For Programmers: Free Programming Magazines  


Home > Archive > Clipper > May 2005 > Recompile questions









You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

 

Author Recompile questions
Chris TenBraak

2005-05-10, 8:55 pm

Its been a long time since I had to work on someone elses code so here I go.

De-compiled at 5.01 app. Added some basic SET PRINTER TO code to clean up
despooling.

When I try to recompile using 5.2e I get the following errors:

..RTLink for Clipper Dynamic Overlay Linker / Pre-Linker Version 3.14B
(C) Copyright Pocket Soft Inc., 1988-1991. All Rights Reserved.

warning wrt0011: Public symbol '__GETCLASS' doubly defined

UNDEFINED SYMBOL(S) AFTER LIBRARY SEARCH:
SYMBOL FIRST REFERENCE
------ ---------------
'READKILL' VIC.OBJ
'__TOTBROWS' VIC.OBJ
'__TOTBCOLU' VIC.OBJ
'__COLOR' VIC.OBJ
'__GETSETSU' CLIPPER.LIB

warning wrt0022: .EXE may not execute properly -- undefined symbols
280K
2 warning message(s)

Of course it won't execute.
Any thoughts?
TIA
Chris


Ross McKenzie

2005-05-10, 8:55 pm

On Thu, 5 May 2005 21:39:02 -0500, "Chris TenBraak"
<ctenbraak@cox.net> wrote:

>Its been a long time since I had to work on someone elses code so here I go.
>
>De-compiled at 5.01 app. Added some basic SET PRINTER TO code to clean up
>despooling.
>
>When I try to recompile using 5.2e I get the following errors:
>
>.RTLink for Clipper Dynamic Overlay Linker / Pre-Linker Version 3.14B
>(C) Copyright Pocket Soft Inc., 1988-1991. All Rights Reserved.
>
>warning wrt0011: Public symbol '__GETCLASS' doubly defined
>
>UNDEFINED SYMBOL(S) AFTER LIBRARY SEARCH:
> SYMBOL FIRST REFERENCE
> ------ ---------------
> 'READKILL' VIC.OBJ
> '__TOTBROWS' VIC.OBJ
> '__TOTBCOLU' VIC.OBJ
> '__COLOR' VIC.OBJ
> '__GETSETSU' CLIPPER.LIB
>
>warning wrt0022: .EXE may not execute properly -- undefined symbols
>280K
>2 warning message(s)
>
>Of course it won't execute.
>Any thoughts?
>TIA
>Chris
>


Hello Chris,

I have done "one or two" of these before <g>.

As I understand it, you have Clipper 5.2e plus the source recovered
via decompilation.

First step is to remove (to a safe place) all of the source code whose
function names start with the double underscore. These are internals
from your original 5.01 compiler or associated libraries. 5.2e will
automatically substitute its own versions as needed.

Unless you are certain that the original errorsys and getsys had been
altered for some special functionality, eg timed exits from a read,
also remove them. 5.2e will use its own.

Then recompile and identify what functions, if any, are causing a
failure to rebuild. Then come back here...or feel free to email me.

Regards,

Ross McKenzie
ValuSoft
Melbourne Australia

valusoft AT optushome DOT com DOT au

So long ... and thanks for all the fish.
Chris TenBraak

2005-05-10, 8:55 pm

You know, I could of re-written this monster in S87 in the amount of time
I've devoted to trying to re-compile this app.

I've cleaned up a bunch but I'll probably have to start over somewhat now
that I'm getting a feel for this.

It's still throwing a fit over READKILL.

"Ross McKenzie" <NoJunk_valusoft@optushome.com.au> wrote in message
news:427bfdf2.1329765@news...
> On Thu, 5 May 2005 21:39:02 -0500, "Chris TenBraak"
> <ctenbraak@cox.net> wrote:
>
go.[color=darkred]
up[color=darkred]
>
> Hello Chris,
>
> I have done "one or two" of these before <g>.
>
> As I understand it, you have Clipper 5.2e plus the source recovered
> via decompilation.
>
> First step is to remove (to a safe place) all of the source code whose
> function names start with the double underscore. These are internals
> from your original 5.01 compiler or associated libraries. 5.2e will
> automatically substitute its own versions as needed.
>
> Unless you are certain that the original errorsys and getsys had been
> altered for some special functionality, eg timed exits from a read,
> also remove them. 5.2e will use its own.
>
> Then recompile and identify what functions, if any, are causing a
> failure to rebuild. Then come back here...or feel free to email me.
>
> Regards,
>
> Ross McKenzie
> ValuSoft
> Melbourne Australia
>
> valusoft AT optushome DOT com DOT au
>
> So long ... and thanks for all the fish.



Ray Marron

2005-05-10, 8:55 pm

Can you get a copy of 5.01 to recompile it? Otherwise, you're going to
have to do a lot more than add some printer code in order to get it to
work in 5.2 -- you'll have to find out where __GETCLASS is defined in
the code and figure how it differs from the 5.2 implementation, as well
as figure out what all those undefined symbols are supposed to do.

If you had the original source code, I imagine those differences would
most likely be resolved by the preprocessor. That may be the first
place to look: the 5.01 .ch files to see what commands preprocess to
those symbols.

Good luck!

Ross McKenzie

2005-05-10, 8:55 pm

On Fri, 6 May 2005 21:38:09 -0500, "Chris TenBraak"
<ctenbraak@cox.net> wrote:

>You know, I could of re-written this monster in S87 in the amount of time
>I've devoted to trying to re-compile this app.
>
>I've cleaned up a bunch but I'll probably have to start over somewhat now
>that I'm getting a feel for this.
>
>It's still throwing a fit over READKILL.
>


Chris,

READKILL is a function within the Clipper.lib file, so it should be
finding it. So...can you be more specific about its complaint?

Alternatively, send me your .lnk file and the prg that has this
readkill code used. I may see something behind your trees.

Regards,

Ross McKenzie
ValuSoft
Melbourne Australia

valusoft AT optushome DOT com DOT au

So long ... and thanks for all the fish.
Sponsored Links







Also available: Server administration forum archive | Web Design forum archive | Software forum archive | Hardware reviews archive

Copyright 2008 codecomments.com