Decompile issues

[Patrick Jackman]

I have a 40 MB Access 2002 front end. I use the decompile, compact,
recompile and compact procedure prior to releasing new versions to keep the
application size to a minimum.

Lateley I've been getting run time errors after using this process. There
are no compile errors as the project will successfully compile.

I traced one of the problems and found that a string litteral passed byval
to a function was resulting in an empty string within the function. As an
example:

Function SomeFunction(ByVal vstrParam As String) As Long
SomeFunction = Len(vstrParam)
End Sub

?SomeFunction("A String")
results in Zero!

My workaround today was to release the application decompiled and it seems
to be running without any problems.

Has anyone else run into wacky issues with decompile/recompile?

Has anyone run into wacky issues after installing WinXP SP3?

Patrick
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Patrick Jackman
Vancouver, BC

 

[Klatuu]

eliminate the Decompile step. The Decompile is really only a tool to try to
retieve a corrupt database. It can cause problems.

 

[Patrick Jackman]

Thanks for the advice. I thought the use of decompile was a common practice.
I've been using it for several years without issues until now.

Patrick.

eliminate the Decompile step. The Decompile is really only a tool to try to
retieve a corrupt database. It can cause problems.

 

[Douglas J. Steele]

See what MichKa says about decompiling at
http://www.trigeminal.com/usenet/usenet004.asp , specifically the Risks To
Decompile near the end.

 

[Patrick Jackman]

Great article Doug. Thanks for the link.

Patrick.

See what MichKa says about decompiling at
http://www.trigeminal.com/usenet/usenet004.asp , specifically the Risks To
Decompile near the end.

 

[David W. Fenton]

See what MichKa says about decompiling at
http://www.trigeminal.com/usenet/usenet004.asp , specifically the
Risks To Decompile near the end.

In my experience, while Michael's point is well-taken, it doesn't
happen often. The only time I've ever lost anything in a decompile
was when something was already corrupted. I decompile regularly and
have not lost anything, ever, from that. It keeps your code running
clean and flushes out problems as soon as they develop. One thing,
perhaps, that many people overlook is that they definitely need to
turn off COMPILE ON DEMAND, as during development, this vastly
increases the chance of code corruption.

 

[Patrick Jackman]

This is the first I have heard about turning off COMPILE ON DEMAND. I've
ignored this setting so it's always been True and BACKGROUND COMPILE also
True. So thanks very much for that information. Where can I read more about
this David?

Patrick.

See what MichKa says about decompiling at
http://www.trigeminal.com/usenet/usenet004.asp , specifically the
Risks To Decompile near the end.

In my experience, while Michael's point is well-taken, it doesn't
happen often. The only time I've ever lost anything in a decompile
was when something was already corrupted. I decompile regularly and
have not lost anything, ever, from that. It keeps your code running
clean and flushes out problems as soon as they develop. One thing,
perhaps, that many people overlook is that they definitely need to
turn off COMPILE ON DEMAND, as during development, this vastly
increases the chance of code corruption.

 

[David W. Fenton]

This is the first I have heard about turning off COMPILE ON
DEMAND. I've ignored this setting so it's always been True and
BACKGROUND COMPILE also True. So thanks very much for that
information. Where can I read more about this David?

MichKa's article on decompile basically explains why at the
beginning, in teh section on all the states of decompilation.
Turning off COMPILE ON DEMAND means that *you* have more control
over when your code is compiled.