tiny-cobol-users

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

[<ber...@cr...>]
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[Tiny-cobol-users] Run Time Error ??

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

Hi all,

Can somebody help me on this message :

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 

all seem to me good, but the move is not made 
why ?

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

Re: [Tiny-cobol-users] problem with SORT

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

Hi Bernard,

Le Tue, 03 Sep 2002 07:42:54 +0200
Bernard Giroud <bg...@fr...> a =E9crit:

> Andrew Cameron a =E9crit :
>=20
> > Hi,
> >
> > I see that someone has broken this by changes made recently to the
> > compiler as
> > the test08 code compiles with an older version I had of TC.
> >
> > Unfortunatly I have not worked with the compiler code for some
> > time.
> >
> > David,
> >
> > Would you be able to look at htcobgen.c to determine why
> > SORT-RETURN is not working.
> >
> > Regards
> > Andrew Cameron
> >
>=20
> I'll have a look, unless somebody already has.
>=20
> Bernard
>=20
>=20
I don't think that somebody have a look at this.

The sort-return don't works because it is not defined
in the tests programs.

But it's not the real problem, wich is on the USING=20
clause in the sort.

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

Re: [Tiny-cobol-users] problem with SORT

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

Andrew Cameron a écrit :

> Hi,
>
> I see that someone has broken this by changes made recently to the compiler
> as
> the test08 code compiles with an older version I had of TC.
>
> Unfortunatly I have not worked with the compiler code for some time.
>
> David,
>
> Would you be able to look at htcobgen.c to determine why SORT-RETURN
> is not working.
>
> Regards
> Andrew Cameron
>

I'll have a look, unless somebody already has.

Bernard

Re: [Tiny-cobol-users] Lock files/registers from a alternative way with TinyCOBOL

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

Andrew Cameron a écrit :

> Hi,
>
> You may want to have a look at Lockserver.
> http://sourceforge.net/projects/lockserver/
>
> LockServer 0.9b - 7/14/2000
> ---------------------------
>

Snip ..

>
> Regards
> Andrew Cameron
>

That seems interesting: not too complicated at first glance,
although pthreaded. Has anybody tried to compile it yet?

Bernard

Re: [Tiny-cobol-users] Lock files/registers from a alternative way with TinyCOBOL

[Andrew C. <apc...@so...>]

Hi,

You may want to have a look at Lockserver.
http://sourceforge.net/projects/lockserver/



LockServer 0.9b - 7/14/2000
---------------------------

o What is LockServer?
-    Where I work, we have need of row locking for a database that doesn't
provide it.  The solution I came up with was to write a generic lock server
that could provide the functionality we needed.

o Ok, so how does it work?
-    The server itself is network based, and so far it seems to handle load
reasonably well.  Essentially, you make a request, it gets queued.  The
server will return where you are in the queue.

o How does it REALLY work?
-    The storage mechanism is basically implemented as a hash of linked
lists.  At this point, it would be generous to describe the hash algorithm
as pathetic; anybody who wants to offer suggestions is certainly welcome to.
The protocol the lock server uses is described in the PROTOCOL document that
should accompany this package.

o What platforms does LockServer run on?
-    To date, the LockServer has only been tried on linux (Intel and Sparc).
Portability was not a design concern so I'd be very interested to hear of
platforms it does and does not run on.  Anybody wishing to work on making
it run on other platforms is encouraged to do so.  Giving me access to
other platforms to allow me to port it would also be appreciated.  You can
contact me a co...@tu....

o How do I use LockServer?
-    At this point, the only client software I know is some proprietary
code that some coworkers wrote and cannot (unfotunately) be released under
the GPL.  If anybody wishes to make client code to access LockServer, I
would be happy to make it available on my ftp server.  Guidelines and
technical information on the LockServer are available in the PROTOCOL
document.


Thanks to Eric Johnson and Robert Holak for development ideas and testing.
---------------------

Regards
Andrew Cameron


----- Original Message -----
From: "Hudson Reis" <hud...@so...>
To: "Lista SourceForge" <tin...@li...>
Cc: <apc...@so...>; "Husni Ali Husni" <hu...@gl...>
Sent: 01 September 2002 09:33
Subject: [Tiny-cobol-users] Lock files/registers from a alternative way with
TinyCOBOL


> Hi Andrew,
>
> A brasilian user(Husni Ali Husni) suggested to me an alternative way to
lock files/registers in TinyCOBOL. Talking to Rildo, he said to me to ask
you. The Husli Ali Husni idea is this:
>
> Portuguese:
> -----------
>    Que tal usar para lock de arquivos um subprograma em C que valida a
existencia
> de um arquivo mestre e o arquivo mestre é um arquivo indexado que contem
> uma chave de pesquisa para o arquivo ou até mesmo o registro.
>
> English:
> --------
>    What do you think to use a C subprogram(to lock of files/registers)
that validate the existence of a master file and the master file is a
indexed file that contains a search key to file or register.
>
> Do you think is this a good idea since lock isn't supported by TinyCOBOL?
>
> Thanks
> Hudson
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Tin...@li...
> https://lists.sourceforge.net/lists/listinfo/tiny-cobol-users

Re: [Tiny-cobol-users] How to reduce the time for found a shared library ?

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

Hi Hudson,

Le Sun, 1 Sep 2002 13:06:11 -0300
Hudson Reis <hud...@so...> a =E9crit:

> Hi Alain,
>
...
>=20
> If your doubt isn't this, sorry for don't have helped you. :(
>=20
I have no doubt, but my problem was elsewhere (sure I made a
mistake) and now the time is good.
In any cases, thanks for your help.

> > Perhaps for the same reason the "sort" is broken ?
>=20
> You talk about the SORT verb? If yes, unfortunatelly the verb SORT
> not have working, the problem is this:
>=20
> htcobol  -c  -P -D -I../copybooks -I. test08u.cob
> test08u.cob:    67: variable SORT-RETURN not defined,0
>=20
this is the first error, because the sort-return is not defined
(there are more controls in TC now than when Test08 cobs where written)
but, if you add in WORKING-STORAGE SECTION.
77 sort-return pic 99.
you can see the real problem :
a SORT with INPUT and OUTPUT PROCEDURE(S) is well compiled,
but a sort with USING a file name is not.
This is not a mistake in the test08 cobol program but
I suppose, somewhere in the compiler because most of the=20
time USING refer a FIELD name in W-S or Linkage section
but the sort USING refer to a FILE NAME in the FILE SECTION.

Anyway, it's not a big problem because we can sort a file
by a call "system" ...

> Regards and sorry for anyway...
> Hudson.
>=20
Other thing, you have certainly termcap (/etc/termcap)
and /usr/share/terminfo .
try=20
$ locate terminfo
(if you have slocate on your system)
and
$ apropos terminfo
$ man terminfo=20
$ man infocmp

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

Re: [Tiny-cobol-users] How to reduce the time for found a shared library ?

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

Hi Alain,

> Really not : I have not downloaded the last version from CVS,
> (because I am on a big work and, even I think that Ferran's job
> is a very good idea, I have no time to test it now), but, i think
> that the problems I have seen have nothing to do with Ferran's work
> but come from the parser (?) or other component.

Well, you didn't talking about to reduce the time for found a shared library? With the runtime library unified and as a shared library, the calls to COBOL shared libraries are much more fast that before(with the runtime as a static libraries).

If your doubt isn't this, sorry for don't have helped you. :(

> Perhaps for the same reason the "sort" is broken ?

You talk about the SORT verb? If yes, unfortunatelly the verb SORT not have working, the problem is this:

htcobol  -c  -P -D -I../copybooks -I. test08u.cob
test08u.cob:    67: variable SORT-RETURN not defined,0

Regards and sorry for anyway...
Hudson.

Re: [Tiny-cobol-users] Problems using function keys

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

Hi Jim,

Sorry to not answered yet..

> You can change the time delay by editing the terminfo file and "tic"ing it.
> Its a while since I have done this, so I can't remember which parameters you
> change, but it is pretty easy if you set aside a few hours to study it.

Well, I didn't know that! I'm a beginner in linux, then I don't know use the files terminfo and termcap. If you have this files, could you send this files to me? 
 
Thanks for your informations, are very good!. 

Best regards!
Hudson

Re: [Tiny-cobol-users] Interfacing TinyCOBOL programs with cgi-util library

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

Hi Ferran,

> The variable can be any type, I use pic x(n) but pic a(n) or pic 9(n) 
> should work.
> The size is not important, the size of parameter-wor only needs to be 
> big enought to contain all the address and the null char. If there is an > empty space after this, the C routine won't use it.

Ok. I got it.

> To get the return value, you should use the giving or returning clause 
> of the call statement.
> 
> call "cgi_goodemailaddress" using   parameter-wor giving return-wor.
> 
> I recomend that return-wor were a pic 9(6) comp, this picture is the equivalent to a C int.

This way, it shows nothing in return-wor variable, even if the e-mail address is true. Talking with Husni(a brasilian TinyCOBOL user), I notice that cgi_goodemailaddress routine not return values from a COBOL call.

anyway, thanks for your help!

Hudson

[Tiny-cobol-users] Lock files/registers from a alternative way with TinyCOBOL

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

Hi Andrew,

A brasilian user(Husni Ali Husni) suggested to me an alternative way to lock files/registers in TinyCOBOL. Talking to Rildo, he said to me to ask you. The Husli Ali Husni idea is this:

Portuguese:
-----------
   Que tal usar para lock de arquivos um subprograma em C que valida a existencia
de um arquivo mestre e o arquivo mestre é um arquivo indexado que contem
uma chave de pesquisa para o arquivo ou até mesmo o registro.

English:
--------
   What do you think to use a C subprogram(to lock of files/registers) that validate the existence of a master file and the master file is a indexed file that contains a search key to file or register.
   
Do you think is this a good idea since lock isn't supported by TinyCOBOL?

Thanks 
Hudson

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

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

Hi Boris,

> 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.
> 
> I don't exactly understand what you are trying to do, but the following
> changes should be made.

Well boris, I want to create a program like the development/test.code/tdb03, for example, but I've had problems in SEARCH verb, and the results in browser are very strange... :(

I noticed that my example didn't explain well. Sorry.
I will try to use this way below to show the values in browser... 

>        identification division
>        program-id. validate2.
>        author. Hudson Reis.
> 
>        data division.
>        working-storage section.
>        01 HtmlLineChars.
>           03 HtmlEntry occurs 5 indexed by i.
>              03 HtmlChar           pic x(001) value spaces.
> 
>        procedure division.
> 	SET I TO 1.
> 
>            search HtmlEntry varying i
> *            when HtmlLineChars(i:2) equal " "
> 		WHEN HtmlChar(i) = SPACES
> 
>                   display "Passed"
>            end-search
>            stop run.
> 
> Now, if what you are really trying to do is search pairs of character, the
> SEARCH verb is inappropriate.  You would want something like:
> 	PROCEDURE DIVISION.
> 	SET i TO 1.
> 	PERFORM  VARYING I FROM 1 BY 1 UNTIL i > 4
> 		IF HtmlChar(i) = SPACES AND HtmlChar(i+ 1) = SPACES
> 			DISPLAY "Passed"
> 		END-IF
> 	END-PERFORM.
> 	STOP RUN.

Thanks 
Hudson

Re: [Tiny-cobol-users] Extra stuff in assembler

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

Thank you all, I'm sure that'll do the trick.  I've left town for a long
weekend, and I'll try it when I get back.

On Fri, 30 Aug 2002 18:21:49 +0200 eurlix <eu...@li...> wrote:

Hi Chris,

Le 30 Aug 2002 09:42:23 -0500
Chris Borgnæs <cbo...@mi...> a écrit:

...
>        FILE-CONTROL.
>            SELECT WAtkTabFile ASSIGN TO WAtkTabFile

I think that this works better if you write :

           SELECT WAtkTabFile ASSIGN TO "WAtkTabFile"
or
           SELECT WAtkTabFile ASSIGN TO WAtkTabLabel
  and add in Working-storage section
  77 watktabLABEL pic x(??) value "what_you_want"

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

Re: [Tiny-cobol-users] Postgres interface

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

Hi Alain,

On Fri, 30 Aug 2002, eurlix wrote:

> Ok, (really I am the boss of nobody), but thank you !
> As Linus said "if you make a mistake on the web, it is for eternity"
> so my bads things (joins) may made laugh many peoples.
> Tant pis !

Well, I don't care to make mistakes. Whom has never made a mistake
throw the first stone! Besides, we only learn when we try (with or
without mistakes) to do something. So, I prefer trying to learn :^)

> First, a question :
> the new guide of PostgreSQL  (7.2.1 it has changed ans is better)
> said : it is better to use PQsendQuery and PQgetResult than
> PQexec.
> Have you an idea about this ?

Let's think a little. With PQsendQuery() we have a multiple-SQL-statement
query, that may return several answers, so we must check all the
answers (by calling PQgetResult()) till NULL is returned. Where do we
return all the collected answers to cobol? (or just the last answer?!)
Of course, this means we have a faster way to query many things.
On the other side, it is more complex, as we are not able to do another
query until this procedure finish. This is the tradeoff.
What do you prefer to use?

> Second, Using your own code, I have written intfpg.c ,
> adding some zones to sql_get_tuple and a rustine "frdb_num"
> this works, but I think is it not "clean"
> frdb_num.bad seem to me more clean, but don't work.
> (the comments are in french, but I suppose you understand its).
> your help is welcome !

Well, I will have to read it first. Let me answer that later.

> third, what I try to do :
> with some utilities I analyze the data copybooks of a file
> and generate a shell for creation of the table and a cobol subprog
> to access to it in the pg db
> with three tables
>  one for the length of the zones
>  one for the number of decimals
>  one for said if the zone is signed, sign leading or not signed ;
> this works almost automatically but is a little heavy ...

What are you trying to tell me? You are trying to map a cobol file
(record structure) into an automatically generated table?
There are a number of issues there. First, all COMP variables must
be converted into textual data (say, by sprintf()). In some places,
(for instance in the "oids") postgres don't like to have it included
in quotes (like '1234'), but for other numeric data, it must be
quoted. There are many more issues (exceptions to the rule)...

> Sure you can have some ideas in order to make it more light .
> If you have sometime and are interested, I can send you my
> procs in order to discuss about them.

A simple idea that comes to my mind (for just encapsulating a cobol
record in a sql table) is to use blobs (binary large objects). This
way you can save everything in native cobol format. But, as I would
say, it depends on the kind of application you have in mind.

Let us discuss more, so I can have a better picture of it.

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] Postgres interface

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

Hi  Rildo,

Le Fri, 30 Aug 2002 11:50:54 -0300 (BRT)
Rildo Pragana <ri...@pr...> a =E9crit:

> Hi Alain,
>=20
...
>=20
> Of course, boss :^)
> Tell me what you want.
>=20

Ok, (really I am the boss of nobody), but thank you !
As Linus said "if you make a mistake on the web, it is for eternity"
so my bads things (joins) may made laugh many peoples.
Tant pis !

First, a question :
the new guide of PostgreSQL  (7.2.1 it has changed ans is better)
said : it is better to use PQsendQuery and PQgetResult than
PQexec.=20
Have you an idea about this ?

Second, Using your own code, I have written intfpg.c ,
adding some zones to sql_get_tuple and a rustine "frdb_num"
this works, but I think is it not "clean"
frdb_num.bad seem to me more clean, but don't work.
(the comments are in french, but I suppose you understand its).
your help is welcome !

third, what I try to do :
with some utilities I analyze the data copybooks of a file
and generate a shell for creation of the table and a cobol subprog=20
to access to it in the pg db
with three tables=20
 one for the length of the zones
 one for the number of decimals
 one for said if the zone is signed, sign leading or not signed ;
this works almost automatically but is a little heavy ...

Sure you can have some ideas in order to make it more light .
If you have sometime and are interested, I can send you my=20
procs in order to discuss about them.

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

Re: [Tiny-cobol-users] How to reduce the time for found a shared library ?

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

Hi Hudson,

Le Fri, 30 Aug 2002 04:50:36 -0300
Hudson Reis <hud...@so...> a =E9crit:

> Hi Eurlix,
>=20
> You try to make this tests with the new modifications that Ferran
> does(Runtime unified and as a shared library)? I've had much
> problems with shared libraries, but after this, some problems was
> solved.
>=20
> Regards,
> Hudson

Really not : I have not downloaded the last version from CVS,
(because I am on a big work and, even I think that Ferran's job
is a very good idea, I have no time to test it now), but, i think
that the problems I have seen have nothing to do with Ferran's work
but come from the parser (?) or other component.

Perhaps for the same reason the "sort" is broken ?

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

[Tiny-cobol-users] test

[Hudson R. <hu...@ti...>]

Re: [Tiny-cobol-users] How to reduce the time for found a shared library ?

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

Hi Eurlix,

> It was a very good idea to add in TC the possibility of add
> a comment on a line with "*>"
> but if you write something like :
> 
> copy "../copyboooks/cust.desc". *> description of customers data
> the copy is not found and the program is not compiled
> 
> and
> 
> program-id. dusty. *> this subprog clean the computer after works hours
> the compil is ok, but the rustine is never found
> and after ~5 minutes you have a SIG 11.
> 
> An other bad idea is to put the cobols ".so" in the same directory
> as the executables. This increase the time to found the rustines.

You try to make this tests with the new modifications that Ferran does(Runtime unified and as a shared library)? I've had much problems with shared libraries, but after this, some problems was solved.

Regards,
Hudson

Re: [Tiny-cobol-users] Extra stuff in assembler

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

Hi Chris,

On 30 Aug 2002, Chris Borgn=E6s wrote:

> Here is an excerpt.  If this is inadequate I can provide the entire
> thing, but it's essentially multiple iterations of the same logic for
> both files.
>
>        IDENTIFICATION DIVISION.
>        PROGRAM-ID. BuildWAtkTab.
>        ENVIRONMENT DIVISION.
>        INPUT-OUTPUT SECTION.
>        FILE-CONTROL.
>            SELECT WAtkTabFile ASSIGN TO WAtkTabFile
>                               ORGANIZATION INDEXED
>                               ACCESS RANDOM
>                               RECORD KEY WAtkTabKey.

Because the way our compiler works, if you want to assign the filename
to a literal name, you must enclose it in double-quotes, like:

SELECT WAtkTabFile ASSIGN TO "WAtkTabFile"


In other situation, you may want to have the file name into
another variable, like this:

SELECT WAtkTabFile ASSIGN TO WAtkTabFile
=2E..
=0977 WAtkTabFile=09PIC X(80) VALUE "my_file_name".

Both are acceptable, but the compiler only assumes you give a explicit
filename if you enclose it inside double-quotes ("...").

What happened, is that your variable (WAtkTabFile) was not defined later,
so it had no storage class. Sorry if our compiler didn't show a better
error detection.

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] OK, I'm able to compile.

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

Hi Rildo,

Le Thu, 29 Aug 2002 16:00:58 -0300 (BRT)
Rildo Pragana <ri...@pr...> a =E9crit:

> Hi all,
>=20
> Thanks for all the feedback and help I received from all of you.
> I'm able to compile now, but with the following command line:
>=20
> ./configure --with-libdb=3D3
>=20
> (I must not forget that again!i :^)
>
 to write "make cleandist; ./configure --yours_options" in a shell
 can help your mind ;-)
=20
> Though it don't work well for "--with-libdb=3D2",
> I don't know exactly why!
> Well, I'm not going to spend time on that stuff. Let me do
> something more useful.
>=20
> 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
>=20
With the help of tdb02a.c, I have writted some programs to access
a postgreSQL db like it was an isam file (and more), but
as you know, i am not fluent in C ...

Have you now some minutes in order to help me ?
=20
Regards,=20
--=20
Alain Lucari    Eurlix
1, rue Reine Elisabeth von Witellsbach
       (Reine des Belges)
13001 Marseille
FRANCE

Re: [Tiny-cobol-users] How to reduce the time for found a shared library ?

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

Hi Ferran and All,

Le Thu, 22 Aug 2002 19:17:47 +0200
Ferran Pegueroles <fer...@re...> a =E9crit:

> Sorry for the late answer, I was on halydays.

no matter, it's not a good idea to have problems in August ;-)
Anyway, thanks for the help.

It was a very good idea to add in TC the possibility of add
a comment on a line with "*>"
but if you write something like :

copy "../copyboooks/cust.desc". *> description of customers data
the copy is not found and the program is not compiled

and

program-id. dusty. *> this subprog clean the computer after works hours
the compil is ok, but the rustine is never found
and after ~5 minutes you have a SIG 11.

An other bad idea is to put the cobols ".so" in the same directory
as the executables. This increase the time to found the rustines.

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

Re: [Tiny-cobol-users] Extra stuff in assembler

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

Hi Chris,

Le 30 Aug 2002 09:42:23 -0500
Chris Borgn=E6s <cbo...@mi...> a =E9crit:

...
>        FILE-CONTROL.
>            SELECT WAtkTabFile ASSIGN TO WAtkTabFile

I think that this works better if you write :

           SELECT WAtkTabFile ASSIGN TO "WAtkTabFile"
or
           SELECT WAtkTabFile ASSIGN TO WAtkTabLabel
  and add in Working-storage section
  77 watktabLABEL pic x(??) value "what_you_want"

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

Re: [Tiny-cobol-users] Extra stuff in assembler

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

Here is an excerpt.  If this is inadequate I can provide the entire
thing, but it's essentially multiple iterations of the same logic for
both files.

       IDENTIFICATION DIVISION.
       PROGRAM-ID. BuildWAtkTab.
       ENVIRONMENT DIVISION.
       INPUT-OUTPUT SECTION.
       FILE-CONTROL.
           SELECT WAtkTabFile ASSIGN TO WAtkTabFile
                              ORGANIZATION INDEXED
                              ACCESS RANDOM
                              RECORD KEY WAtkTabKey.
           SELECT WeaponTypeFile ASSIGN TO WeaponTypeFile
                                 ORGANIZATION INDEXED
                                 ACCESS RANDOM
                                 RECORD KEY WSName.
       DATA DIVISION.
       FILE SECTION.
       FD  WAtkTabFile.
       01  WAtkTab.
           05  WAtkTabKey.
               10  WCd         PIC XX.
               10  WAtkTabAT   PIC 99 COMP.
           05  CritMinRoll OCCURS 5 TIMES
                               PIC 999 COMP.
           05  Div             PIC 99  COMP.
           05  CritTyp         PIC XXX.
       FD  WeaponTypeFile.
       01  WeaponType.
           05  WSName          PIC XX.
           05  WName           PIC X(15).
           05  WUse            PIC XXX.
           05  WtMin           PIC 9V9  COMP.
           05  WtMax           PIC 99V9 COMP.
           05  LnMin           PIC 9V99 COMP.
           05  LnMax           PIC 99V9 COMP.
           05  Fum             PIC 9    COMP.
           05  RngInfo OCCURS 5 TIMES.
               10  Range       PIC  999 COMP.
               10  RangeMod    PIC S99  COMP.
       PROCEDURE DIVISION.
           OPEN OUTPUT WAtkTabFile WeaponTypeFile.

           INITIALIZE WeaponType.
           MOVE 'Long Bow' TO WName.
           MOVE 'MS' TO WUse.
           MOVE 'lb' TO WSName.
           MOVE 2.0 TO WtMin.
           MOVE 3.0 TO WtMax.
           MOVE 5.00 TO LnMin.
           MOVE 7.0 TO LnMax.
           MOVE 5 TO Fum.
           MOVE 10 TO Range(1).
           MOVE +20 TO RangeMod(1).
           MOVE 100 TO Range(2).
           MOVE 200 TO Range(3).
           MOVE -30 TO RangeMod(3).
           MOVE 300 TO Range(4).
           MOVE -40 TO RangeMod(4).
           MOVE 400 TO Range(5).
           MOVE -50 TO RangeMod(5).
           WRITE WeaponType INVALID KEY
               DISPLAY 'Failed to Write to WeaponTypeFile'
               CLOSE WAtkTabFile WeaponTypeFile
               STOP RUN.

           MOVE 1 TO WAtkTabAT.
           MOVE 'lb' TO WCd.
           MOVE 88 TO CritMinRoll(1).
           MOVE 93 TO CritMinRoll(2).
           MOVE 98 TO CritMinRoll(3).
           MOVE 104 TO CritMinRoll(4).
           MOVE 124 TO CritMinRoll(5).
           MOVE 3 TO Div.
           MOVE 'P  ' TO CritTyp.
           WRITE WAtkTab INVALID KEY
               DISPLAY 'Failed to Write to WAtkTabFile'
               CLOSE WAtkTabFile WeaponTypeFile
               STOP RUN.

           CLOSE WAtkTabFile WeaponTypeFile.
           STOP RUN.

On Fri, 2002-08-30 at 03:31, bg...@fr... wrote:

    En r=E9ponse =E0 Chris Borgn=E6s <cbo...@mi...>:
   =20
    > OK, that solved part of the problem.  Thanks, I didn't know the prope=
r
    > gcc options to use.  I still get this error though:
    > as -o BuildWAtkTab.o BuildWAtkTab.s
    > BuildWAtkTab.s: Assembler messages:
    > BuildWAtkTab.s:35: Error: suffix or operands invalid for `mov'
    > BuildWAtkTab.s:44: Error: suffix or operands invalid for `mov'
    >=20
    > Paradoxically, when I remove those #sec:0 references to eliminate the
    > above error, the gcc step still chokes:
    > gcc -o BuildWAtkTab BuildWAtkTab.o -lhtcobol -lhtcobol2 -lm -ldb -ldl
    > BuildWAtkTab.o: In function `main':
    > BuildWAtkTab.o(.text+0x4c): undefined reference to `ww_base0'
    > BuildWAtkTab.o(.text+0x68): undefined reference to `ww_base0'
    > collect2: ld returned 1 exit status
    >=20
   =20
    >From what I can see and what I remember having coded, those
    sections 0 should not appear UNLESS there is something the
    compiler can't handle (some sort of panic!).
   =20
    What is you code looking like? In particular how are
    the files defined? If you post a minimum sample which shows
    the problem, I'll look at it.
   =20
    > On Thu, 2002-08-29 at 17:46, Rildo Pragana wrote:
    >=20
    >     Hi Chris,
    >    =20
    >     On 29 Aug 2002, Chris Borgn=E6s wrote:
    >    =20
    >     > It's been years and years since I really looked at assembler, b=
ut
    > I
    >     > found an oddity.  I try to build my program and I get the
    > following
    >     > error:
    >     > BuildWAtkTab.s: Assembler messages:
    >     > BuildWAtkTab.s:35: Error: suffix or operands invalid for `mov'
    >     > BuildWAtkTab.s:44: Error: suffix or operands invalid for `mov'
    >     >
    >     > I do a htcobol -S and look at the result and this is what I see=
:
    >     >
    >     > .Linite_main:
    >     >  leal	main, %eax
    >     > 	pushl	%eax
    >     > 	leal	.Lend_pgm_main, %eax
    >     > 	pushl	%eax
    >     > 	pushl	$3
    >     > 	movl	$ww_base0+4 #sec:0, %eax
    >     > 	pushl	%eax
    >     > 	movl	$w_base0+4, %eax
    >     > 	pushl	%eax
    >     > 	movl	$s_base0+4, %eax
    >     > 	pushl	%eax
    >     > 	call	cob_open
    >     > 	addl	$16, %esp
    >     > 	pushl	$3
    >     > 	movl	$ww_base0+38 #sec:0, %eax
    >     > 	pushl	%eax
    >     > 	movl	$w_base0+20, %eax
    >     > 	pushl	%eax
    >     > 	movl	$s_base0+38, %eax
    >     > 	pushl	%eax
    >     > 	call	cob_open
    >     >
    >     > The as command seems to work fine once I remove the #sec:0
    > entries, but
    >     > gcc fails:
    >     > gcc -o BuildWAtkTab BuildWAtkTab.o -lhtcobol -lm
    >     > BuildWAtkTab.o: In function `main':
    >     > BuildWAtkTab.o(.text+0x4c): undefined reference to `ww_base0'
    >     > BuildWAtkTab.o(.text+0x68): undefined reference to `ww_base0'
    >     > /usr/local/lib/libhtcobol.a(fileio.o): In function `cob_open':
    >     > fileio.o(.text+0x563): undefined reference to `__db185_open'
    >     > fileio.o(.text+0x5d6): undefined reference to `__db185_open'
    >     > /usr/local/lib/libhtcobol.a(fileio.o): In function `sort_open':
    >     > fileio.o(.text+0x385f): undefined reference to `__db185_open'
    >     > collect2: ld returned 1 exit status
    >     >
    >     > What's wrong?
    >     >
    >    =20
    >     Well, from the errors shown by gcc, it seems you missed to includ=
e
    >     -ldb to link with libdb. It is needed if your program uses any
    > kind
    >     of file access, even if no indexes are in use.
    >    =20
    >     best regards,
    >     Rildo
    >    =20
    > ------------------------------------------------------------------
    >      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
    >    =20
    >    =20
    >    =20
    >     -------------------------------------------------------
    >     This sf.net email is sponsored by:ThinkGeek
    >     Welcome to geek heaven.
    >     http://thinkgeek.com/sf
    >     _______________________________________________
    >     Tin...@li...
    >=20
    > https://lists.sourceforge.net/lists/listinfo/tiny-cobol-users
    >=20
    >    =20
    >=20
   =20
   =20
    Bernard
   =20

Re: [Tiny-cobol-users] Off topic

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

En réponse à David Essex <de...@ar...>:

> Effective August 28 2002, GMT, I will no longer participate in any
> further
> development of TC.
> 
> 
> 
> David Essex
> 

Sorry for my late reaction, but rather late than never!

I am very sad about your decision! And I hope that
you will still find some time to help us with
your comments!

But in any case, I thank you from the whole COBOL
world (intersected or unioned with OpenSource world?)
for all the work that you already did in that project!

Keep well!

Bernard

Re: [Tiny-cobol-users] Extra stuff in assembler

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

En réponse à Chris Borgnæs <cbo...@mi...>:

> OK, that solved part of the problem.  Thanks, I didn't know the proper
> gcc options to use.  I still get this error though:
> as -o BuildWAtkTab.o BuildWAtkTab.s
> BuildWAtkTab.s: Assembler messages:
> BuildWAtkTab.s:35: Error: suffix or operands invalid for `mov'
> BuildWAtkTab.s:44: Error: suffix or operands invalid for `mov'
> 
> Paradoxically, when I remove those #sec:0 references to eliminate the
> above error, the gcc step still chokes:
> gcc -o BuildWAtkTab BuildWAtkTab.o -lhtcobol -lhtcobol2 -lm -ldb -ldl
> BuildWAtkTab.o: In function `main':
> BuildWAtkTab.o(.text+0x4c): undefined reference to `ww_base0'
> BuildWAtkTab.o(.text+0x68): undefined reference to `ww_base0'
> collect2: ld returned 1 exit status
> 

From what I can see and what I remember having coded, those
sections 0 should not appear UNLESS there is something the
compiler can't handle (some sort of panic!).

What is you code looking like? In particular how are
the files defined? If you post a minimum sample which shows
the problem, I'll look at it.

> On Thu, 2002-08-29 at 17:46, Rildo Pragana wrote:
> 
>     Hi Chris,
>     
>     On 29 Aug 2002, Chris Borgnæs wrote:
>     
>     > It's been years and years since I really looked at assembler, but
> I
>     > found an oddity.  I try to build my program and I get the
> following
>     > error:
>     > BuildWAtkTab.s: Assembler messages:
>     > BuildWAtkTab.s:35: Error: suffix or operands invalid for `mov'
>     > BuildWAtkTab.s:44: Error: suffix or operands invalid for `mov'
>     >
>     > I do a htcobol -S and look at the result and this is what I see:
>     >
>     > .Linite_main:
>     >  leal	main, %eax
>     > 	pushl	%eax
>     > 	leal	.Lend_pgm_main, %eax
>     > 	pushl	%eax
>     > 	pushl	$3
>     > 	movl	$ww_base0+4 #sec:0, %eax
>     > 	pushl	%eax
>     > 	movl	$w_base0+4, %eax
>     > 	pushl	%eax
>     > 	movl	$s_base0+4, %eax
>     > 	pushl	%eax
>     > 	call	cob_open
>     > 	addl	$16, %esp
>     > 	pushl	$3
>     > 	movl	$ww_base0+38 #sec:0, %eax
>     > 	pushl	%eax
>     > 	movl	$w_base0+20, %eax
>     > 	pushl	%eax
>     > 	movl	$s_base0+38, %eax
>     > 	pushl	%eax
>     > 	call	cob_open
>     >
>     > The as command seems to work fine once I remove the #sec:0
> entries, but
>     > gcc fails:
>     > gcc -o BuildWAtkTab BuildWAtkTab.o -lhtcobol -lm
>     > BuildWAtkTab.o: In function `main':
>     > BuildWAtkTab.o(.text+0x4c): undefined reference to `ww_base0'
>     > BuildWAtkTab.o(.text+0x68): undefined reference to `ww_base0'
>     > /usr/local/lib/libhtcobol.a(fileio.o): In function `cob_open':
>     > fileio.o(.text+0x563): undefined reference to `__db185_open'
>     > fileio.o(.text+0x5d6): undefined reference to `__db185_open'
>     > /usr/local/lib/libhtcobol.a(fileio.o): In function `sort_open':
>     > fileio.o(.text+0x385f): undefined reference to `__db185_open'
>     > collect2: ld returned 1 exit status
>     >
>     > What's wrong?
>     >
>     
>     Well, from the errors shown by gcc, it seems you missed to include
>     -ldb to link with libdb. It is needed if your program uses any
> kind
>     of file access, even if no indexes are in use.
>     
>     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
>     
>     
>     
>     -------------------------------------------------------
>     This sf.net email is sponsored by:ThinkGeek
>     Welcome to geek heaven.
>     http://thinkgeek.com/sf
>     _______________________________________________
>     Tin...@li...
> 
> https://lists.sourceforge.net/lists/listinfo/tiny-cobol-users
> 
>     
> 


Bernard