye@peterskye.com >
Reply-To:	scoug-programming@scoug.com
To:	scoug-programming@scoug.com
Subject:	SCOUG-Programming: PL/I most important statement (was: call(TZ) ? - PL/I)

Content Type: text/plain

Lynn H. Maxson wrote:
>
> obviously they couldn't get it to
> work and the project was dropped.

No, I'm sorry, it's not obvious that they couldn't get it to work.
There are many reasons for dropping a project.

> It was to use Fortran . . . REXX and APL assign data
> attributes, e.g. string or arithmetic, dynamically
> based on its use in an assignment statement.

No. Rexx treats everything as a string.

> Begin-end of PL/I would allow [dynamic
> attributes], Fortran has no such equivalent.

This has nothing to do with writing an APL compiler or interpreter in
Fortran.

It _does_ matter if you're trying to translate APL to Fortran.

Are you confusing compilation, interpretation, and translation?

> You don't have to get too deep into the parse verb
> in REXX before you're way over what Fortran supports.

Luckily for us Fortran is a programming language, not a library of
pre-formulated routines. Thus a programmer can write, in Fortran, a
parse routine. Surely Fortran has some sort of substring function.

> APL supports intermixing arithmetic, string, and
> logical operators in an expression. Fortran does not.

Why do you feel this prevents a Fortran program from processing an APL
source file?

If you go back about 50 years, Dr. Turing demonstrated the most feeble
of digital machine designs and then proved it could do anything a more
robust machine could do.

> As SL/I syntax follows the same rules as PL/I you can
> write a PL/I parser in any manner you desire. Frankly I
> would probably start with understanding how it is done
> in LPEX (Live Parsing EXecutive) that comes with your
> IBM PL/I compiler. For my money LPEX is the starting
> point for developing the Developer's Assistant, which
> is editor plus syntax analysis plus semantic analysis
> plus logical organizer plus interpretive or compile
> code generator plus automated test data generator.

This would be a fascinating presentation for the entire SCOUG audience.
Lynn, can _you_ speak on this sometime in the near future? You
suggested Steven, but SL/I and the Developer's Assistant is _your_
concept. The general interest here is that it can be done with existing
tools, inexpensively, within a moderate time frame -- the Programming
SIG is the better venue for discussing the actual implementation.

Your presentation outline might be:

1. Overview of Developer's Assistant
2. Overview of SL/I
3. Available tools.
a. LEX
b. YACC
c. LPEX
4. Creating the working environment.
a. SCOUG's Programming SIG
b. Open Source
c. Remote server tools such as CVS
5. The dollar costs for the project
6. Projected man-hours for the project

So I'm asking you: Can you speak once again on SL/I and the Developer's
Assistant, with the thrust being how LEX, YACC and LPEX could jointly be
used as the tools for this project?

- Peter

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

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.