Re: Thoughts on MF COBOL, Unix, and RDBMS

"Richard" <riplin@[EMAIL PROTECTED]
> wrote in message 
news:c831a358-1634-4d38-8f91-7dc76c570638@[EMAIL PROTECTED]
 May 5, 11:03 am, "James J. Gavan" <jgavandeletet...@[EMAIL PROTECTED]
> wrote:
> Pete Dashwood wrote:
> > I have some programs written in Fujitsu COBOL (generated by a tool)
that 
> > run
> > in a Windows environment, read ISAM and write a RDB using embedded SQL

> > and
> > ODBC.
> > <snip>
>
> Caveat - other than being a name I haven't a clue what C-ISAMs are.
>
> Pete : You *think* the compiler *might* be Microsoft rebadged MF COBOL
:-
>

That's not what he said, he said it may be [a Unix version of] the
same compiler that [the DOS/Windows/OS2 version] was rebadged by
Microsoft. ie MF Cobol/2 version 3.x.


> (a) Are you dealing with an end-user who only knows the bare bones, and
> whether amicably or for whatever other reason, has parted company with
> the contractors who wrote the application; or,
>
> (b) They do have programmers
>
> Both (a) and (b) above could deny the compiler doesn't have SQL -
> because it was never used. (See explanation below).
>
> Unfortunately I chucked my copies of Microsoft COBOL Versions 4 and 5
> away ages ago, but I still have diskettes and manuals for Micro Focus
> (DOS) Version 3.1.
>
> Can you establish which compiler they have :-
>
> - does the term 'Programmer's Workbench' ring bells with their
programmers
> - I think when the application starts up it should contain a banner
> saying 'Microsoft COBOL Version .....'
>
> Back to Micro Focus (DOS) Version 3.1. If you asked me, "Does it have
> SQL" my response would have been "No" - FALSE. It does. Just that I
> wasn't aware of it, because it's not whistles and bells like Net
Express.
>
> Briefly, this is what V 3.1 says about SQL :-
>
> -----------------------------------------------------------------------
> Programs can be created using SQL to access DBs that comply with IBM's
> SQL Precompiler Services Architecture
>
> - IBM - IBM OS/2 Extended Edition DB Manager - refer to IBM
do***entation
>
> - Microsoft SQL Server with Embedded SQL for COBOL Developers Kit -
> refer to Microsoft for do***entation
>
> The sup****t is dependent on various system directive (some 17), all of
> which begin "SQL".
> -----------------------------------------------------------------------

You will notice the requirement for a precompiler. You will notice
that the 'SQL' directive lists some of the database engines but does
not indicate that these are actually included with the compiler.

> Whether the ODBC drivers are obtained from IBM or MS I don't know. As
> there are just two, they *might* be included in the Version 3.1 package.

No. They were not.


> To provide increased compatibility M/F provides the following extensions
> - INCLUDE, DECLARE TABLE, UNDECLARED HOST VARIABLES, QUALIFIED HOST
> VARIABLES, HOST VARIABLE GROUPS AND INDICATOR ARRAYS, THE NOT OPERATOR
> (symbol); the symbol used is the top right corner of a box.

These are not in the Cobol/2 version 3.2 manual so I doubt they are in
3.1 (or 3.4 the last 3.x).

> Back to Microsoft. I switched from RM/COBOL V 1.? with Radio Shack to a
> Compaq PC with Microsoft COBOL Version 3. DEFINITELY a Microsoft
> product, regardless of who may have written the compiler for them. The
> manuals were written in plain English - nothing like the gobbledy-gook
> from the Committee for Programming Language 22, Version 4.

Microsoft versions 1 through 3 were written by Microsoft, originally
for CP/M in about 1978.

Microsoft version 4.0 was a rebadged MicroFocus Cobol/2 version 3.0
with MS 4.5 being MF 3.1 and MS 5.0 being MF 3.2.

> I upgraded to MS Versions 4 and 5, which as Richard pointed out years
> ago were re-badged M/F. The LRMs were lifted from the text of PL22.4,
> well under one of its previous names. Essentially the only difference is
> that the "IDE" feature was Microsoft, viewing things through the
> Programmers Workbench and programs were compiled/linked using the MS
> Makefile feature. As V 4.0 was released MS said "Don't use Comp-5 to
> store data". Then when they issued V 5.0 that warning became apparent -
> the package included a glossy from an organisation called 'Micro
> Focus',(whom I had never heard of), where at a price you could switch
> from MS to MF using MF Version (DOS) 3.1.
>
> Now it is a BIG ASSUMPTION - if MF 3.1 has SQL, I think it's a fair bet
> that MS Versions 4 and 5 also had that feature. If the user has the
> diskettes for the compiler, without actually re-installing, it should be
> possible, going through the Setup program, which allows you to select
> modules you want, to determine whether or not there is a module for SQL.

While MF 3.1 'sup****ts' SQL it doesn't include it.


> I don't think Net Express Open ESQL would help you. Granted it's a menu
> selection item in the N/E IDE, and although a separate sub-component

See, even the much later N/E has SQL as a sub-component and only
includes one of these.


> it
> must be fairly integrated with N/E, particularly the compiler to be able
> to accept the SQL family of RESERVED WORDS. It is not a separate module,
> but part of the whole N/E package.

So, it is bundled. MF 3.x came as various 'bundlings', the cheapest
was just the compiler (which is all I bought), next was the Compiler +
Toolset', then 'Workbench'.

To increase revenue they later only sold the 'Workbench'.

> If you can establish the name/version of the compiler, then I would
> think that initially writing CSVs would be the way to go; then update a 
> DB.
>
> Bill - can you add anything relative to how close MS Versions 4.0 and
> 5.0 are to MF Version 3.?

They dropped OS/2 sup****t in 5.0, and had PWB (yuck) instead of
Toolset (yuck), but some tools made it through. For example XM was
freely distributable in MS but there was no 'run-time' runcob so
programs had to be built as .EXE or .DLE and could not be .int
or .gnt. Inside thge .EXE they could be OPT(0) though.


> Richard - we all know you are a pack-rat and don't throw anything away
> :-). Did you use MS V 4.0 or 5.0 and have you got the manuals/diskettes

Yes, of course, but that is irrelevant.

> ? Can you expand/clarify on what I have written ?

You mistook what Peter said and went downhill from there.

[Pete]

Yes, there was some confusion, but I did find Jimmy's comments interesting

and I appreciate that he made them.

Your own knowledge of the history of this stuff never ceases to amaze me, 
Richard :-), and I found your comments above useful too.

We are fortunate to have some very clever people contributing to this 
Newsgroup.

As someone who normally works alone these days (at least on technical
stuff) 
it is really good to be able to gather other opinions and comments.

Even if it isn't directly pertinent, such interaction often sparks another

idea.

Pete.
-- 
"I used to write COBOL...now I can do anything."





and solutions.