tiny-cobol-users

[Tiny-cobol-users] Bugs in DISPLAY

[Hudson R. <hud...@so...>]

Hi all,

I've had problems in DISPLAY verb.

1. Using a DISPLAY verb with literals, if the string isn't terminated(if the double quotes are opened and don't closed), the scanner shows the following message: "scanner token too large". 
If the string need be continued in other line and don't have the I/O screen statements(as line/position statement), the scanner shows the same message above.
Example programs: disp02.cob, disp03.cob

2. Using the DISPLAY verb, if I insert the statements auto and update, the compiler not show errors in compilation. Should will show?
Example programs: disp04.cob.

3. Using the DISPLAY verb without any way of I/O screen, and after, use a ACCEPT verb using a I/O screen(as line/position statement), it don't show the values showed by display.
Example programs: disp05.cob.

Thanks.
Hudson


-- 
Hudson Reis                                       hud...@so...

[Tiny-cobol-users] Bugs in ACCEPT

[Hudson R. <hud...@so...>]

Hi all,

I've had problems in ACCEPT verb.

1. In execution, the no-echo statement don't work.
Example program: accept03.cob

2. If I don't use nothing type of statement in accept(auto, update, etc), or nothing type of screen I/O, in execution, isn't respected the size of picture and permit that the user type any character (numeric/alphanumeric,etc). Also, function keys don't work.
Example program: accept00.cob, accept01.cob

3. When I press Alt+TAB in any part from my program, it's exit from that operation. Why?

Thanks.
Hudson

Re: [Tiny-cobol-users] bugs in occurs <n> times indexed by <index > ?

[Hudson R. <hud...@so...>]

Hi all,

> > This program is a good start.  Binyamin has already pointed out the problem
> > with not initializing i.  There is another problem in that you may, at some
> > point, exceed the bounds of the table by specifying an additional offset.
> > 

You'll be correct. I'm have specified an additional offset in my table..

Sorry and thanks for your help.
Hudson

[Tiny-cobol-users] CGI interface

[<tb...@nf...>]

Hello.

I've already written in C a CGI interface library for Gnu fortran
compiler (g77) on Intel platform. May be you can grab some in it...

http://krabulator.free.fr/devel/cgi77.html

Thierry.

RE: [Tiny-cobol-users] Introduction and call for ideas

[<bko...@ma...>]

I don't think there is any project that it too small for SourceForge.
Besides, what better way to gain exposure and input?

If you are limiting this to a cgi interface, then you're right about GNUe
being less than useful.

I suspect the relevant thing for me would be a simple interface to access
the data string.  Since I have yet to do any CGI work, I could not make any
more specific comment.

Good Luck.

-----Original Message-----
From: Wim Niemans [mailto:ni...@pb...]
Sent: Friday, September 13, 2002 4:56 AM
To: Boris Kortiak
Subject: RE: [Tiny-cobol-users] Introduction and call for ideas


> Have you checked what is available at http://sourceforge.net yet?  Not
only
> are there a variety of open source projects going on, but it seems to be
an
> excellent tool for managing projects and keeping a repository of project
> information.  This is what TinyCOBOL uses.  

I think this project is too tiny for sourceforge. Maybe later, when a
roadmap is 
clear, and some things are coded already.

> You may want to also check out the gnue project at http://www.gnue.org to
> see what has already been done with respect to warehouse management and
> other enterprise software.  You may find some useful stuff, although it
will
> not be written in COBOL.

Well, I intend to build a cgi interface completely written in Cobol.
Though there is a xml parser there, written in cobol.

> Good luck and keep us informed about your project.

First of all, I will need some input to decide what is relevant to cobol 
programmers. I do not intend to write an interface for all that legacy code
out 
there. But the idea to have cobol as a cgi-option tickles me.
After all, a runtime cobol program has a much less footprint than perl, php
or 
other popular scripting languages.

Cheers,
wim niemans

Re: [Tiny-cobol-users] Tiny Cobol and CGI

[Wim N. <ni...@pb...>]

> Is there a way to use tiny cobol in a cgi-bin with
> data from a html form?  I have used rm\cobol by
> www.liant.com and this options is available.  It is
> very usefull and if its available please let me know.

Yes and No. I am using the ACU cobol compiler with a very nice interface.
The cgi-form is declared in working-storage as external, and can be read 
using the ACCEPT statement. In the same concept one can DISPLAY the 
form, and point to a html template.

The trick here, is that the compile can use the field-names for mapping to the 
input-fields of the html-form. Which is hard to do in the runtime.
RMCobol has a runtime module doing the same on a call-by-call basis (for 
each field).
The RMComponent is known as CGIEX and is commercially available for 
others compilers as well.

My point. Since all this cgi and web thingies are different per compiler, I 
intend to develop a cgi interface (as described here and above) completely 
written in Cobol.
I"ve done similar things and a basic parser is already available.

What I basically need in the cobol-syntax is the statement
MOVE A TO B WITH CONVERSION.

Let me know your needs, ideas, hints, tricks, etc.

Cheers,
wim niemans

Re: [Tiny-cobol-users] Introduction and call for ideas

[Wim N. <ni...@pb...>]

> I belive it is a good idea.
> I as well am an indipendent software developer and think that portability is essential.
> Commercial java products I tested, result being poor and extremely slow in running. And mixed-mode's one seem to be a long way to go.
> As I planned, starting from zero, to build a product of my own, I think thar co-operating would do things easier.
> for direct mailing you can address to st...@fo....
> Regards,
> Marcello Marsile

Well, a road map would be:

1. CGI input from the GET-method.
2. CGI output using templates with simple substitution.
3. CGI input from the POST method
4. CGI input for uploading files.
5. CGI output using templates with simple scripting.

I need your ideas, related to COBOL. Do not hesitate to send them.

Cheers
wim niemans

[Tiny-cobol-users] Tiny Cobol and CGI

[Rod S. <rod...@ya...>]

Is there a way to use tiny cobol in a cgi-bin with
data from a html form?  I have used rm\cobol by
www.liant.com and this options is available.  It is
very usefull and if its available please let me know.
Thank you,
rod...@ya... 

__________________________________________________
Do you Yahoo!?
Yahoo! News - Today's headlines
http://news.yahoo.com

Re: [Tiny-cobol-users] Our homepage was updated? (Complementing)

[<bur...@ya...>]

Hi

Yes I have to admit to making a couple of minor
changes, I was thinking of also spliting the download
pages up into sub pages one for each platform that we
support , I dont know what your thoughts are on this ?

Peter

 --- Hudson Reis <hud...@so...> wrote: > Hi
again,
> 
> > Hi, 
> > 
> > Look the our official TinyCOBOL homepage, I notice
> that it was updated... Is 
> > this true? 
> 
> This question is this: I Need to update the
> brazilian TinyCOBOL homepage?
> 
> Regards
> Hudson
> 
> 
>
-------------------------------------------------------
> In remembrance
> www.osdn.com/911/
> _______________________________________________
> Tin...@li...
>
https://lists.sourceforge.net/lists/listinfo/tiny-cobol-users 

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com

Re: [Tiny-cobol-users] Our homepage was updated? (Complementing)

[Hudson R. <hud...@so...>]

Hi again,

> Hi, 
> 
> Look the our official TinyCOBOL homepage, I notice that it was updated... Is 
> this true? 

This question is this: I Need to update the brazilian TinyCOBOL homepage?

Regards
Hudson

[Tiny-cobol-users] Our homepage was updated?

[<hud...@so...>]

Hi, 

Look the our official TinyCOBOL homepage, I notice that it was updated... Is 
this true? 

Regards.
Hudson

[Tiny-cobol-users] Introduction and call for ideas

[Wim N. r. <ni...@pb...>]

Hello all,

My name is wim niemans.
My origin is Eindhoven, Netherlands. Momentary I reside in Bonn, 
Germany.

I am an independent software developer (since 1979) and use mainly 
Cobol for application development (since 1972).
You may read (in Dutch) about me at www.nlbox.com/vrinternet

My point of focus is standardized Cobol. Systems I write do have 
source compatibilty between Cobol platforms. Paradigma is 'write 
once - compile everywhere (where Cobol is the environment)'. This 
as oposed to the Java paradigma 'write once-run anywhere'.

Current project is to port a wms (see www.tinywms.com) to an 
embedded Linux environment. Lately I did discover that the jargon for 
this activity is 'reconfigurable'.

Because this (cobol written) system run through a webserver and 
MUST use CGI, and I don't like C-calls (or whatever non-cobol calls), 
I intend to develop a true Cobol (written) interface for CGI. 
Comparable to the perl components cgilib.pl and/or CGI.pm.
However, there might be some stuff around. In a week there will be a 
specification available, though this will cover only my needs for CGI. 
Therefor I call for ideas, contributions, suggestions, hints, and 
whatelse might be usefull.

All resulting source code will be donated to each contributor. 
Intention is to create a true open source Cobol component.

Sincerely,
wim niemans

[Tiny-cobol-users] Re: [ tiny-cobol-Bugs-567343 ] ACCEPT DATE

[Bernard G. <bg...@fr...>]

no...@so... a écrit :

> Bugs item #567343, was opened at 2002-06-11 10:30
> You can respond by visiting:
> https://sourceforge.net/tracker/?func=detail&atid=100383&aid=567343&group_id=383
>
> Category: None
> Group: None
> >Status: Closed
> Resolution: Fixed
> Priority: 5
> Submitted By: Nobody/Anonymous (nobody)
> Assigned to: Giroud, Bernard (bgiroud)
> Summary: ACCEPT DATE
>
> Initial Comment:
> I declare W01-DATE 9(08)  in the working section
>
> the result of ACCEPT W01-DATE FROM DATE is
>
> '020611  ' instead of '00020611'
>
> ----------------------------------------------------------------------
>
> Comment By: Giroud, Bernard (bgiroud)
> Date: 2002-09-10 07:36
>
> Message:
> Logged In: YES
> user_id=1011
>
> The accept routines would not follow the rules for numeric
>
> move. Fixed in CVS as of 10 Sep 2002.
>
> ----------------------------------------------------------------------
>
> You can respond by visiting:
> https://sourceforge.net/tracker/?func=detail&atid=100383&aid=567343&group_id=383

Re: [Tiny-cobol-users] Compile time error

[David E. <de...@ar...>]

At 08:01 PM 09/09/02 +0100, Peter Burgess wrote:
 

>I was wondering if anyone can point out what I am
>doing wrong, I am compiling a cobol program (ex
>HP3000) which has file i/o, so I have berkleys db
>Library installed versions 2.? and 3.? , but I get the
>following error message:-
>
>$ htcobol pay2.cob
> ... ld: cannot find -ldb
>collect2: ld returned 1 exit status
>
>I am working via cygwin and understand I need to
>change the libraries to point to ldb2 or ldb3 , the
>only problem is I cannot find them on my system , what
>could be the possible issue ? Is this called something
>else other than ldb2 or ldb3 on cygwin ?

AFAIK, Berkeley’s DB is not part of the basic Cygwin distribution.

You can find version 1.85 binaries for both Cygwin and MinGW on the
download TC web pages. 

The Cygwin version can be installed using the Cygwin 'setup.exe'.

Hope this helps.





David Essex

[Tiny-cobol-users] Compile time error

[<bur...@ya...>]

Hi 

I was wondering if anyone can point out what I am
doing wrong, I am compiling a cobol program (ex
HP3000) which has file i/o, so I have berkleys db
Library installed versions 2.? and 3.? , but I get the
following error message:-

$ htcobol pay2.cob
/usr/lib/gcc-lib/i686-pc-cygwin/2.95.3-5/../../../../i686-pc-cygwin/bin/ld:
cann
ot find -ldb
collect2: ld returned 1 exit status

I am working via cygwin and understand I need to
change the libraries to point to ldb2 or ldb3 , the
only problem is I cannot find them on my system , what
could be the possible issue ? Is this called something
else other than ldb2 or ldb3 on cygwin ?

Thanks in advance for your help.

Peter Burgess

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com

Re: [Tiny-cobol-users] Run Time Error ??

[eurlix <eu...@li...>]

Hi Bernars,

Le Thu, 05 Sep 2002 19:22:49 +0200
Bernard Giroud <bg...@fr...> a =E9crit:

...
> I'm not really surprised after debugging a little:
> - your variable names indicate you are trying to get seperatly
> the program name and the wtest zone, which means that
> your intro must look something like (notice the sharp after
> the pgm name)::
> ECLAIR #0#vis${USERNUM}#lst${USERNUM}#${SorV}#  2>eclair.como
>=20
> - but that's not enough: if you do a display of intro, before the
> unstring, you don't get much; to get it right you need the
> following: ECLAIR "#0#vis${USERNUM}#lst${USERNUM}#${SorV}#"=20
> 2>eclair.como
>=20
> Hope this helps!
>=20
...

I think that you are wrong :
 ECLAIR =3D $0
 0#  ...  #$Sorv is an unique variable, "unstriged"  by ECLAIR
 (in fact, the differents variables should be separated
  by space and loaded one by one ... but I dont know how
  write a "shift" in cobol :-)

In spite of with I think, I have tried what you write and
the result is the same.
REMARKS : the first variable is well positionned in all cases,
  unless the second is not moved.

  (the 2>eclair.como is not for display the intro, but for
  redirect std-error in the program)

Hope this helps !

best regards,
--=20
Alain Lucari    Eurlix
1, rue Reine Elisabeth von Witellsbach
       (Reine des Belges)
13001 Marseille
FRANCE

Re: [Tiny-cobol-users] Indexes in Linkage

[Rildo P. <ri...@pr...>]

Hi Chris,

On 7 Sep 2002, Chris Borgn=E6s wrote:

> That's good news, as long as the area for the index isn't part of
> linkage.  Is it?

No, it is "hidden" in the regular data area.
Let's see how this apply to your code. Here is a test program with only
your data:

 IDENTIFICATION DIVISION.
 PROGRAM-ID. TESTING.
 ENVIRONMENT DIVISION.
 INPUT-OUTPUT SECTION.
 FILE-CONTROL.
 DATA DIVISION.
 LINKAGE SECTION.
    01 LnkWAtkTab SYNC.
    05 LnkWAtkRec OCCURS 20 TIMES INDEXED BY WAR-Idx.
     10 LnkCritMinRoll OCCURS 5 TIMES INDEXED BY
            CMR-Idx PIC 999 COMP.
     10 LnkDiv PIC 99 COMP.
     10 LnkCritTyp PIC XXX.

 PROCEDURE DIVISION.
 STOP RUN.

If you compile it with listing output enabled, you may see at the end of
listing, the section which each variable belongs to. Section 3 (according
defines in compiler/htcoboly.h) is SEC_DATA or SEC_WORKING, for working
storage variables. Your CMR_IDX and WAR_IDX variables will be placed
there. All your other variables (supposing the only data defined is yours
supplied in the previous e-mail) will be placed in section 5, or SEC_ARGS.

best regards,
Rildo
 ------------------------------------------------------------------
 Rildo Pragana                   FPGA/uControllers * Linux * tcl/tk
 R.Joaquim Nabuco,92/302 Derby   http://www.pragana.net
 Recife, PE - Brazil 52011-000   +55-81-3223-5694

Re: [Tiny-cobol-users] Indexes in Linkage

[Rildo P. <ri...@pr...>]

Hi Chris,

On 7 Sep 2002, Chris Borgn=E6s wrote:

> First of all, I'd like to point out a feature of Cobol that most other
> languages aren't good at - storage handling.
> Take a look at this Linkage Section:
>        LINKAGE SECTION.
>        01  LnkWAtkTab SYNC.
>            05  LnkWAtkRec OCCURS 20 TIMES INDEXED BY WAR-Idx.
>                10  LnkCritMinRoll OCCURS 5 TIMES INDEXED BY CMR-Idx.
>                                PIC 999 COMP.
>                10  LnkDiv      PIC 99  COMP.
>                10  LnkCritTyp  PIC XXX.
>
> The sync is necessary so C knows where a structure begins - on a word or
> half-word boundary.  Without it the call returns gibberish.  If it were
> Cobol calling this module the sync wouldn't be necessary!
>
> Here's the problem: where are the indexes' storage placed?  I've got
> stuctures in C that don't include them:
>
> typedef struct {
>  char WCd[2];
> =09int=09 CritMinRoll[5];
> =09int=09 Div;
> =09char=09CritTyp[3];
> } WAtkTab;
>
> WAtkTab=09Weapon[20];

In current TC's implementation, the index variable gets a separate storage
area. If you want to change the index, you must explicitly pass it to your
C subroutine.

best regards,
Rildo
 ------------------------------------------------------------------
 Rildo Pragana                   FPGA/uControllers * Linux * tcl/tk
 R.Joaquim Nabuco,92/302 Derby   http://www.pragana.net
 Recife, PE - Brazil 52011-000   +55-81-3223-5694

[Tiny-cobol-users] Indexes in Linkage

[Chris <cbo...@mi...>]

First of all, I'd like to point out a feature of Cobol that most other
languages aren't good at - storage handling.
Take a look at this Linkage Section:
       LINKAGE SECTION.
       01  LnkWAtkTab SYNC.
           05  LnkWAtkRec OCCURS 20 TIMES INDEXED BY WAR-Idx.
               10  LnkCritMinRoll OCCURS 5 TIMES INDEXED BY CMR-Idx.
                               PIC 999 COMP.
               10  LnkDiv      PIC 99  COMP.
               10  LnkCritTyp  PIC XXX.

The sync is necessary so C knows where a structure begins - on a word or
half-word boundary.  Without it the call returns gibberish.  If it were
Cobol calling this module the sync wouldn't be necessary!

Here's the problem: where are the indexes' storage placed?  I've got
stuctures in C that don't include them:

typedef struct {
 char WCd[2];
	int	 CritMinRoll[5];
	int	 Div;
	char	CritTyp[3];
} WAtkTab;

WAtkTab	Weapon[20];

Re: [Tiny-cobol-users] Re: Bug Report

[Rildo P. <ri...@pr...>]

Hi,

On Thu, 5 Sep 2002 hud...@so... wrote:

> >            SELECT Student-File ASSIGN TO "GPA.DAT".
> >            SELECT Out-File     ASSIGN TO SYS$OUTPUT.
>
> I think that the problem is the output name to Out-File (SYS$OUTPUT).
> Modify this name and it will work.

Not really that. The problem is because the file name is not within
double quotes. The first SELECT is right. The second must be something
like
           SELECT Out-File     ASSIGN TO "SYS$OUTPUT".

Anyway, I suspect this is not what the programmer wants. We don't have
the standard output as a file name. Instead, we have STD-OUTPUT (and
STD-ERROR as well) that may be used for DISPLAY statements.

The other form of assignment is with a named variable. Say we declare

SELECT OutFile ASSIGN to W-FILENAME.

then in the data division we must define this variable:

  77 W-FILENAME     PIC X(80).

or whatever we like. We must of course, move the actual file name
to that variable, before executing any OPEN statement, or use a VALUE
clause at its declaration.

best regards,
Rildo
 ------------------------------------------------------------------
 Rildo Pragana                   FPGA/uControllers * Linux * tcl/tk
 R.Joaquim Nabuco,92/302 Derby   http://www.pragana.net
 Recife, PE - Brazil 52011-000   +55-81-3223-5694

RE: [Tiny-cobol-users] Bug Report

[Bill G. <bi...@cs...>]

On Thu, 5 Sep 2002 bko...@ma... wrote:

> I think the problem may be in the second SELECT statement.  IIRC, TC only
> allows the use of a literal or a data element as the object of an
> assignment.  The '$' in the ASSIGN clause is probably throwing the compiler
> off.

That was it.  And a quick look at my old (and unused for many years) COBOL
book verified that the "$" is not valid in that use.

>
> What are you trying to accomplish with this SELECT statement?  (It looks
> like you want the files to go to the standard output.)

SYS$OUTPUT is the VMS way of saying stdout.  Yet another case of a vendor
allowing non-standard code to pass through their compiler.  Of course,
these are the same people who complain about lack of portability.

Thanks everyone who answered.  Looks like this compiler will do the
trick for this sememster.  I'll be running more code through it myself
and, of course, if anyone can break it, a bunch of undergrads can.
I'll report back any problems that turn up.

All the best.

bill

-- 
Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves
bi...@cs...     |  and a sheep voting on what's for dinner.
University of Scranton   |
Scranton, Pennsylvania   |         #include <std.disclaimer.h>

[Tiny-cobol-users] Re: Bug Report

[<hud...@so...>]

Hi Bill, 

> Here is the error message I receive:
>        server2# htcobol -F -L/usr/local/lib w1.cob
>        w1.s: Assembler messages:
>        w1.s:226: Error: suffix or operands invalid for `mov' 
> 
> Here is the assembler generated (the error being rather obvious): 
> 
> .LB_MAIN_PROGRAM_0:
>         pushl   $3
>         movl    $ww_base0+0 #sec:0, %eax
>         pushl   %eax 
> 
> Notice the extraneous "#sec:0" in the movl..... 
> 
> And here's the COBOL that causes it.  The testing I have done points to
> it being something in the "OPEN OUTPOUT" statement. 
> 
>        INPUT-OUTPUT SECTION.
>        FILE-CONTROL.
>            SELECT Student-File ASSIGN TO "GPA.DAT".
>            SELECT Out-File     ASSIGN TO SYS$OUTPUT.

I think that the problem is the output name to Out-File (SYS$OUTPUT).
Modify this name and it will work. 

>        INPUT-OUTPUT SECTION.
>        FILE-CONTROL.
>            SELECT Student-File ASSIGN TO "GPA.DAT".
>            SELECT Out-File     ASSIGN TO SYSOUTPUT.

or 

>        INPUT-OUTPUT SECTION.
>        FILE-CONTROL.
>            SELECT Student-File ASSIGN TO "GPA.DAT".
>            SELECT Out-File     ASSIGN TO "SYSOUTPUT".

or 

>        INPUT-OUTPUT SECTION.
>        FILE-CONTROL.
>            SELECT Student-File ASSIGN TO "GPA.DAT".
>            SELECT Out-File     ASSIGN TO external "SYSOUTPUT".
 

Hope this helps. 

Hudson.

[Tiny-cobol-users] Bug Report

[Bill G. <bi...@cs...>]

I am testing Tiny Cobol in anticipation of its being used this semester
by our students.  In doing this, I have uncovered a bug.  I am using
FreeBSD and the bug first appeared in the version in The Ports collection,
which is 0.57.  I then manula built and installed 0.58.  The same bug
is there also.

Here is the error message I receive:
       server2# htcobol -F -L/usr/local/lib w1.cob
       w1.s: Assembler messages:
       w1.s:226: Error: suffix or operands invalid for `mov'

Here is the assembler generated (the error being rather obvious):

.LB_MAIN_PROGRAM_0:
        pushl   $3
        movl    $ww_base0+0 #sec:0, %eax
        pushl   %eax

Notice the extraneous "#sec:0" in the movl.....

And here's the COBOL that causes it.  The testing I have done points to
it being something in the "OPEN OUTPOUT" statement.


       IDENTIFICATION DIVISION.
       PROGRAM-ID. Grade-Point-Averages.
       AUTHOR.  R. McCloskey and A. Programmer.
      * Assisted by: no one
       INSTALLATION.  U of S.
       DATE-WRITTEN.  September 1, 1992.
      ***********************************************************
      * Program Abstract:
      *   < Here should be a general description of
      *     the purpose/functionality of the program >
      ***********************************************************

       ENVIRONMENT DIVISION.

       INPUT-OUTPUT SECTION.
       FILE-CONTROL.
           SELECT Student-File ASSIGN TO "GPA.DAT".
           SELECT Out-File     ASSIGN TO SYS$OUTPUT.

       DATA DIVISION.

       FILE SECTION.

       FD  Student-File
           DATA RECORD IS Student-Record.
       01  Student-Record.
           02 SSN                  PIC X(9).
           02 Name                 PIC X(20).
           02 Attempted-Hours      PIC 99.
           02 Quality-Points       PIC 99V99.

       FD  Out-File
           DATA RECORD IS Print-Line.
       01  Print-Line   PIC X(80).

       WORKING-STORAGE SECTION.

       01 Eof                   PIC X     VALUE 'F'.
       01 Gpa                   PIC 9V99.
       01 Deans-List-Threshold  PIC 9V9   VALUE 3.5.

       01 Col-Head-Line-1.
          02 FILLER PIC X(37) VALUE SPACES.
          02 FILLER PIC X(5)  VALUE 'Hours'.

       01 Col-Head-Line-2.
          02 FILLER PIC X(6)  VALUE SPACES.
          02 FILLER PIC A(3)  VALUE 'SSN'.
          02 FILLER PIC X(12) VALUE SPACES.
          02 FILLER PIC X(4)  VALUE 'Name'.
          02 FILLER PIC X(10) VALUE SPACES.
          02 FILLER PIC X(15) VALUE 'Attempted'.
          02 FILLER PIC X(2)  VALUE SPACES.
          02 FILLER PIC X(3)  VALUE 'GPA'.
          02 FILLER PIC X(5)  VALUE SPACES.
          02 FILLER PIC X(7)  VALUE 'Remarks'.

       01 Out-Line.
          02 FILLER               PIC X(3)  VALUE SPACES.
          02 Out-SSN              PIC X(9).
          02 FILLER               PIC X(5)  VALUE SPACES.
          02 Out-Name             PIC X(20).
          02 FILLER               PIC X(2)  VALUE SPACES.
          02 Out-Attempted-Hours  PIC Z9.
          02 FILLER               PIC X(10) VALUE SPACES.
          02 Out-GPA              PIC 9.99.
          02 FILLER               PIC X(5)  VALUE SPACES.
          02 Out-Remarks          PIC X(20).


       PROCEDURE DIVISION.

       Main-Program.
           OPEN OUTPUT Out-File
           OPEN INPUT  Student-File
           PERFORM Print-Col-Headings
           PERFORM Read-a-Record
           PERFORM UNTIL Eof = 'T'
              PERFORM Process-a-Record
              PERFORM Read-a-Record
           END-PERFORM
           CLOSE Student-File
           CLOSE Out-File
           STOP RUN
           .

       Read-a-Record.
           READ Student-File
              AT END MOVE 'T' TO Eof
           END-READ
           .

       Process-a-Record.
           DIVIDE Attempted-Hours INTO Quality-Points GIVING Gpa
           IF Gpa IS GREATER THAN Deans-List-Threshold
             MOVE "Dean's List" TO Out-Remarks
           ELSE
             MOVE SPACES TO Out-Remarks
           END-IF
           MOVE SSN TO Out-SSN
           MOVE Name TO Out-Name
           MOVE Attempted-Hours TO Out-Attempted-Hours
           MOVE Gpa TO Out-GPA
           PERFORM Write-Record
           .

       Write-Record.
           WRITE Print-Line FROM Out-Line
           .

       Print-Col-Headings.
           WRITE Print-Line FROM Col-Head-Line-1 BEFORE ADVANCING 1 LINE
           WRITE Print-Line FROM Col-Head-Line-2 BEFORE ADVANCING 1 LINE
           .

       END-PROGRAM.

Any help greatly appreciated and I will be glad to test any patch you
may come up with.

All the best and good luck.

bill

-- 
Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves
bi...@cs...     |  and a sheep voting on what's for dinner.
University of Scranton   |
Scranton, Pennsylvania   |         #include <std.disclaimer.h>

Re: [Tiny-cobol-users] Run Time Error ??

[Bernard G. <bg...@fr...>]

eurlix a écrit :

> Hi Bernard,
>
> Le Tue, 3 Sep 2002 14:11:48 +0200
> ber...@cr... a écrit:
>
> > Funny ! The area that is shown should be correct for a source
> > field.
> >
> > What are the descriptions of your fields (pictures), and are we
> > seeing the data of the source or destination field?
> >
> > Bernard Giroud
> > Crédit Lyonnais (Suisse) SA
> >
>
> Not really "funny", but we must laugh with we has ?
>
> Sure, we are seeing the source field, because the destination
> field is empty :
>
> with the folloving fields,
>
>  01  wtest pic 9 value 1.
>      88 debug value 1.
>  01  WLABEL PIC X(32) .
>  01  F-LISTF PIC X(32).
>  01  VERIF PIC X(1).
> the run of :
> ECLAIR 0#vis${USERNUM}#lst${USERNUM}#${SorV}#  2>eclair.como
>
>      ACCEPT intro   FROM COMMAND-LINE.
>      MOVE SPACE TO VERIF.
>      UNSTRING intro  DELIMITED BY '#'  INTO wprogname wtest WLABEL F-LISTF VERIF.
> Run Time Error - Invalid Data Content
> Field Description: len = 8, type = X, dec/pscale = 0/0, all = 0, just_r = 0, signs = 0/0
> Data Dump, Address = 080C4F27
> 0000: 76 69 73 31 37 35 31 39
>        v  i  s  1  7  5  1  9
>
>      display wprogname " wtest(" wtest ") " wlabel " " f-listf " " verif upon std-error.
> give :
> ECLAIR wtest(0) lst17519                         S
>
> look that "lst17519" is in place of "f-listf" and "S" in place of "verif" !
>

I'm not really surprised after debugging a little:
- your variable names indicate you are trying to get seperatly
the program name and the wtest zone, which means that
your intro must look something like (notice the sharp after
the pgm name)::
ECLAIR #0#vis${USERNUM}#lst${USERNUM}#${SorV}#  2>eclair.como

- but that's not enough: if you do a display of intro, before the
unstring, you don't get much; to get it right you need the following:
ECLAIR "#0#vis${USERNUM}#lst${USERNUM}#${SorV}#"  2>eclair.como

Hope this helps!

>
> I have replaced the accept from command-line by :
>    684    355 [00]:      move space to wtest wlabel f-listf verif
>    685    356 [00]:      call 'GETVAR' using wdebug wtest
>    686    357 [00]:      string "vis" usnum into wlabel
>    687    358 [00]:      string "lst" usnum into f-listf
>    688    359 [00]:      call 'GETVAR' using SorV verif
>    689    360 [00]:      display wprogname " par GETVAR : wtest(" wtest ") " wlabel " " f-listf " " verif upon std-error.
>    wich give :
> ECLAIR par GETVAR : wtest(0) vis17798                         lst17798                         S
> this is good.
>
> but, I have the same problem in an other program, where I can't do the same.
>
> In the sample above, the destination fields are not initialized, but in
> the other program they are padded to space before a simple "move"
> and the result is a very curious thing (more funny ;-).
>
> BTW : You can have more explanations on the problem with sort in
>   mu last mail to Hudson, in TC list, subject "How to reduce ...".
>

OK! I'll have a look this week!

>
> Regards,
> --
> Alain Lucari    Eurlix
> 1, rue Reine Elisabeth von Witellsbach
>        (Reine des Belges)
> 13001 Marseille
> FRANCE

Bernard

Re: [Tiny-cobol-users] Run Time Error ??

[eurlix <eu...@li...>]

Hi Bernard,

Le Tue, 3 Sep 2002 14:11:48 +0200
ber...@cr... a =E9crit:

> Funny ! The area that is shown should be correct for a source
> field.
> =20
> What are the descriptions of your fields (pictures), and are we
> seeing the data of the source or destination field?
>=20
> Bernard Giroud
> Cr=E9dit Lyonnais (Suisse) SA
>=20

Not really "funny", but we must laugh with we has ?

Sure, we are seeing the source field, because the destination
field is empty :

with the folloving fields,

 01  wtest pic 9 value 1.
     88 debug value 1.
 01  WLABEL PIC X(32) .
 01  F-LISTF PIC X(32).
 01  VERIF PIC X(1).
the run of :
ECLAIR 0#vis${USERNUM}#lst${USERNUM}#${SorV}#  2>eclair.como

     ACCEPT intro   FROM COMMAND-LINE.
     MOVE SPACE TO VERIF.
     UNSTRING intro  DELIMITED BY '#'  INTO wprogname wtest WLABEL F-LIST=
F VERIF.
Run Time Error - Invalid Data Content
Field Description: len =3D 8, type =3D X, dec/pscale =3D 0/0, all =3D 0, =
just_r =3D 0, signs =3D 0/0
Data Dump, Address =3D 080C4F27
0000: 76 69 73 31 37 35 31 39=20
       v  i  s  1  7  5  1  9=20

     display wprogname " wtest(" wtest ") " wlabel " " f-listf " " verif =
upon std-error.
give :    =20
ECLAIR wtest(0) lst17519                         S      =20

look that "lst17519" is in place of "f-listf" and "S" in place of "verif"=
 !=20

I have replaced the accept from command-line by :
   684    355 [00]:      move space to wtest wlabel f-listf verif
   685    356 [00]:      call 'GETVAR' using wdebug wtest
   686    357 [00]:      string "vis" usnum into wlabel
   687    358 [00]:      string "lst" usnum into f-listf
   688    359 [00]:      call 'GETVAR' using SorV verif
   689    360 [00]:      display wprogname " par GETVAR : wtest(" wtest "=
) " wlabel " " f-listf " " verif upon std-error.
   wich give :
ECLAIR par GETVAR : wtest(0) vis17798                         lst17798   =
                      S
this is good.

but, I have the same problem in an other program, where I can't do the sa=
me.

In the sample above, the destination fields are not initialized, but in
the other program they are padded to space before a simple "move"
and the result is a very curious thing (more funny ;-).

BTW : You can have more explanations on the problem with sort in
  mu last mail to Hudson, in TC list, subject "How to reduce ...".

Regards, =20
--=20
Alain Lucari    Eurlix
1, rue Reine Elisabeth von Witellsbach
       (Reine des Belges)
13001 Marseille
FRANCE