Re: Forms 3.0 fortran user-exit problem

From: Mike Crawford <mcrawfor_at_bio.ri.ccf.org>
Date: Fri, 12 Aug 1994 11:56:11 GMT
Message-ID: <1994Aug12.115611.21482_at_bme.ri.ccf.org>


In article 3vj_at_b8.b8.ingr.com, mlcampb1_at_mikec.b8.ingr.com (Mike Campbell) writes:
>
>Well, to make a long story short, ORACLE support won't return my phone
>calls about a logged TAR; can't even get a call back after raising the
>prioity to critical. It's been 10 days now and the TAR still hasn't been
>assigned to anyone yet so I am going to try the net to see if anyone can
>answer my question.
>
>The basic problem is that when I compile a Pro*FORTRAN user exit into
>forms 3.0 and execute it it will core dump whenever I try to open a file
>and write to it. I know that a Pro*C user exit will work fine, but can't
>seem to get a FORTRAN to work. The user exit I am using is the sample
>concat.pfo one that gets delivered and I am just adding in a few lines
>of code to open a file, write a text string to it, and close the file.
>When the user exit fires, it creates the file, but core dumps before
>anything gets written to it. By the way, the user exit works fine
>without my additional code so I know that the problem is in what I am
>trying to do.

What operating system are you running under ?  Unix ?

If it is unix, here's a potential 'work around' to try until you get a real answer. ( since I don't have a answer for the real problem)

Write the Pro*Fortran routine as a stand-alone program that can be run outside the form. Where you would normally call the user-exit, replace it with a HOST(command_string, NO_SCREEN) call. This will call the Pro*Fortran program as if you were running it from the system prompt.

This was our solution to eliminating user-exits altogether.

Good luck, hope you find a solution to the real problem.

mwc


"you don't pay me to turn a screw, you pay be to know what screw to turn." Received on Fri Aug 12 1994 - 13:56:11 CEST

Original text of this message