dolf@fwi.uva.nl (Dolf Starreveld) (11/06/89)
I am currently working on a program that needs to copy a 'dctb' resource
from one file to another. Since the program must also copy other resources
I have written a general purpose "CopyResource" routine. The routine itself
was extensively tested and seemed to work, but .....
The copying of the 'dctb' resource was only recently added and things
started to go wrong. After lots of debugging I found out that after executing
the following statement:
inHandle = GetResource('dctb', 10000);
"inHandle" contains a valid handle, but not a handle to a resource. I entered
TMON to find out more and:
1) Just before executing the statement the 'dctb' resource is not
loaded in the heap.
2) Just after the statement, the 'dctb' resource is loaded, but
immediately following it, the heap contains another relocatable
block of exactly the same size of the resource (48 bytes).
3) The GetResource call returns a handle to this last block instead
of to the actually loaded resource.
Since my program continues to do a GetResInfo call on the handle returned,
I get into problems. The handle is not a handle to a resource and the
call sets ResError != 0. Furthermore, the &resID is set to -1 and the
&resType is set to four zero bytes. This is what finally causes another
program using the copied resource file to crash.
I moved the GetResource call to very early in my program (just after ToolBox
initialization), just to be sure the heap was not already corrupt. I also
tried using other values for the ID (100 instead of 10000). The net result
was the same in all cases: two blocks appear on the heap and the wrong
handle is returned to me.
Is this a bug in the system software? Am I doing something wrong?
Any suggestions?
BTW. For those who think it is important, I am using THINK C 4.0 on an
8Mb Mac IIcx, system 6.0.4b15.
Before you start screaming about using 6.0.4b15, you should know that
the problems were already there when I still used 6.0.3. I have not tested
with other versions.
--dolf
Dolf Starreveld Phone: +31 20 592 5056/+31 20 592 5022, TELEX: 10262 HEF NL
EMAIL: dolf@fwi.uva.nl (dolf%fwi.uva.nl@hp4nl.nluug.nl)
SNAIL: Dept. of Math. and Computing Science, University of Amsterdam,
Kruislaan 409, NL-1098 SJ Amsterdam, The Netherlandshan@apple.COM (Byron Han, Project Scapegoat) (11/07/89)
In article <232@fwi.uva.nl> dolf@fwi.uva.nl (Dolf Starreveld) writes: > I have written a general purpose "CopyResource" routine. The routine itself > was extensively tested and seemed to work, but ..... > The copying of the 'dctb' resource was only recently added and things > started to go wrong. After lots of debugging I found out that after executing > the following statement: > inHandle = GetResource('dctb', 10000); > "inHandle" contains a valid handle, but not a handle to a resource. I entered > TMON to find out more and: > 1) Just before executing the statement the 'dctb' resource is not > loaded in the heap. > 2) Just after the statement, the 'dctb' resource is loaded, but > immediately following it, the heap contains another relocatable > block of exactly the same size of the resource (48 bytes). > 3) The GetResource call returns a handle to this last block instead > of to the actually loaded resource. > > <deleted> > > Is this a bug in the system software? Am I doing something wrong? > Any suggestions? Well, this is kind of a strange quirk in the operating system. Do this instead: SetResLoad(FALSE); hDCTB := GetResource('dctb', foobar); SetResLoad(TRUE); LoadResource(hDCTB); This should work in place of a simple GetResource. +-------------------------------------------------------------------------+ | Disclaimer: Apple has no connection with my postings. | +-------------------------------------------------------------------------+ Byron Han, CommToolbox Scapegoat "DeAnza 3 - R.I.P. - 10/17/89 5:04PM" Apple Computer, Inc. -------------------------------- 20525 Mariani Ave, MS 69L Internet: han@apple.COM Cupertino, CA 95014 UUCP:{sun,voder,nsc,decwrl}!apple!han -------------------------------- GENIE:BYRONHAN CompuServe:72167,1664 ATTnet: 408-974-6450 Applelink:HAN1 HAN1@applelink.apple.COM ---------------------------------------------------------------------------
alan@Apple.COM (Alan Mimms) (11/07/89)
In article <232@fwi.uva.nl> dolf@fwi.uva.nl (Dolf Starreveld) writes: > >I am currently working on a program that needs to copy a 'dctb' resource >from one file to another. Since the program must also copy other resources >I have written a general purpose "CopyResource" routine. The routine itself >was extensively tested and seemed to work, but ..... >The copying of the 'dctb' resource was only recently added and things >started to go wrong. After lots of debugging I found out that after executing >the following statement: > inHandle = GetResource('dctb', 10000); >"inHandle" contains a valid handle, but not a handle to a resource. I entered >TMON to find out more and: >1) Just before executing the statement the 'dctb' resource is not > loaded in the heap. >2) Just after the statement, the 'dctb' resource is loaded, but > immediately following it, the heap contains another relocatable > block of exactly the same size of the resource (48 bytes). >3) The GetResource call returns a handle to this last block instead > of to the actually loaded resource. > >Since my program continues to do a GetResInfo call on the handle returned, >I get into problems. The handle is not a handle to a resource and the >call sets ResError != 0. Furthermore, the &resID is set to -1 and the >&resType is set to four zero bytes. This is what finally causes another >program using the copied resource file to crash. > >I moved the GetResource call to very early in my program (just after ToolBox >initialization), just to be sure the heap was not already corrupt. I also >tried using other values for the ID (100 instead of 10000). The net result >was the same in all cases: two blocks appear on the heap and the wrong >handle is returned to me. > >Is this a bug in the system software? Am I doing something wrong? >Any suggestions? > >BTW. For those who think it is important, I am using THINK C 4.0 on an >8Mb Mac IIcx, system 6.0.4b15. >Before you start screaming about using 6.0.4b15, you should know that >the problems were already there when I still used 6.0.3. I have not tested >with other versions. > >--dolf > >Dolf Starreveld Phone: +31 20 592 5056/+31 20 592 5022, TELEX: 10262 HEF NL >EMAIL: dolf@fwi.uva.nl (dolf%fwi.uva.nl@hp4nl.nluug.nl) >SNAIL: Dept. of Math. and Computing Science, University of Amsterdam, > Kruislaan 409, NL-1098 SJ Amsterdam, The Netherlands SOME version of the system software introduced a "feature" in which 'dctb' and 'actb' resources are CLONED when they're retrieved by GetResource and related calls. A colleague of mine discovered this recently. You MAY be able to get around this with either Get1Resource calls or with the sequence SetResLoad (false); GetResource (...); SetResLoad (true); LoadResource (...); Anyone in System Software care to comment? Also, please note that you really SHOULD go ahead an upgrade to 6.0.4 final since there were a few bugs fixed in the last few versions of 6.0.4 after b15 (I THINK). Hope this helps. -- Alan Mimms My opinions are generally Communications Product Development Group pretty worthless, but Apple Computer they *are* my own... "The company has new jobs and Jobs has a new company" -- Harry Anderson
dolf@fwi.uva.nl (Dolf Starreveld) (11/07/89)
alan@Apple.COM (Alan Mimms) writes: [Lots deleted] > Also, please note that you really SHOULD go ahead an upgrade to 6.0.4 > final since there were a few bugs fixed in the last few versions of > 6.0.4 after b15 (I THINK). Thanks, both to you and Byron, for the info about dctb cloning. The reason I have not yet upgraded to 6.0.4. final is that Apple Holland is refusing to give it to the Dutch developers. They insist that you need this only if you have a Portable or a IIci. Anyway, tomorrow I am visiting a guy at Apple Holland and will try again. --dolf Dolf Starreveld Phone: +31 20 592 5056/+31 20 592 5022, TELEX: 10262 HEF NL EMAIL: dolf@fwi.uva.nl (dolf%fwi.uva.nl@hp4nl.nluug.nl) SNAIL: Dept. of Math. and Computing Science, University of Amsterdam, Kruislaan 409, NL-1098 SJ Amsterdam, The Netherlands
pepke@loligo (Eric Pepke) (11/07/89)
In article <5058@internal.Apple.COM> han@apple.COM (Byron Han, Project Scapegoat) writes: > >Well, this is kind of a strange quirk in the operating system. > >Do this instead: >SetResLoad(FALSE); >hDCTB := GetResource('dctb', foobar); >SetResLoad(TRUE); >LoadResource(hDCTB); Please give a complete definition of the quirk, including what types of resources it affects under what conditions. I trust that I will not have to change all my code that gets all kinds of resources to do this charade; however, nor do I want to have timebombs ticking in my code. Eric Pepke INTERNET: pepke@gw.scri.fsu.edu Supercomputer Computations Research Institute MFENET: pepke@fsu Florida State University SPAN: scri::pepke Tallahassee, FL 32306-4052 BITNET: pepke@fsu Disclaimer: My employers seldom even LISTEN to my opinions. Meta-disclaimer: Any society that needs disclaimers has too many lawyers.
keith@Apple.COM (Keith Rollin) (11/15/89)
In article <338@vsserv.scri.fsu.edu> pepke@loligo.UUCP (Eric Pepke) writes: >In article <5058@internal.Apple.COM> han@apple.COM (Byron Han, Project Scapegoat) writes: >> >>Well, this is kind of a strange quirk in the operating system. >> >>Do this instead: >>SetResLoad(FALSE); >>hDCTB := GetResource('dctb', foobar); >>SetResLoad(TRUE); >>LoadResource(hDCTB); > >Please give a complete definition of the quirk, including what types of >resources it affects under what conditions. I trust that I will not have >to change all my code that gets all kinds of resources to do this charade; >however, nor do I want to have timebombs ticking in my code. I haven't seen anyone respond to this yet, so I'll give what little information I have. I hadn't responded earlier, as I don't know the full details of the problem, and hate to speak knowledgably on things I don't know about. I think that this whole thing was an attempt to fix a bug in the dialog manager. It was leaving 'dctb' resources in memory after the dialog was disposed of. Apparently, the fix involved a patch to GetResource that would detach the 'dctb'. This is what leads us to the current problem. Needless to say, we in DTS were upset about this when we found out. There were many developers that were messed up by this patch. I think that some guys here are working on another fix that should fix the fix. -- ------------------------------------------------------------------------------ Keith Rollin --- Apple Computer, Inc. --- Developer Technical Support INTERNET: keith@apple.com UUCP: {decwrl, hoptoad, nsc, sun, amdahl}!apple!keith "Argue for your Apple, and sure enough, it's yours" - Keith Rollin, Contusions