FROM : Bill Bumgarner
DATE : Wed Feb 06 23:21:37 2008
On Feb 6, 2008, at 2:12 PM, Jonathan Hess wrote:
> I'm not sure that this is your problem, but I would expect to see a
> call to [pluginBundle load] right after pluginBundle is set to
> [NSBundle bundleWithPath: path]. Having an instance of NSBundle
> doesn't imply that the code for the bundle has been loaded into your
> address space.
-principalClass should load the bundle and this check....
> if (pluginClass = [pluginBundle principalClass]) {
.... would barf it didn't.
I would suggest that OP put in some logging or debugging only code
that prints something indicative of which if() fell through on failure.
>> I'm quite running out of ideas what is going on. The more because
>> in addition the gcc debugger seems to have problems with this on
>> its own:
Some suggestions:
- Check the bundle's path; is it different in the failure case vs.
the success case?
- does the download process do some kind of a "all done, move into
place" operation?
- could there be a race condition where your loading code is
activated before the cleanup-after-download is done?
>> BFD: BFD 2.16.91 20050815 internal error, aborting at /SourceCache/
>> gdb/gdb-768/src/bfd/cache.c line 517 in bfd_cache_lookup_worker
>> BFD: Please report this bug.
>> The Debugger has exited with status 1.The Debugger has exited with
>> status 1.
Please file a bug on that...
b.bum
DATE : Wed Feb 06 23:21:37 2008
On Feb 6, 2008, at 2:12 PM, Jonathan Hess wrote:
> I'm not sure that this is your problem, but I would expect to see a
> call to [pluginBundle load] right after pluginBundle is set to
> [NSBundle bundleWithPath: path]. Having an instance of NSBundle
> doesn't imply that the code for the bundle has been loaded into your
> address space.
-principalClass should load the bundle and this check....
> if (pluginClass = [pluginBundle principalClass]) {
.... would barf it didn't.
I would suggest that OP put in some logging or debugging only code
that prints something indicative of which if() fell through on failure.
>> I'm quite running out of ideas what is going on. The more because
>> in addition the gcc debugger seems to have problems with this on
>> its own:
Some suggestions:
- Check the bundle's path; is it different in the failure case vs.
the success case?
- does the download process do some kind of a "all done, move into
place" operation?
- could there be a race condition where your loading code is
activated before the cleanup-after-download is done?
>> BFD: BFD 2.16.91 20050815 internal error, aborting at /SourceCache/
>> gdb/gdb-768/src/bfd/cache.c line 517 in bfd_cache_lookup_worker
>> BFD: Please report this bug.
>> The Debugger has exited with status 1.The Debugger has exited with
>> status 1.
Please file a bug on that...
b.bum
| Related mails | Author | Date |
|---|---|---|
| Alexander Griekspo… | Feb 6, 18:36 | |
| Jonathan Hess | Feb 6, 23:12 | |
| Alexander Griekspo… | Feb 6, 23:20 | |
| Bill Bumgarner | Feb 6, 23:21 | |
| Bill Bumgarner | Feb 6, 23:26 | |
| Alexander Griekspo… | Feb 6, 23:32 |






Cocoa mail archive

