tiny-cobol-users

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

[Rildo P. <rpr...@ac...>]

Hi Andrew,

On Mon, 13 Dec 1999, Andrew Cameron wrote:

> I am cc'ing this reply to the list so that the others can also give you
> some input if they so desire.
> 
> Our first target it to implement the full COBOL 74 Standard. After this we
> will implement Cobol 85 Followed by Cobol 2000.

Bravo! I like to see someone optimistic here :)
Even if we have a very long way to cruise...
 
> Our Plan was to make it a GNU Cobol Compiler until we discovered that one
> was in progress and had been for quite a long time.

Yes, but nothing prevent us from a merge with the other projects. Tim
Josling could be preparing our next parser/scanner version  and we release
our tiny beta compiler when it's ready to use.
 
> Please join our cobol mailing list tin...@so... and
> take part in our project. The more people we have the better. The web page
> is http://tiny-cobol.sourceforge.net

I would like to see RMS as a subscriber. Oh, I wanted :)

> To become a developer create yourself a username on
> http://www.sourceforge.net and send an e-mail to the list asking that your
> username be given developer status.

Oh no, Andrew. I don't think he is going to help us. Anyway, you have
caught the attention :)

> You will then have access to the latest Source code etc.

:)   What can I say?

>  On Sun, 12 Dec 1999, Richard Stallman wrote:
                        ^^^^^^^^^^^^^^^^
It's hard to believe, indeed.

> > Date: Sun, 12 Dec 1999 00:33:34 -0700 (MST)
> > From: Richard Stallman <rm...@gn...>
> > To: an...@an...
> > Subject: Re: Tiny Cobol
> > 
> > Hooray, finally!  We have had Cobol on the task list for about a decade.
> > 
> > How much of Cobol does it handle, now?  Are you hoping to handle the
> > whole Cobol language eventually?
> > 
> > What kind of language does this compiler convert the Cobol into?
> > 
> > Do you think you might be interested in making this a GNU package?
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Of course. Yeeeeeees! 

You are doing a great job as a public relations.  
The job's yours, Andrew :)

regards,
 ----------------------------------------------------------------
 Rildo Pragana                 FPGA/uControllers * Linux * tcl/tk
 P.O. Box 721 Camaragibe PE    http://members.xoom.com/rpragana
 Brazil 54792-990              +55-81-459-1776

Re: [Tiny-cobol-users] (no subject)

[Rildo P. <rpr...@ac...>]

Hi Jim,

On Mon, 13 Dec 1999, Noeth, Jim wrote:

> I have been working on the run time routines to facilitate the cobol move
> verb, and have them very close to being finished. My concern, though, is
> that these routines have considerable overhead (especially when moving
> numeric fields) to figure out what type of move is involved, how much to
> pad, how much to truncate, etc.. I'm afraid that our move routines are going
> to be very innefficient. What I would like to do is to modify the 'gen_move'
> routine to generate the code directly, all of the information to do these
> 'overhead' calculations is available at compile time, and we could generate
> code that is very specific to each move statement.

We can do a simple 2-way classification. If the fields are both binaries,
we generate a simple move at compile-time. If not, we call your routine.
What you think?

There are some optimizations I have already introduced, but I can try to
do more. (or you can too, if you want. Look for the functions gen_move,
gen_loadvar, gen_loadloc at the file htcobgen.c)
I really think we are more concerned with having something usable fast.
Then we make gradual performance improvements.

> I realize that this will make the compiler less portable (and a little more
> difficult to debug), I believe that I can write the code generation routines
> such that they can be easily modified for different platforms. Let me know
> what you think.

We can leeave the more efficient code generation for the next version of
the compiler.  We need a beta version soon, to have more contributors
helping us. We only will have a really portable compiler when we output C
language code (or some intermediary language), perhaps with a multi-pass
compiler.

regards,
 ----------------------------------------------------------------
 Rildo Pragana                 FPGA/uControllers * Linux * tcl/tk
 P.O. Box 721 Camaragibe PE    http://members.xoom.com/rpragana
 Brazil 54792-990              +55-81-459-1776

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

[Ted R. <te...@ac...>]

Cobol is COBOL; Gnu is GNU.

First-time surrealists are often confused by the similarities between fish and telephones.

[Tiny-cobol-users] Re: Tiny Cobol

[Andrew C. <an...@an...>]

Hi,

I am cc'ing this reply to the list so that the others can also give you
some input if they so desire.

Our first target it to implement the full COBOL 74 Standard. After this we
will implement Cobol 85 Followed by Cobol 2000.

So far the basic Compiler is up and running with new features being added
every day. Currently there are six of us working on the project. We are
looking for more people.

There is enough of the compiler implemented to write a fair amount of test
programs. 

The current system creates Gnu Assembler which is then assembled and
linked with the library routines. A later version may output C code.

Our Plan was to make it a GNU Cobol Compiler until we discovered that one
was in progress and had been for quite a long time.

Please join our cobol mailing list tin...@so... and
take part in our project. The more people we have the better. The web page
is http://tiny-cobol.sourceforge.net

To become a developer create yourself a username on
http://www.sourceforge.net and send an e-mail to the list asking that your
username be given developer status.

You will then have access to the latest Source code etc.

regards
Andrew
 
 On Sun, 12 Dec 1999, Richard Stallman wrote:

> Date: Sun, 12 Dec 1999 00:33:34 -0700 (MST)
> From: Richard Stallman <rm...@gn...>
> To: an...@an...
> Subject: Re: Tiny Cobol
> 
> Hooray, finally!  We have had Cobol on the task list for about a decade.
> 
> How much of Cobol does it handle, now?  Are you hoping to handle the
> whole Cobol language eventually?
> 
> What kind of language does this compiler convert the Cobol into?
> 
> Do you think you might be interested in making this a GNU package?
> 

-----------------------------------------------------------------------------

Andrew Cameron
Internet : an...@an...
X.400    : C=ZA G=Andrew S=Cameron Admd=TELKOM400

----------------------------------------------------------------------------

[Tiny-cobol-users] Move implementation.

[Ted R. <te...@ac...>]

Well, as a COBOL/S370 assembler fan, I can assure you that is the way the
IBM COBOL compiler works.

Having a 'generalized' MOVE routine would involve much overhead and
require a lot of tweaking and changes whenever someone comes up with a
new idea.

I vote for a 'lean and mean' type-specific MOVE.  It would localize
changes and result in faster code.

First-time surrealists are often confused by the similarities between fish and telephones.

On Mon, 13 Dec 1999, Noeth, Jim wrote:

> To all,
> 
> I have been working on the run time routines to facilitate the cobol move
> verb, and have them very close to being finished. My concern, though, is
> that these routines have considerable overhead (especially when moving
> numeric fields) to figure out what type of move is involved, how much to
> pad, how much to truncate, etc.. I'm afraid that our move routines are going
> to be very innefficient. What I would like to do is to modify the 'gen_move'
> routine to generate the code directly, all of the information to do these
> 'overhead' calculations is available at compile time, and we could generate
> code that is very specific to each move statement.
> 
> I realize that this will make the compiler less portable (and a little more
> difficult to debug), I believe that I can write the code generation routines
> such that they can be easily modified for different platforms. Let me know
> what you think.
> 
> Jim Noeth - SWAT/Technology Services
> A. G. Edwards and Sons, Inc.
> 
> 
> _______________________________________________
> Tiny-cobol-users mailing list
> Tin...@li...
> http://lists.sourceforge.net/mailman/listinfo/tiny-cobol-users
> 

[Tiny-cobol-users] (no subject)

[Noeth, J. <jm...@ag...>]

To all,

I have been working on the run time routines to facilitate the cobol move
verb, and have them very close to being finished. My concern, though, is
that these routines have considerable overhead (especially when moving
numeric fields) to figure out what type of move is involved, how much to
pad, how much to truncate, etc.. I'm afraid that our move routines are going
to be very innefficient. What I would like to do is to modify the 'gen_move'
routine to generate the code directly, all of the information to do these
'overhead' calculations is available at compile time, and we could generate
code that is very specific to each move statement.

I realize that this will make the compiler less portable (and a little more
difficult to debug), I believe that I can write the code generation routines
such that they can be easily modified for different platforms. Let me know
what you think.

Jim Noeth - SWAT/Technology Services
A. G. Edwards and Sons, Inc.

[Tiny-cobol-users] Re: Tiny Cobol

[Andrew C. <an...@an...>]

Hi,

Thank you for informing us of where we have gone wrong in our definitions.

I am forwarding this to our mailing list so that the relevant people can
update our Web page as well as Document.

Regards
Andrew Cameron

On 12 Dec 1999, Jonas Oberg wrote:

> Date: 12 Dec 1999 03:27:43 +0100
> From: Jonas Oberg <jo...@gn...>
> To: Andrew Cameron <an...@an...>
> Subject: Re: Tiny Cobol
> 
> Andrew Cameron <an...@an...> writes:
> 
> > The Url for this Compiler is http://tiny-cobol.sourceforge.net
> 
> It is unfortunate that you refer to this as "freeware". That
> term was first used in the 80'th to mean software that was
> delivered gratis only as binaries with no source code. Today
> it has no particular agreed on definition, so it's best to use
> something else to describe what you mean, such as "free software".
> 
> I also see from your pages that you refer to the operating system
> that your compiler runs on as "Linux", which is unfortunate since
> that is really only the name of the kernel that Linus Torvalds created
> in the early 90'th. The actual operating system is a combination of
> the Linux kernel together with tools that has been created by and for
> the GNU project who has been working for 16 years to create a free
> operating system.  You can give credit to both the GNU project and
> Linus Torvalds by calling this system the GNU/Linux system. Or a
> Linux-based GNU system.
> 

-----------------------------------------------------------------------------

Andrew Cameron
Internet : an...@an...
X.400    : C=ZA G=Andrew S=Cameron Admd=TELKOM400

----------------------------------------------------------------------------

[Tiny-cobol-users] Re: Tiny Cobol (fwd)

[Andrew C. <an...@an...>]

Hi,

I have asked the gnu project to list us on their web page.

They have done so.

see below

---------- Forwarded message ----------
Date: Sat, 11 Dec 1999 20:05:04 -0600
From: Neelakanth Nadgir <ne...@gn...>
To: Andrew Cameron <an...@an...>
Cc: web...@ww...
Subject: Re: Tiny Cobol


On Sat, 11 Dec 1999, Andrew Cameron wrote:

> Hi,
> 
> Please add a link on your  page
> http://www.gnu.org/software/software.html to our Tiny Cobol Compiler.
> The Url for this Compiler is http://tiny-cobol.sourceforge.net
> 
> It is released under GPL and is currently under development.
> 
> 

thanks or your submission. I have added it to our software page.

-neelakanth
(ne...@gn...)

RE: [Tiny-cobol-users] Changes in switches.

[Rildo P. <rpr...@ac...>]

Hi Glen,

On Sun, 12 Dec 1999, Glen Colbert wrote:

> I changed all of the relevent Makefiles in the test.code directory.  I think
> this fixes what you were after.  If I missed someplace and you notice,
> please let me know.

Thank you. Was exactly that.

regards,
 ----------------------------------------------------------------
 Rildo Pragana                 FPGA/uControllers * Linux * tcl/tk
 P.O. Box 721 Camaragibe PE    http://members.xoom.com/rpragana
 Brazil 54792-990              +55-81-459-1776

RE: [Tiny-cobol-users] Changes in switches.

[Glen C. <gco...@US...>]

> -----Original Message-----
> From: tin...@so...
> [mailto:tin...@so...]On Behalf Of Rildo
> Pragana
> Sent: Sunday, December 12, 1999 7:00 PM
> To: tin...@ma...
> Subject: [Tiny-cobol-users] Changes in switches.
> Glen have done some changes in the compiler's switches, so almost all
> makefiles are wrong.  Please someone correct it soon or we will have
> problems with the beta (alpha?) testers.
>
I changed all of the relevent Makefiles in the test.code directory.  I think
this fixes what you were after.  If I missed someplace and you notice,
please let me know.

Glen

[Tiny-cobol-users] Changes in switches.

[Rildo P. <rpr...@ac...>]

Hi,

Glen have done some changes in the compiler's switches, so almost all
makefiles are wrong.  Please someone correct it soon or we will have
problems with the beta (alpha?) testers.

I have also changed a little the code for performs, to eliminate the
exit_paragraph function call. This will not be perceived by you, except
that the program is faster yet, a little larger (but just some
bytes per paragraph) and easier to debug. I have done this because there
was no way for gdb placing a breakpoint when single-stepping.  It seems to
be working now (well, almost, at least in some cases).

regards,
 ----------------------------------------------------------------
 Rildo Pragana                 FPGA/uControllers * Linux * tcl/tk
 P.O. Box 721 Camaragibe PE    http://members.xoom.com/rpragana
 Brazil 54792-990              +55-81-459-1776

RE: [Tiny-cobol-users] Newbie compile "challenges"

[Rildo P. <rpr...@ac...>]

Hi Glen,

On Sun, 12 Dec 1999, Glen Colbert wrote:

> The first cause of this problem is that there is more than one libdb in the
> Linux distributions ;(.  One is an MIT library that is used for medical
> research using real-time monitoring (I think) of cardiac patients.  The
> other is the Berkley DB library.
> 
> I am currently using libdb-1.85.4 (the a.out static library).  When I frist
> installed this however, the db.h file in /usr/include was for the wrong
> library.

The RedHat distro have a  db_185.h include file that could replace our
db.h at the htcoblib.h header file. I suggest you make something like

#if LIB_DB_185
#include <db_185.h>
#else
#include <db.h>
#endif

But, unfortunately, I don't have a RedHat system to test. (mine is home
grown).

> The binary distributions of libdb-1.85-4 include the correct header file and
> a usable library.  You may want to move these to /usr/include and /lib
> before trying to build.

If you don't have the file db_185.h, you can rename the libdb's db.h and
place it there (/usr/include).
 
> I have moved my Cobol libraries and executables to a directory
> /opt/cobol/{bin|lib} including the libdb stuff and changed my -I and -L
> directives to point here.  This (I hope) will prevent another package from
> installing the wrong libraries over libdb.

Again, you can make something like the above at the Makefile, so we have
only one makefile.
Please, as you have a RedHat, do this and make the patches availbale
through our cvs server.  If you need assistance in modifying the makefile,
I can try to help you.  We can also have 2 makefiles: the default and
another "Makefile.RedHat", so the user do "make -f Makefile.RedHat" to
compile it.

> I also had several problems when using mixed static and dynamic libraries.
> I have currently removed the dynamic (libcobol,libdb,..so) libraries from my
> development environment.

Perhaps you didn't put the library in your ld.so directory path. You must
check the file /etc/ld.so.conf to see where it looks for libraries.  If
you can't (because ld.so.conf is not there), the safer place is /usr/lib.
Put your library there and do "ldconfig" to put it in the cache.

This way, we can make more friendly the installation of Tiny Cobol :)
Please report if you succeed.

regards,
 ----------------------------------------------------------------
 Rildo Pragana                 FPGA/uControllers * Linux * tcl/tk
 P.O. Box 721 Camaragibe PE    http://members.xoom.com/rpragana
 Brazil 54792-990              +55-81-459-1776

[Tiny-cobol-users] Introduction

[Ted R. <arc...@db...>]

This message was sent from Geocrawler.com by "Ted Rolle" <te...@ac...>
Be sure to reply to that address.

I'm a geek.  Pure and simple.
I've coded COBOL since the late 1960s to the
present.
I'm currently working with COBOL, S/370 assembler,
C (not C++), PHP4, HTML, LaTeX...

Geocrawler.com - The Knowledge Archive

RE: [Tiny-cobol-users] Newbie compile "challenges"

[Ted R. <te...@ac...>]

Thanks for the help... off to help someone move...  I'll get back to it
tonight, probably.

First-time surrealists are often confused by the similarities between fish and telephones.

RE: [Tiny-cobol-users] Newbie compile "challenges"

[Glen C. <gco...@US...>]

Hi Ted.  Been here, done this...

> -----Original Message-----
> From: tin...@so...
> [mailto:tin...@so...]On Behalf Of Ted Rolle
> Sent: Sunday, December 12, 1999 9:16 AM
> Here is the result of my compile of Tiny COBOL:
>
> Making all in compiler
> make[1]: Entering directory `/home/ted/development/compiler'
> make[1]: Nothing to be done for `all'.
> make[1]: Leaving directory `/home/ted/development/compiler'
> Making all in lib
> make[1]: Entering directory `/home/ted/development/lib'
> gcc  -g -Wall -DPICTURE_TESTING  -I/usr/include -I../compiler -I.
> -c fileio.c
> fileio.c: In function `cob_open':
> fileio.c:106: warning: implicit declaration of function `dbopen'
> fileio.c:106: warning: assignment makes pointer from integer


The first cause of this problem is that there is more than one libdb in the
Linux distributions ;(.  One is an MIT library that is used for medical
research using real-time monitoring (I think) of cardiac patients.  The
other is the Berkley DB library.

I am currently using libdb-1.85.4 (the a.out static library).  When I frist
installed this however, the db.h file in /usr/include was for the wrong
library.

The binary distributions of libdb-1.85-4 include the correct header file and
a usable library.  You may want to move these to /usr/include and /lib
before trying to build.

I have moved my Cobol libraries and executables to a directory
/opt/cobol/{bin|lib} including the libdb stuff and changed my -I and -L
directives to point here.  This (I hope) will prevent another package from
installing the wrong libraries over libdb.

I also had several problems when using mixed static and dynamic libraries.
I have currently removed the dynamic (libcobol,libdb,..so) libraries from my
development environment.

I hope this helps.

Glen

RE: [Tiny-cobol-users] END-IF

[Ted R. <te...@ac...>]

I think that the ANSI-85 extensions that I'd miss the most are
reference modification, END-IF, CONTINUE, and END-PERFORM.

Reference modification:

	05  I                     PIC S9(07) COMP SYNC.
	05  J                     PIC S9(07) COMP SYNC.
	05  K                     PIC S9(07) COMP SYNC.
	05  L                     PIC S9(07) COMP SYNC.
	05  FIELD-1               PIC  X(08).
	05  FIELD-2               PIC  X(24).

MOVE FIELD-1 (1:5) TO FIELD-2 (3:I)
MOVE FIELD-1 (I:J) TO FIELD-2 (K:L)

Is everyone aware that there is an on-line COBOL language reference at IBM
Boulder?  I don't have the URI handy here at home, but do have it at work.
I'll post it from there.

First-time surrealists are often confused by the similarities between fish and telephones.

On Sun, 12 Dec 1999, Ted Rolle wrote:

> Well, just as long as we don't paint ourselves into a corner on the
> ANSI-85 grammar...
> 
> First-time surrealists are often confused by the similarities between fish and telephones.
> 
> 
> 
> _______________________________________________
> Tiny-cobol-users mailing list
> Tin...@li...
> http://lists.sourceforge.net/mailman/listinfo/tiny-cobol-users
> 

[Tiny-cobol-users] Newbie compile "challenges"

[Ted R. <te...@ac...>]

I'm keeping track of all of these problems so that they can be answered in
the README/INSTALL.  Now, as soon as I get CVS access... ;->

Here is the result of my compile of Tiny COBOL:

Making all in compiler
make[1]: Entering directory `/home/ted/development/compiler'
make[1]: Nothing to be done for `all'.
make[1]: Leaving directory `/home/ted/development/compiler'
Making all in lib
make[1]: Entering directory `/home/ted/development/lib'
gcc  -g -Wall -DPICTURE_TESTING  -I/usr/include -I../compiler -I. -c fileio.c
fileio.c: In function `cob_open':
fileio.c:106: warning: implicit declaration of function `dbopen'
fileio.c:106: warning: assignment makes pointer from integer without a cast
fileio.c: In function `cob_close':
fileio.c:131: too few arguments to function
fileio.c: In function `cob_read':
fileio.c:144: `recno_t' undeclared (first use in this function)
fileio.c:144: (Each undeclared identifier is reported only once
fileio.c:144: for each function it appears in.)
fileio.c:144: parse error before `recno'
fileio.c:160: `recno' undeclared (first use in this function)
fileio.c:160: parse error before `)'
fileio.c:160: parse error before `)'
fileio.c:198: warning: passing arg 2 from incompatible pointer type
fileio.c:198: warning: passing arg 4 makes pointer from integer without a cast
fileio.c:198: too few arguments to function
fileio.c: In function `cob_read_next':
fileio.c:209: `R_NEXT' undeclared (first use in this function)
fileio.c:237: structure has no member named `seq'
fileio.c: In function `cob_read_prev':
fileio.c:248: `R_PREV' undeclared (first use in this function)
fileio.c:284: structure has no member named `seq'
fileio.c: In function `cob_write':
fileio.c:296: `recno_t' undeclared (first use in this function)
fileio.c:296: parse error before `recno'
fileio.c:317: `recno' undeclared (first use in this function)
fileio.c:317: parse error before `)'
fileio.c:317: parse error before `)'
fileio.c:342: warning: passing arg 2 from incompatible pointer type
fileio.c:342: warning: passing arg 4 makes pointer from integer without a cast
fileio.c:342: too few arguments to function
fileio.c: In function `cob_delete':
fileio.c:353: `recno_t' undeclared (first use in this function)
fileio.c:353: parse error before `recno'
fileio.c:372: `recno' undeclared (first use in this function)
fileio.c:372: parse error before `)'
fileio.c:372: parse error before `)'
fileio.c:386: warning: passing arg 2 from incompatible pointer type
fileio.c:386: warning: passing arg 3 makes pointer from integer without a cast
fileio.c:386: too few arguments to function
fileio.c: In function `cob_start':
fileio.c:395: `recno_t' undeclared (first use in this function)
fileio.c:395: parse error before `recno'
fileio.c:396: `R_CURSOR' undeclared (first use in this function)
fileio.c:410: `recno' undeclared (first use in this function)
fileio.c:410: parse error before `)'
fileio.c:410: parse error before `)'
fileio.c:424: structure has no member named `seq'
fileio.c: In function `cob_rewrite':
fileio.c:443: `recno_t' undeclared (first use in this function)
fileio.c:443: parse error before `recno'
fileio.c:479: `recno' undeclared (first use in this function)
fileio.c:479: parse error before `)'
fileio.c:479: parse error before `)'
fileio.c:502: warning: passing arg 2 from incompatible pointer type
fileio.c:502: warning: passing arg 4 makes pointer from integer without a cast
fileio.c:502: too few arguments to function
fileio.c: In function `sort_open':
fileio.c:561: `BTREEINFO' undeclared (first use in this function)
fileio.c:561: parse error before `b'
fileio.c:563: `b' undeclared (first use in this function)
fileio.c:563: `R_DUP' undeclared (first use in this function)
fileio.c:584: warning: assignment makes pointer from integer without a cast
fileio.c: In function `sort_release':
fileio.c:633: warning: passing arg 2 from incompatible pointer type
fileio.c:633: warning: passing arg 4 makes pointer from integer without a cast
fileio.c:633: too few arguments to function
fileio.c: In function `sort_return':
fileio.c:643: `R_NEXT' undeclared (first use in this function)
fileio.c:645: structure has no member named `seq'
make[1]: *** [fileio.o] Error 1
make[1]: Leaving directory `/home/ted/development/lib'
make: *** [all] Error 2

>-------------------------------------------------------------- db.0

Here's the result of the db compile:
cc -c -D__DBINTERFACE_PRIVATE -O6 -m486 -fomit-frame-pointer -I. -Iinclude -I../../hash ../../hash/hash.c -o static/hash.o
In file included from ../../hash/hash.c:55:
../../hash/hash.h:106: field `__errno_location' declared as a function
../../hash/hash.c: In function `flush_meta':
../../hash/hash.c:508: parse error before `('
../../hash/hash.c: In function `hash_get':
../../hash/hash.c:539: parse error before `('
../../hash/hash.c: In function `hash_put':
../../hash/hash.c:556: parse error before `('
../../hash/hash.c:560: parse error before `('
../../hash/hash.c: In function `hash_delete':
../../hash/hash.c:577: parse error before `('
../../hash/hash.c:581: parse error before `('
../../hash/hash.c: In function `hash_seq':
../../hash/hash.c:732: parse error before `('
make: *** [hash.o] Error 1



First-time surrealists are often confused by the similarities between fish and telephones.

RE: [Tiny-cobol-users] END-IF

[Ted R. <te...@ac...>]

Well, just as long as we don't paint ourselves into a corner on the
ANSI-85 grammar...

First-time surrealists are often confused by the similarities between fish and telephones.

RE: [Tiny-cobol-users] END-IF

[Glen C. <gco...@US...>]

> -----Original Message-----
> From: tin...@so...
> [mailto:tin...@so...]On Behalf Of Ted Rolle
> Subject: [Tiny-cobol-users] END-IF
>
> Is there one?

I hope this answer isn't too discouraging, but.....

The scope of _this_ phase of the Tiny-Cobol project is to implement an
ANSI-74 release of Cobol.  As such, ANSI-85 scope terminators such as END-IF
and END-PERFORM are not a part of the current development effort.  Once we
have a functional release of the earlier Cobol grammar, we have planned to
implement the newer 85 release.

HOWEVER - Much of the ANSI-85 grammar is included in the compiler source
code.  It is not as if we will need to do a redesign to include the newer
Cobol features.

Glen

RE: [Tiny-cobol-users] RPM Experience

[Glen C. <gco...@US...>]

> -----Original Message-----
> From: tin...@so...
> [mailto:tin...@so...]On Behalf Of Ted Rolle
> Sent: Saturday, December 11, 1999 10:48 PM
>
> Um... it takes all of those?
>
> I'm still trying to get it to compile...
>
> > byacc, lex, and all of the other tools.
> >
>
Ted,
  You make my point very well.  People who want to use 'Cobol' aren't
necessisarily concerned with making a compiler, they just want to compile
Cobol source code.  It would be nice if we could release a binary executable
of the compiler for those with this level of interest. It can be very
discouraging to have to hunt around for the right version of make, lex,
yacc, libdb, cobpp, gcc, gas, and all of the other tools that it takes to
put the compiler together.

 It would be nice to release an RPM package that just installs a Cobol
compiler, ready to compile _Cobol_ source code, on the user's system.  I've
started work on this, but I keep getting build error messages.  It would be
nice to comunicate with someone who knows how to set up the control files
for an rpm build.

By the way Ted, If you are having problems, please send me copies of your
output and I'll do what I can to help you get a compiler compiled.

Glen

[Tiny-cobol-users] END-IF

[Ted R. <te...@ac...>]

Is there one?

First-time surrealists are often confused by the similarities between fish and telephones.

Re: [Tiny-cobol-users] RPM Experience

[Ted R. <te...@ac...>]

Um... it takes all of those?

I'm still trying to get it to compile...

First-time surrealists are often confused by the similarities between fish and telephones.

> byacc, lex, and all of the other tools.
> 

[Tiny-cobol-users] Command Line Option Changes

[Glen C. <gco...@US...>]

Please don't throw things at me...

I removed the -l -L command line options.  This may require a rework of some
of your make files.  These options will be used to pass information to the
linker in the near future.  The old -L function is now accessed by -P.

I also added -D to generate and keep debugging information, and -B
static|dynamic

Glen

[Tiny-cobol-users] va_start help

[Glen C. <gco...@US...>]

I am trying to build on Solaris, but I'm having a problem with the varargs
stuff.  I keep getting errors related to __va_builtin_ stuff.  I have gotten
close, (compiled, but link-edit of cobol code fails).

Any hints as to what I need to do to get the varargs stuff to work on
Solaris??

Glen

[Tiny-cobol-users] RPM Experience

[Glen C. <gco...@US...>]

Does anyone have any experience making RPM packages?  I think we should make
an RPM of the compiled executable for distribution for those who want to
use/test Cobol, but do not want to go through the hastle of setting up
byacc, lex, and all of the other tools.

Glen