Dialects 
Author Message
 Dialects

Hi

I would like to compile a list of PLI "dialects" as part of my PLI-2000 project.

I'd welcome  feedback from others on:

1) Number of distinct "dialects" that have been implemented and are in use.
2) Possibly size of user base for each dialect.

It might be useful for me to perhaps support more than one, possibly a compile time switch for dialect.

Thks

Hugh



Mon, 21 Mar 2005 10:26:07 GMT  
 Dialects
Most compilers I'm aware of follow the IBM "standard" - basically ANSI
PL/I with some stuff from ANSI Subset G (SELECT) plus extensions.  Of
course the two ANSI/ISO standards constitute two different dialects.
Multics PL/I is probably the most different.  Now Enterprise PL/I has
added a lot of stuff borrowed from C, some of it useful.
Quote:

> Hi

> I would like to compile a list of PLI "dialects" as part of my PLI-2000 project.

> I'd welcome  feedback from others on:

> 1) Number of distinct "dialects" that have been implemented and are in use.
> 2) Possibly size of user base for each dialect.

> It might be useful for me to perhaps support more than one, possibly a compile time switch for dialect.

> Thks

> Hugh



Mon, 21 Mar 2005 18:53:30 GMT  
 Dialects

Quote:

> Hi

> I would like to compile a list of PLI "dialects" as part of my PLI-2000 project.

> I'd welcome  feedback from others on:

> 1) Number of distinct "dialects" that have been implemented and are in use.
> 2) Possibly size of user base for each dialect.

In addition to the Multics dialect, a computer science student (whose
name eludes me but it's in Robin's PL/I FAQ) designed a set of object-
oriented extensions for PL/I.

I think parsing these extensions is likely to be the easiest
part of adding them to your compiler. But it might be a good
place to start...

-dq



Tue, 22 Mar 2005 03:38:10 GMT  
 Dialects

Quote:

> I would like to compile a list of PLI "dialects" as part of my PLI-2000 project.

In college about 9x10**8 years ago we used a "teaching" dialect called
PL/C, if I remember correctly. It was from the Univ. of Waterloo in the
Great White North, I think.

And there's also the IBM dialect in which much of their MVS operating
system used to be written, PL/S. I haven't touched a mainframe in ten
years so I don't know where it stands today, but I used to spend lots of
time reading the microfiche, refitting those dreaded "local mods".
....back when men were men and mainframes were hacked with vigor.
:-)

Quote:
> Thks
> Hugh

* Nick Geovanis
| IT Computing Svcs      Computing's central challenge:
| Northwestern Univ          How not to make a mess of it.

+------------------->


Tue, 22 Mar 2005 05:40:51 GMT  
 Dialects
Thanks for the info.

It seems we have:

1) IBM "Mainframe" PLI (Is the ANSI 76 standard or a variation?)
2) PL/S
3) PL/C
4) PL/I + OO Extensions
5) Multics PL/I (Is this same as Stratus (ie VOS)  PL/I ?)
6) ANSI 1981
7) ANSI 1987
8) Enterprise (?)
9) DEC VAX PL/I

I believe that there is not in fact; any implementation of the 1976 ANSI standard, IBM's is (I think) a variation,
despite the fact that they drove the definition of the 1976 standard.

I used to use IBM PL/I on a sys 360 (or 370, I cant recall).

I know it was very powerful, controlled storage, cross sections of arrays etc, features that are absent from the others.

To my knowledge IBM Mainframe PL/I is the only implementation that supports controlled storage etc, although their PC
offering may do I really dont know.

Any further info on all this is much appreciated, I want to pin down the feature set of my compiler, I need some
reference, currently is the 1987 standard + a few bits n pieces.

Thks
Hugh


Quote:

>> I would like to compile a list of PLI "dialects" as part of my PLI-2000 project.

>In college about 9x10**8 years ago we used a "teaching" dialect called
>PL/C, if I remember correctly. It was from the Univ. of Waterloo in the
>Great White North, I think.

>And there's also the IBM dialect in which much of their MVS operating
>system used to be written, PL/S. I haven't touched a mainframe in ten
>years so I don't know where it stands today, but I used to spend lots of
>time reading the microfiche, refitting those dreaded "local mods".
>....back when men were men and mainframes were hacked with vigor.
>:-)

>> Thks
>> Hugh

>* Nick Geovanis
>| IT Computing Svcs      Computing's central challenge:
>| Northwestern Univ          How not to make a mess of it.

>+------------------->



Tue, 22 Mar 2005 07:36:42 GMT  
 Dialects
PL/S is about as close to PL/I as Rexx is, in the other direction.  I'd
suggest not considering it.

PL/C is basically a somewhat stripped-down PL/I subset (no CONTROLLED
storage, no Record I/O, IIRC).  What made it different were the
debugging features and the quality of the diagnostics.

I think VOS PL/I and Multics PL/I are close, at least the compilers came
from the same source (TSI).

The manuals for IBM's Enterprise (aka VisualAge) PL/I and "classic" (OS
PL/2 2.3, PL/I for MVS and VM 1.1) are available online.  I'd really
like to see a manual for Multics PL/I, but lots of source is available.

Quote:

> Thanks for the info.

> It seems we have:

> 1) IBM "Mainframe" PLI (Is the ANSI 76 standard or a variation?)
> 2) PL/S
> 3) PL/C
> 4) PL/I + OO Extensions
> 5) Multics PL/I (Is this same as Stratus (ie VOS)  PL/I ?)
> 6) ANSI 1981
> 7) ANSI 1987
> 8) Enterprise (?)
> 9) DEC VAX PL/I

> I believe that there is not in fact; any implementation of the 1976 ANSI standard, IBM's is (I think) a variation,
> despite the fact that they drove the definition of the 1976 standard.

> I used to use IBM PL/I on a sys 360 (or 370, I cant recall).

> I know it was very powerful, controlled storage, cross sections of arrays etc, features that are absent from the others.

> To my knowledge IBM Mainframe PL/I is the only implementation that supports controlled storage etc, although their PC
> offering may do I really dont know.

> Any further info on all this is much appreciated, I want to pin down the feature set of my compiler, I need some
> reference, currently is the 1987 standard + a few bits n pieces.

> Thks
> Hugh



> >> I would like to compile a list of PLI "dialects" as part of my PLI-2000 project.

> >In college about 9x10**8 years ago we used a "teaching" dialect called
> >PL/C, if I remember correctly. It was from the Univ. of Waterloo in the
> >Great White North, I think.

> >And there's also the IBM dialect in which much of their MVS operating
> >system used to be written, PL/S. I haven't touched a mainframe in ten
> >years so I don't know where it stands today, but I used to spend lots of
> >time reading the microfiche, refitting those dreaded "local mods".
> >....back when men were men and mainframes were hacked with vigor.
> >:-)

> >> Thks
> >> Hugh

> >* Nick Geovanis
> >| IT Computing Svcs      Computing's central challenge:
> >| Northwestern Univ          How not to make a mess of it.

> >+------------------->



Tue, 22 Mar 2005 08:49:29 GMT  
 Dialects



Quote:
>1) Number of distinct "dialects" that have been implemented and are
>in use.

I'll only list the IBM dialects:

 1. PL/I (F) - obsolete but has some use

 2. OS "Optimizing" Compiler V1 - obsolete but has some use

 3. OS "Optimizing" Compiler V2 - obsolete but has some use

 4. "PL/I for this and that"; it went through about three names,
    with new language being added at each release and name
    change - obsolete but has some use

 5. Visual Age PL/I - obsolete but has some use

 6. Enterprise PL/I - current

--
     Shmuel (Seymour J.) Metz, SysProg and JOAT
     Atid/2, Team OS/2, Team PL/I

Any unsolicited commercial junk E-mail will be subject to legal
action.  I reserve the right to publicly post or ridicule any
abusive E-mail.

I mangled my E-mail address to foil automated spammers; reply to
domain Patriot dot net user shmuel+news to contact me.  Do not



Tue, 22 Mar 2005 12:46:16 GMT  
 Dialects
VA PL/I (AIX, NT, OS/2, and now MVS) supports controlled storage, cross
sections of arrays, and other features absent from many other subset-based
compilers. However, iSUB support is currently limited to arrays of scalars.


Quote:
> Thanks for the info.

> It seems we have:

> 1) IBM "Mainframe" PLI (Is the ANSI 76 standard or a variation?)
> 2) PL/S
> 3) PL/C
> 4) PL/I + OO Extensions
> 5) Multics PL/I (Is this same as Stratus (ie VOS)  PL/I ?)
> 6) ANSI 1981
> 7) ANSI 1987
> 8) Enterprise (?)
> 9) DEC VAX PL/I

> I believe that there is not in fact; any implementation of the 1976 ANSI

standard, IBM's is (I think) a variation,
Quote:
> despite the fact that they drove the definition of the 1976 standard.

> I used to use IBM PL/I on a sys 360 (or 370, I cant recall).

> I know it was very powerful, controlled storage, cross sections of arrays

etc, features that are absent from the others.
Quote:

> To my knowledge IBM Mainframe PL/I is the only implementation that

supports controlled storage etc, although their PC
Quote:
> offering may do I really dont know.

> Any further info on all this is much appreciated, I want to pin down the

feature set of my compiler, I need some
Quote:
> reference, currently is the 1987 standard + a few bits n pieces.

> Thks
> Hugh

> On Thu, 3 Oct 2002 16:40:51 -0500, Nicholas Geovanis


> >> I would like to compile a list of PLI "dialects" as part of my PLI-2000
project.

> >In college about 9x10**8 years ago we used a "teaching" dialect called
> >PL/C, if I remember correctly. It was from the Univ. of Waterloo in the
> >Great White North, I think.

> >And there's also the IBM dialect in which much of their MVS operating
> >system used to be written, PL/S. I haven't touched a mainframe in ten
> >years so I don't know where it stands today, but I used to spend lots of
> >time reading the microfiche, refitting those dreaded "local mods".
> >....back when men were men and mainframes were hacked with vigor.
> >:-)

> >> Thks
> >> Hugh

> >* Nick Geovanis
> >| IT Computing Svcs      Computing's central challenge:
> >| Northwestern Univ          How not to make a mess of it.

> >+------------------->



Tue, 22 Mar 2005 13:37:40 GMT  
 Dialects


10/03/2002

Quote:
>In college about 9x10**8 years ago we used a "teaching" dialect
>called PL/C, if I remember correctly. It was from the Univ. of
>Waterloo in the Great White North, I think.

Cornell. You're thinking of, e.g., WATFOR, WATFIV.

Quote:
>And there's also the IBM dialect in which much of their MVS
>operating system used to be written, PL/S.

Neither the first not the last of the series. Off the top of my head,
BSL -> PL/S -> PL/S II -> PL/AS -> PL/X. It includes imbedded
assembler and a nonstandard but useful substr(string,first:last)
syntax. It also has useful extensions to DECLARE.

--
     Shmuel (Seymour J.) Metz, SysProg and JOAT
     Atid/2, Team OS/2, Team PL/I

Any unsolicited commercial junk E-mail will be subject to legal
action.  I reserve the right to publicly post or ridicule any
abusive E-mail.

I mangled my E-mail address to foil automated spammers; reply to
domain Patriot dot net user shmuel+news to contact me.  Do not



Tue, 22 Mar 2005 13:10:31 GMT  
 Dialects

Quote:
> Thanks for the info.

> It seems we have:

> 1) IBM "Mainframe" PLI (Is the ANSI 76 standard or a variation?)
> 2) PL/S
> 3) PL/C
> 4) PL/I + OO Extensions
> 5) Multics PL/I (Is this same as Stratus (ie VOS)  PL/I ?)
> 6) ANSI 1981
> 7) ANSI 1987
> 8) Enterprise (?)
> 9) DEC VAX PL/I

Regarding PL/C:  that was supposed to have about 1/3 the resources of full
PL/I-F.  Also, at the university of Toronto 20 years ago, Professors Hume
and Holt had an even smaller subset (about 10% the resources of the full
language) known as SP/k, used to teach those who were learning computer
programming for non-professional intentions.


Wed, 23 Mar 2005 04:29:12 GMT  
 Dialects
What about PL/M? Seems like it was an ourtgrowh of PL/I

On Fri, 4 Oct 2002 16:29:12 -0400, "Alexander Damyanovich"

Quote:



>> Thanks for the info.

>> It seems we have:

>> 1) IBM "Mainframe" PLI (Is the ANSI 76 standard or a variation?)
>> 2) PL/S
>> 3) PL/C
>> 4) PL/I + OO Extensions
>> 5) Multics PL/I (Is this same as Stratus (ie VOS)  PL/I ?)
>> 6) ANSI 1981
>> 7) ANSI 1987
>> 8) Enterprise (?)
>> 9) DEC VAX PL/I

>Regarding PL/C:  that was supposed to have about 1/3 the resources of full
>PL/I-F.  Also, at the university of Toronto 20 years ago, Professors Hume
>and Holt had an even smaller subset (about 10% the resources of the full
>language) known as SP/k, used to teach those who were learning computer
>programming for non-professional intentions.



Wed, 23 Mar 2005 04:42:11 GMT  
 Dialects
I would say PL/M is closer to PL/S than PL/I.  Also, besides SP/k
there's PL/zero, another subset.  
Quote:

> What about PL/M? Seems like it was an ourtgrowh of PL/I

> On Fri, 4 Oct 2002 16:29:12 -0400, "Alexander Damyanovich"



> >> Thanks for the info.

> >> It seems we have:

> >> 1) IBM "Mainframe" PLI (Is the ANSI 76 standard or a variation?)
> >> 2) PL/S
> >> 3) PL/C
> >> 4) PL/I + OO Extensions
> >> 5) Multics PL/I (Is this same as Stratus (ie VOS)  PL/I ?)
> >> 6) ANSI 1981
> >> 7) ANSI 1987
> >> 8) Enterprise (?)
> >> 9) DEC VAX PL/I

> >Regarding PL/C:  that was supposed to have about 1/3 the resources of full
> >PL/I-F.  Also, at the university of Toronto 20 years ago, Professors Hume
> >and Holt had an even smaller subset (about 10% the resources of the full
> >language) known as SP/k, used to teach those who were learning computer
> >programming for non-professional intentions.



Wed, 23 Mar 2005 07:03:11 GMT  
 Dialects

Quote:

> I know it was very powerful, controlled storage, cross sections of arrays etc, features that are absent from the others.

> To my knowledge IBM Mainframe PL/I is the only implementation that supports controlled storage etc, although their PC
> offering may do I really dont know.

I used IBM mainframe compilers (F, Optimizing) for 30 years and I own Personal
PL/I for OS/2.  Those are the only implementations i have direct knowledge of
and documentation for.  I believe but am not certain that Personal PL/I is
essentially the same as some earlier (not current) version of VA PL/I without
the facilities to easily interface with mainframe products such as DB2.  It
seems to have most if not all of the language features that I have seen
mentioned in posts about VA PL/I on this newsgroup.

Personal PL/I certainly has controlled and based storage, cross sections and all
the other familiar language features present in the F and Optimizing compilers.
  The workstation PL/I products have a number of language extensions (such as
type definitions, packages, additional ways of defining data values, and many
new built in functions) beyond the language recognized by the Optimizing
compiler.  (Most if not all of these are also in the current mainframe version I
believe).
Needless to say I have not had occasion to try them all in personal PL/I, but
everything I have tried has been there.

So far I know of only three areas in which Personal PL/I is less capable that
the Optimizing Compiler:

The maximum floating point precision is 80 instead of 109.  This, of course, is
due to the capabilities of the underlying hardware and is to be expected.

Aggregates may not be passed as arguments if data type differences between
argument and parameter would require the creation of a temporary copy.  This is
undoubtedly due to implementor laziness.  On the other hand since it is seldom
if ever a good idea it is not much of an omission.

The multitasking capabilities have been considerably altered.  In particular:

It is not possible to attach tasking options to IO statements.  A new task or
thread is initiated by the ATTACH statement instead of simply adding one or more
tasking options to a CALL statement.  The procedure invoked must be non nested
(i.e., level 1 or external).  It appears that only a single by value pointer can
be passed.  This is not completely clear because a few lines down the
documantation says that arguments (plural) may be passed to the thread in the
usual way.  I have not done an experiment to find out which sentence is correct.

The EVENT variable has been eliminated and the WAIT statement can only WAIT on
task (thread) completion rather that on task completion, IO operation
completion, or arbitrary programmer defined events. Although it is still
possible to share data between tasks (threads) the programmer is on his own and
must accomplish this using OS/2 API calls.  In other words, synchronization is
no longer integrated into the language, a great leap backwards.

I believe that most of these statements are also true of the VA compiler but
having no access to documentation, I am not sure.



Wed, 23 Mar 2005 08:22:09 GMT  
 Dialects
Unrelated, except for the similarity of the first two letters of the name.

Quote:
> What about PL/M? Seems like it was an ourtgrowh of PL/I

> On Fri, 4 Oct 2002 16:29:12 -0400, "Alexander Damyanovich"



> >> Thanks for the info.

> >> It seems we have:

> >> 1) IBM "Mainframe" PLI (Is the ANSI 76 standard or a variation?)
> >> 2) PL/S
> >> 3) PL/C
> >> 4) PL/I + OO Extensions
> >> 5) Multics PL/I (Is this same as Stratus (ie VOS)  PL/I ?)
> >> 6) ANSI 1981
> >> 7) ANSI 1987
> >> 8) Enterprise (?)
> >> 9) DEC VAX PL/I

> >Regarding PL/C:  that was supposed to have about 1/3 the resources of
full
> >PL/I-F.  Also, at the university of Toronto 20 years ago, Professors Hume
> >and Holt had an even smaller subset (about 10% the resources of the full
> >language) known as SP/k, used to teach those who were learning computer
> >programming for non-professional intentions.



Thu, 24 Mar 2005 15:43:39 GMT  
 Dialects



Quote:
> Aggregates may not be passed as arguments if data type differences between
> argument and parameter would require the creation of a temporary copy.

This was resolved in later versions of VA PL/I (and probably in the current
Personal PL/I).


Thu, 24 Mar 2005 15:45:47 GMT  
 
 [ 27 post ]  Go to page: [1] [2]

 Relevant Pages 

1. Simple architecture for mixins for all Smalltalk dialects

2. Regional Dialects

3. Dialects & debates (was Re: (no subject))

4. Can any Smalltalk dialects produce stand-alone DLLs?

5. New Smalltalk dialect prototype

6. Dialect in comp.lang.apl

7. Dialects in comp.lang.apl

8. Syntax differences between ST dialects

9. Question about Dialects, and the VM

10. Dialect/Version detection

11. All ST dialects - web access to live workspace/transcript/class browser

12. San Francisco, Internet asp, any dialect OK

 

 
Powered by phpBB® Forum Software