Re: [orbitcpp-list] Another bug with #included idl file

[Sam C. <sa...@to...>]

Kuba <kp...@po...> wrote:
> Hi,
>=20
> I have found another problem, when you're #including one idl from another.

[ Example IDL snipped ]

> The reason for that is fact, that in IDL_Tree analyzed by compiler there =
are two subtrees for module A. When searching for SomeStruct function looku=
pLocal
> from language.cc looks into the first module A subtree only, so it doesn'=
t find SomeStruct declared in the second subtree.

I just tried this (with ORBit-C++ 0.30.1) using the example IDL you posted
(after correcting "int" to "in"), and it works fine for me. Can you try it
with 0.30.1 and tell me if you still see the problem? I don't want to apply
the patch unless it's actually fixing an existent problem.

> Here is a patch (sorry for some garbage inside, my editor cuts white spac=
es at the end of the lines - the only changed functions are lookupLocal () =
and getScope):

You can tell diff to ignore different kinds of whitespace with command line
arguments like -b, -B and -w. Check the man page for diff for details.
--=20
Sam Couter          |   Internet Engineer   |   http://www.topic.com.au/
sa...@to...    |   tSA Consulting      |
OpenPGP key ID:       DE89C75C,  available on key servers
OpenPGP fingerprint:  A46B 9BB5 3148 7BEA 1F05  5BD5 8530 03AE DE89 C75C