Re: Strange Compile

Steve,

Any chance that this program uses named Activation Group?  Also, what Debug
view did you compile with?

Suppose your program used a named AG (QILE) with debug view *SOURCE.  The
first time you run this application, it loads to AG QILE, and any subsequent
calls to that program will execute the already loaded version of this
program.  You made some changes to the source, recompiled, and tried to
debug the program again.  Unless you specifically reclaimed AG QILE, you're
still running the original version of this program, but now the source
member is different from the source used to compile the original program....

Another issue that has bit me is forgetting to ENDDBG before recompiling the
program....  

Also consider the OPTIMIZE(*FULL) parm...  This has significant effects on
how debug works for this program.  IMO, while developing an application, you
should not use any optimization, until you have completed your testing.
Even then, I'm not inclined to use full optimization, because it limits your
ability to use the debugger to diagnose errors in a production app.  




Eric DeLong
Sally Beauty Company
MIS-Project Manager (BSG)
940-297-2863 or ext. 1863



-----Original Message-----
From: wdsci-l-bounces@xxxxxxxxxxxx
[mailto:wdsci-l-bounces@xxxxxxxxxxxx]On Behalf Of sjones@xxxxxxxxxxxxxx
Sent: Friday, February 17, 2006 9:14 AM
To: wdsci-l@xxxxxxxxxxxx
Subject: [WDSCI-L] Strange Compile


I was having a problem yesterday with a program I was developing  when I 
tried to run it a divide by zero error would occur, but it was pointing to 
a line that was not dividing. 

First thing I thought was somehow I was running an old version, so I 
logged-off/back-on, problem still kept occurring.  I was compiling the 
program using RSE , received the message that it compiled w/no errors.  I 
decided to compile it using PDM and after that I seemed to get the correct 
version of the program when I ran it.   I compared the compile options on 
RSE vs. PDM and the only difference I see is *Eventf, Optimization *Full 
and replace program *yes in the RSE.

Anyone have an idea what was causing the compile in RSE to not replace the 
program?

Thanks

Steve Jones