[omniORB] Fixes for GCC warnings

[Tamas Kerecsen]

Duncan,

If you turn on the -Wextra GCC switch, it produces a bunch of warnings for a
couple of omniORB headers, and for certain types of omniidl-generated code.
GCC complains about the lack of explicit initialization for parent classes
in copy constructors. I think I've found all the instances, and fixed them
all.
I would like to ask you to apply the attached patches to the 4.1 line, so I
don't have to maintain a patched version of omniORB for myself... (Note that
this changes the behavior of the copy constructor slightly, because the
parents will be initialized by copy constructor instead of default
constructor. If you intentionally want to not use the copy constructor for
some reason, I'd be more than happy to put together another patch, one with
explicit calls to the default constructors. GCC, as well as myself, doesn't
care which parent constructor is used, just so long as it is explicitly
stated :))

Thanks in advance,
Tamas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.omniorb-support.com/pipermail/omniorb-list/attachments/20061010/a476b302/attachment.htm

[Tamas Kerecsen]

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: warnings_patch1.patch
Type: application/octet-stream
Size: 3037 bytes
Desc: not available
Url : http://www.omniorb-support.com/pipermail/omniorb-list/attachments/20061010/f5389b8b/warnings_patch1.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: warnings_patch2.patch
Type: application/octet-stream
Size: 1324 bytes
Desc: not available
Url : http://www.omniorb-support.com/pipermail/omniorb-list/attachments/20061010/f5389b8b/warnings_patch2.obj

[Duncan Grisby]

Post by Tamas Kerecsen
If you turn on the -Wextra GCC switch, it produces a bunch of warnings for a
couple of omniORB headers, and for certain types of omniidl-generated code.
GCC complains about the lack of explicit initialization for parent classes
in copy constructors. I think I've found all the instances, and fixed them
all.

Thanks for the patches. I've applied them to the 4.1 tree. I'd made a
similar set of changes in the past, but then backed them out because it
failed to compile on Visual C++ 6. Strangely, the version with your
patches works fine on VC++ 6, so I'm not sure what it was unhappy about
before.

Cheers,

Duncan.

[Tamas Kerecsen]

Duncan,

Thanks for applying the patches. Having them in the official release makes
things much easier for me...
As far as I could dig up from the version history, your earlier changes were
targeted at the default constructors, not the copy constructors. In
addition, my patch calls the parameterized parent copy constructors, not the
default ones (I don't know if it matters to VC++ or not, but I'm guessing it
might).

I'm not sure if you hear it enough or not, but thanks also for all the work
you put into omniORB :)

All the best,
Tamas


Thanks for the patches. I've applied them to the 4.1 tree. I'd made a

Post by Duncan Grisby
similar set of changes in the past, but then backed them out because it
failed to compile on Visual C++ 6. Strangely, the version with your
patches works fine on VC++ 6, so I'm not sure what it was unhappy about
before.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.omniorb-support.com/pipermail/omniorb-list/attachments/20061101/31a9065d/attachment.htm