tiny-cobol-users

[Tiny-cobol-users] Compiler crash on COPY statement

[Jim M. <ro...@vi...>]

This is a follow up to my previous post - which has not hit the mailing 
list yet because I sent it from the wrong email account:)

I have just compiled another program that uses "COPY GLBAL-IO-WS" and 
the compiler crashed again with a Segmentation error.

I copied the copybook to create a new copybook called "STUFFX" and then 
modified the program to read "COPY STUFFX".
The program now compiles.

Is there anything suspicious about the name "GLBAL-IO-WS"?  I have 
copybooks with names like "GLACCNT-IO-WS" and "GLTRAN-IO-WS" and these 
compile okay.

Jim

Re: [Tiny-cobol-users] Compiler crash on COPY statement

[Walter G. <ga...@dm...>]

Hi,

yesterday when i send my suggestion to the list i found the same error...
if you compile the same program in the same directory where you have the 
copy files (don´t use -I or global variable), you will not receive the 
segment fault.
i tried to search the error, but it occurs in parser when it tries to 
allocate 16k and return the pointer.
i only will have time to continue search the allocation problem next 
week....

walter



----- Original Message ----- 
From: "Jim Morcombe" <ro...@vi...>
To: "tiny-cobol-users" <tin...@li...>
Sent: Thursday, August 11, 2005 1:59 AM
Subject: [Tiny-cobol-users] Compiler crash on COPY statement


> This is a follow up to my previous post - which has not hit the mailing 
> list yet because I sent it from the wrong email account:)
>
> I have just compiled another program that uses "COPY GLBAL-IO-WS" and the 
> compiler crashed again with a Segmentation error.
>
> I copied the copybook to create a new copybook called "STUFFX" and then 
> modified the program to read "COPY STUFFX".
> The program now compiles.
>
> Is there anything suspicious about the name "GLBAL-IO-WS"?  I have 
> copybooks with names like "GLACCNT-IO-WS" and "GLTRAN-IO-WS" and these 
> compile okay.
>
> Jim
>
>
>
> -------------------------------------------------------
> SF.Net email is Sponsored by the Better Software Conference & EXPO
> September 19-22, 2005 * San Francisco, CA * Development Lifecycle 
> Practices
> Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
> Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
> _______________________________________________
> Tin...@li...
> https://lists.sourceforge.net/lists/listinfo/tiny-cobol-users
>
>
>
> -- 
> No virus found in this incoming message.
> Checked by AVG Anti-Virus.
> Version: 7.0.338 / Virus Database: 267.10.6/69 - Release Date: 11/8/2005
> 

Re: [Tiny-cobol-users] Compiler crash on COPY statement

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

I could not duplicate any of the problems you described using version 0.63.

I think you may be experiencing an intermittent bug in the TC 
pre-processor reported on several occasions, and which I encountered a 
while back.

Unfortunately, I never did isolate the problem as any small change in an 
option, search path or copybook line and the problem would disappear.

To make matters even more frustrating, the conditions (code) which 
caused the bug on one system usually could not be duplicated on another.

And to make matters really frustrating, the same conditions (code) which 
had caused a bug on my system, would no longer duplicate the problem 
several day later.

Anyway, I think the problem is with the a stray pointer(s) in the 
copybook search sequence code.

The only thing I can suggest is to try some work-around, such as the 
following.
1) Enclosed the copybook names in quotes.
COPY 'some-copybook'.
2) Add an extra path delimiter at the end of the path.
... -I $COBCPY: -I $IO: gltest.cbl
3) Add an extra dummy path at the end of the path.
... -I $COBCPY:dummy -I $IO:dummy gltest.cbl

The TC pre-processor code needs to be cleaned up, and better integrated 
into the main compiler, but that is another story.

Re: [Tiny-cobol-users] Compiler crash on COPY statement

[Jim M. <ro...@vi...>]

Thanks Dave

I'll try your work-arounds.

>
> And to make matters really frustrating, the same conditions (code) 
> which had caused a bug on my system, would no longer duplicate the 
> problem several day later.

I'm glad it happenned to you as well - I thought I was really losing it:)

Jim