orbitcpp-list

Re: [orbitcpp-list] Bonobo.idl

[ERDI G. <ca...@ca...>]

On Mon, 25 Jun 2001, Richard Andrews wrote:

> Don't worry about it. It isn't a problem in this particular case.

> > /usr/share/idl/Bonobo_Unknown.idl:19: unknown identifier
> > ::Bonobo::Unknown

the real problem begins here.

-- 
   .--= ULLA! =---------------------.   `We are not here to give users what
   \     http://cactus.rulez.org     \   they want'  -- RMS, at GUADEC 2001
    `---= ca...@ca... =---'
The time to make up your mind about people is...never.

[orbitcpp-list] C wrapper implementations?

[ERDI G. <ca...@ca...>]

Hi,

Could C CORBA_Objects be wrapped by the IDL compiler into C++ objects?
So when compiling an interface, it would create an implementation that is
initialized by a C object implementing the same interface. I.e. a C++->C
in-proc proxy.

This would be very useful for creating language bindings for libraries
where C CORBA objects are returned by the entry poitns (e.g. libbonob).

-- 
   .--= ULLA! =---------------------.   `We are not here to give users what
   \     http://cactus.rulez.org     \   they want'  -- RMS, at GUADEC 2001
    `---= ca...@ca... =---'
Exactly what time of morning did technology pass me by?

Re: [orbitcpp-list] problems compiling idl with ORBit-cpp, again

[Richard A. <ric...@ix...>]

I think you have two different versions of libstdc++ now.

The C13 stuff is just the mangled C++ function entry point names.

Does the compiler work on other programs with shared libraries?




On Wed, 20 Jun 2001 19:02:16 Margarita de Cabo wrote:
> Hi again,
> 
> I have done my libsdtc++ a sahred library, I have recompiled ORBit-cpp
> with no problems, I hava installed a new version of gcc (2.95.3), and I
> am still
> having problems with no found symbols in liborbit-idl-c++-backend.so.
> In my first atempts the no found reference was C13__ ..., or something
> like that, now the error is:
> 
> orbit-idl -lc++
> --backenddir=/usr/local/packages/ORBit-cpp/sparc-sun-solaris/lib/orbit-idl
> prueba.idl
> 
> ** WARNING **: Module load failed: ld.so.1: orbit-idl: fatal: relocation
> error: file
> /usr/local/packages/ORBit-cpp/sparc-sun-solaris/lib/orbit-idl/liborbit-idl-c++-backend.so:
> symbol __pure_virtual: referenced symbol not found
> 
> ** CRITICAL **: file orbit-idl-driver.c: line 50 (orbit_idl_to_backend):
> assertion `binfo && binfo->op_output' failed.
> 
> ** WARNING **: prueba.idl compilation failed
> 
> I need help, I am desperate.
> 
> Thanks for your time.
> 
> Marga
> 
> 
> begin:vcard 
> n:de Cabo;Margarita 
> tel;fax:34-983-423667
> tel;work:34-983-423660 ext:25536
> x-mozilla-html:TRUE
> org:Universidad de Valladolid;TSCIT
> version:2.1
> email;internet:ma...@te...
> adr;quoted-printable:;;Edificio de las Nuevas
> Tecnolog=EDas	=0D=0ACamino del Cementerio
> s/n		;Valladolid;Valladolid;47011;España
> x-mozilla-cpt:;0
> fn:Margarita de Cabo
> end:vcard
> 

[orbitcpp-list] problems compiling idl with ORBit-cpp, again

[Margarita de C. <ma...@gb...>]

Hi again,

I have done my libsdtc++ a sahred library, I have recompiled ORBit-cpp
with no problems, I hava installed a new version of gcc (2.95.3), and I
am still
having problems with no found symbols in liborbit-idl-c++-backend.so.
In my first atempts the no found reference was C13__ ..., or something
like that, now the error is:

orbit-idl -lc++
--backenddir=/usr/local/packages/ORBit-cpp/sparc-sun-solaris/lib/orbit-idl
prueba.idl

** WARNING **: Module load failed: ld.so.1: orbit-idl: fatal: relocation
error: file
/usr/local/packages/ORBit-cpp/sparc-sun-solaris/lib/orbit-idl/liborbit-idl-c++-backend.so:
symbol __pure_virtual: referenced symbol not found

** CRITICAL **: file orbit-idl-driver.c: line 50 (orbit_idl_to_backend):
assertion `binfo && binfo->op_output' failed.

** WARNING **: prueba.idl compilation failed

I need help, I am desperate.

Thanks for your time.

Marga

Re: [orbitcpp-list] Re: Compiling IDL with ORBitcpp

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

Margarita de Cabo <ma...@gb...> wrote:
> Hi,
> I did what you say, but the error is still there.  More suggestions?  I h=
ave no
> idea of what to try?

If you're still getting messages about not having a shared libstdc++.so,
then you need to solve that problem first. You'll have to check the
documentation for that package to make sure you're building and installing
it correctly.
--=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

Re: [orbitcpp-list] Compiling IDL with ORBitcpp

[Richard A. <ric...@ix...>]

> Well, I solved this problem on my solaris system in a different way. I
> used STLPort (http://www.stlport.org) as the shared library instead of
> libstdc++. Which means that it is possible to use orbitcpp without
> libstdc++.so ;-)

I think I used dinkumware a while back with success. The problem was that I
couldn't link any GUI stuff because it caused GTK to crash on init.

From what I've seen the actual STL content of orbitcpp is pretty small -
the main requirement is a compiler that handles templates and implicit
construction OK.

Re: [orbitcpp-list] Compiling IDL with ORBitcpp

[Laszlo M. <las...@et...>]

On Mon, Jun 11, 2001 at 09:24:58PM +1000, Richard Andrews wrote:
> > Do you have a *shared* libstdc++?
> This seems to be a recurring problem. I think there must be a lot of
> libstdc++ builds out there that have been built with --disable-shared. I
> know the first time I tried to build libstdc++-2.90.8 I had to disable
> shared to get it to build on RH6.2 binutils tools.
> Is there a configure check to ensure libstdc++.so exists?

Well, I solved this problem on my solaris system in a different way. I
used STLPort (http://www.stlport.org) as the shared library instead of
libstdc++. Which means that it is possible to use orbitcpp without
libstdc++.so ;-)

Laszlo

Re: [orbitcpp-list] Compiling IDL with ORBitcpp

[Richard A. <ric...@ix...>]

On Mon, 11 Jun 2001 18:52:39 Laszlo Molnar wrote:

> Do you have a *shared* libstdc++?


This seems to be a recurring problem. I think there must be a lot of
libstdc++ builds out there that have been built with --disable-shared. I
know the first time I tried to build libstdc++-2.90.8 I had to disable
shared to get it to build on RH6.2 binutils tools.

Is there a configure check to ensure libstdc++.so exists?

Re: [orbitcpp-list] Re: Compiling IDL with ORBitcpp

[Laszlo M. <las...@et...>]

On Fri, Jun 08, 2001 at 11:10:53AM +0200, Margarita de Cabo wrote:
> Hi,
> I did what you say, but the error is still there.  More suggestions?  I have no
> idea of what to try?

Do you have a *shared* libstdc++?

Laszlo

Re: [orbitcpp-list] Re: Compiling IDL with ORBitcpp

[Margarita de C. <ma...@gb...>]

Hi,
I did what you say, but the error is still there.  More suggestions?  I have no
idea of what to try?

Thanks

Sam Couter wrote:

> Margarita de Cabo <ma...@gb...> wrote:
> >
> > The only thing is the next warning:
> >
> > *** Warning: This library needs some functionality provided by -lstdc++.
> >
> > *** I have the capability to make that library automatically link in
> > when
> > *** you link to this library.  But I can only do this if you have a
> > *** shared version of the library, which you do not appear to have.
>
> This is probably the cause. This library is needed, and the ORBit-C++ IDL
> library won't link correctly without it.
>
> If you have libstdc++ installed on your system somewhere, add that directory
> to your LD_LIBRARY_PATH, re-link the ORBit-C++ IDL library (cd compiler;
> make clean; make; make install) and try again.
>
> If you don't have it installed... I think it's part of the GNU C++ compiler
> distribution.
> --
> 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
>
>   ------------------------------------------------------------------------
>    Part 1.2Type: application/pgp-signature

Re: [orbitcpp-list] Re: Compiling IDL with ORBitcpp

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

Margarita de Cabo <ma...@gb...> wrote:
>
> The only thing is the next warning:
>=20
> *** Warning: This library needs some functionality provided by -lstdc++.
>=20
> *** I have the capability to make that library automatically link in
> when
> *** you link to this library.  But I can only do this if you have a
> *** shared version of the library, which you do not appear to have.

This is probably the cause. This library is needed, and the ORBit-C++ IDL
library won't link correctly without it.

If you have libstdc++ installed on your system somewhere, add that directory
to your LD_LIBRARY_PATH, re-link the ORBit-C++ IDL library (cd compiler;
make clean; make; make install) and try again.

If you don't have it installed... I think it's part of the GNU C++ compiler
distribution.
--=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

[orbitcpp-list] Re: Compiling IDL with ORBitcpp

[Margarita de C. <ma...@gb...>]

    Hi, again

My problem is that I can compile an idl file, the error while compiling
:

orbit-idl -l c++
--backenddir=/usr/.../ORBit-cpp/sparc-sun-solaris2.7/lib/orbit-idl
file.idl

** WARNING **: Module load failed: ld.so.1: orbit-idl: fatal: relocation
error: file /usr/
.../ORBit-cpp/sparc-sun-solaris2.7/lib/orbit-idl/liborbit-idl-c++-backend.so:
symbol what__C13runtime_error: referenced symbol not found

** CRITICAL **: file orbit-idl-driver.c: line 50 (orbit_idl_to_backend):
assertion `binfo && binfo->op_output' failed.

** WARNING **: file.idl compilation failed

Here is the information you request me in order to be able of having me
the better response for my problem:


> What versions of ORBit and ORBit-C++ are you using?
> What platform are you running on?
>
SunOS merlu 5.7 Generic_106541-06 sun4u sparc SUNW,Ultra-250
ORBit-0.5.7, orbit-cpp-0.30

> What configure options did you use when building them?

./configure --prefix=/usr/local/packages/ORBit-cpp
--exec-prefix=/usr/local/packages/ORBit-cpp/sparc-sun-solaris2.7

> Did you see any error or warning messages when compiling them?
The only thing is the next warning:

*** Warning: This library needs some functionality provided by -lstdc++.

*** I have the capability to make that library automatically link in
when
*** you link to this library.  But I can only do this if you have a
*** shared version of the library, which you do not appear to have.
*** The inter-library dependencies that have been dropped here will be
*** automatically added whenever a program is linked with this library
*** or is declared to -dlopen it.


> What compiler and linker are you using, and which versions?

c++: 2.95.2 (GNU CC)
ld: Software Generation Utilities - Solaris/ELF (3.0) (Solaris 2.7)

Thanks in advance

Marga

[orbitcpp-list] Re: orbitcpp-list digest, Vol 1 #206 - 1 msg

[Kuba <kp...@po...>]

> Here's where I probably show my lack of understanding:
> Isn't this what the #pragma inhibit directives are for?
> 
> Something like:
> 
> #pragma inhibit push
> #include "file.idl"
> #pragma inhibit pop
> 
> IDL continues...
> --=20

That doesn't quite solve the problem, but there were discussion between Bill Hanneman and Michael Meeks (orbit-list, 29-May-2001, subject: unbuildable code generated for inheritance?) and they presented solution (very ugly, though).
Thank you anyway.


-- 

Kamera wideo Sony, 5000 zl
wygraj w konkursie eMarketu i Onet.pl [ http://konkursy.onet.pl/emarket ]

Re: [orbitcpp-list] Compiling idl with ORBit-cpp

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

Margarita de Cabo <ma...@gb...> wrote:
> Hi,
> I am  trying to compile an idl file with ORBit-cpp and I have the
> following error
> orbit-idl -l c++ --backenddir=3D/usr/.../lib/orbit-idl file.idl
>=20
> ** WARNING **: Module load failed: ld.so.1: orbit-idl: fatal: relocation
>=20
> error: file /usr/l.../lib/orbit-idl/liborbit-idl-c++-backend.so: symbol
> what__C13runtime_error: referenced symbol not found

What versions of ORBit and ORBit-C++ are you using?
What configure options did you use when building them?
Did you see any error or warning messages when compiling them?
What platform are you running on?
What compiler and linker are you using, and which versions?

If you can answer these questions, we have a better chance of helping you.
--=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

[orbitcpp-list] Compiling idl with ORBit-cpp

[Margarita de C. <ma...@gb...>]

Hi,
I am  trying to compile an idl file with ORBit-cpp and I have the
following error
orbit-idl -l c++ --backenddir=/usr/.../lib/orbit-idl file.idl

** WARNING **: Module load failed: ld.so.1: orbit-idl: fatal: relocation

error: file /usr/l.../lib/orbit-idl/liborbit-idl-c++-backend.so: symbol
what__C13runtime_error: referenced symbol not found

** CRITICAL **: file orbit-idl-driver.c: line 50 (orbit_idl_to_backend):

assertion `binfo && binfo->op_output' failed.

** WARNING **: file.idl compilation failed



Can anybody explain me what I'm doing wrong

Marga

[orbitcpp-list] Re: orbitcpp-list digest, Vol 1 #205

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

Kuba <kp...@po...> wrote:
> Yes, you're right. BTW, this is only a part of problem I have with ORBit(=
cpp) in my project. If you look into files generated from my example, you'l=
l notice that in files file2*.cc and file2*.h there are declarations of int=
erfaces from file1.idl !

Here's where I probably show my lack of understanding:
Isn't this what the #pragma inhibit directives are for?

Something like:

#pragma inhibit push
#include "file.idl"
#pragma inhibit pop

IDL continues...
--=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

[orbitcpp-list] Re: orbitcpp-list digest, Vol 1 #203 - 1 msg

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

Kuba <kp...@po...> wrote:
> OK, you are right. The example wasn't precise. Try this (doesn't matter w=
hether using orbit-cpp-0.30 or 0.30.1):

 [ example IDL snipped ]

> Put both files into the same directory and compile:
> orbit-idl -l c++ -I. file2.idl

Okay, I've confirmed this and committed your patch.

One thing I noticed: You don't need to have the two module declarations in
seperate files to trigger the bug. Just one file will do:

module A {
	interface Empty {
	};
};

module A {
	struct SomeStruct {
		short foo;
	};

	interface NotEmpty : Empty {
		void SomeFunction (in SomeStruct foo);
	};
};
--=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

[orbitcpp-list] Re: orbitcpp-list digest, Vol 1 #203 - 1 msg

[Kuba <kp...@po...>]

OK, you are right. The example wasn't precise. Try this (doesn't matter whether using orbit-cpp-0.30 or 0.30.1):

file1.idl
module A {
  interface Empty {
  };
};

file2.idl
#include <file1.idl>
module A {
  struct SomeStruct {
    string SomeString;
  };

  interface NotEmpty : Empty {
    void SomeFunction (in SomeStruct inParam);
  };
};

Put both files into the same directory and compile:
orbit-idl -l c++ -I. file2.idl

Here is the patch (thanks for diff options hint!):
------------------------------------------------------------
diff -u -b -B -w compiler/language.cc compiler.new/language.cc
--- compiler/language.cc	Fri Apr  6 11:41:42 2001
+++ compiler.new/language.cc	Sun Apr 22 00:05:24 2001
@@ -275,15 +275,20 @@
 	string::size_type first = 0;
 
 	string::size_type nextscopeq = id.find("::",first);
-	while (nextscopeq != string::npos) {
-		IDLScope *nextscope = scope->getScope(id.substr(first,nextscopeq-first));
-		if (!nextscope) return NULL;
-		scope = nextscope;
-		first = nextscopeq + 2;
-		nextscopeq = id.find("::",first);
+	if (nextscopeq == string::npos) {
+		return scope->getItem(id.substr(first));
+	}
+	else {
+		int spos = 0;
+		IDLScope *nextscope = NULL;
+		while ((nextscope = scope->getScope(id.substr(first,nextscopeq-first),spos)) != NULL) {
+			spos++;
+			IDLElement* element = nextscope->lookupLocal(id.substr(nextscopeq+2,id.length()-nextscopeq-2));
+			if (element != NULL) return element;
+		}
 	}
 
-	return scope->getItem(id.substr(first));
+	return NULL;
 }
 
 
@@ -318,13 +324,22 @@
 
 
 IDLScope *
-IDLScope::getScope(string const &id) const {
+IDLScope::getScope (string const &id, int &spos) const {
 	ScopeList::const_iterator first = m_scopes.begin(),last = m_scopes.end();
+	string IDLIdentifier;
 
+	int pos_counter = 0;
 	while (first != last) {
-		if ((*first)->getIDLIdentifier() == id) return *first;
+		IDLIdentifier = (*first)->getIDLIdentifier();
+		if ((*first)->getIDLIdentifier() == id && pos_counter >= spos) {
+			spos = pos_counter;
+			return *first;
+		}
+
 		first++;
+		pos_counter++;
 	}
+
 	return NULL;
 }
 
diff -u -b -B -w compiler/language.hh compiler.new/language.hh
--- compiler/language.hh	Mon Jan 22 08:44:05 2001
+++ compiler.new/language.hh	Sun Apr 22 00:04:43 2001
@@ -129,7 +129,7 @@
 
 	IDLElement *getItem(IDL_tree node) const;
 	IDLElement *getItem(string const &id) const;
-	IDLScope *getScope(string const &id) const;
+	IDLScope *getScope (string const &id, int &spos) const;
 	IDLScope const *getRootScope() const {
 		if (getParentScope()) return Super::getRootScope();
 		else return this;
------------------------------------------------------------


-- 

Kamera wideo Sony, 5000 zl
wygraj w konkursie eMarketu i Onet.pl [ http://konkursy.onet.pl/emarket ]

Re: [orbitcpp-list] error compiling orbitcpp

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

Antony Suter <an...@mi...> wrote:
>=20
> On further examination I found passing -j2 to make was causing the
> problem. I guess multiple processes (on my SMP system) somehow
> resulted in multiple copies of the data making it into that file.

That means the dependencies aren't specified correctly. Using the -j option
means some dependencies can be built in parallel, but if the Makefiles
aren't correct then sometimes stuff breaks.
--=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

Re: [orbitcpp-list] error compiling orbitcpp

[Antony S. <an...@mi...>]

Sam Couter wrote:
> 
> Antony Suter <an...@mi...> wrote:
> >
> > Why am i getting these types of bugs when compiling?
[snip]
> Try doing a make clean (and make sure the CosNaming.cc and CosNaming.hh
> files get deleted), and make again. You'll only need to make clean from that
> directory to get it to work, but if that directory is messed up, maybe
> others are as well.

On further examination I found passing -j2 to make was causing the
problem. I guess multiple processes (on my SMP system) somehow
resulted in multiple copies of the data making it into that file.

Thanks.

--
- Antony Suter  (an...@mi...)  "Exner"  openpgp:7916EE67
- "...to condense fact from the vapor of nuance."

Re: [orbitcpp-list] error compiling orbitcpp

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

Antony Suter <an...@mi...> wrote:
>=20
> Why am i getting these types of bugs when compiling?

Because your build is screwy for some reason. See below.

> I have gcc 2.96-0.52 from Mandrake Cooker.
> Ignore the locale error unless it is relevant to the c++ error.

It's not, that's just perl.

> CosNaming.cc:352: redefinition of `void=20
> _orbitcpp::stub::CosNaming::NamingContext::bind (const Name &,=20
> CORBA::Object *)'
> CosNaming.cc:66: `void _orbitcpp::stub::CosNaming::NamingContext::bind=20
> (const Name &, CORBA::Object *)' previously defined here
> CosNaming.cc:384: redefinition of `void=20
> _orbitcpp::stub::CosNaming::NamingContext::rebind (const Name &,=20
> CORBA::Object *)'
> CosNaming.cc:98: `void=20
> _orbitcpp::stub::CosNaming::NamingContext::rebind (const Name &,=20
> CORBA::Object *)' previously defined here
> CosNaming.cc:411: redefinition of `void=20
> _orbitcpp::stub::CosNaming::NamingContext::bind_context (const Name &,=20
> _orbitcpp::stub::CosNaming::NamingContext *)'

These errors are ocurring on lines that don't even exist in the file (at
least, not in the one that I just generated). I'd say that somehow you've
ended up with the file concatenated onto itself, or something equally
screwy and confusing.

Try doing a make clean (and make sure the CosNaming.cc and CosNaming.hh
files get deleted), and make again. You'll only need to make clean from that
directory to get it to work, but if that directory is messed up, maybe
others are as well.
--=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

[orbitcpp-list] error compiling orbitcpp

[Antony S. <an...@mi...>]

Why am i getting these types of bugs when compiling?

I have gcc 2.96-0.52 from Mandrake Cooker.
Ignore the locale error unless it is relevant to the c++ error.

---

[antony@gamut name]$ make
/bin/sh ../../libtool --mode=compile c++ -DHAVE_CONFIG_H -I. -I.
-I../../orb -I../.. -I../.. -I/usr/include/glib-1.2
-I/usr/lib/glib/include -I/usr/include     -g -O0 -Wall -c CosNaming.cc
rm -f .libs/CosNaming.lo
c++ -DHAVE_CONFIG_H -I. -I. -I../../orb -I../.. -I../..
-I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/include -g -O0
-Wall -Wp,-MD,.deps/CosNaming.pp -c CosNaming.cc   -fPIC -DPIC
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
	LANGUAGE = "en_US:en",
	LC_ALL = (unset),
	LC_MESSAGES = "en_US",
	LC_TIME = "en_US",
	LC_NUMERIC = "en_US",
	LC_CTYPE = "en_US",
	LC_MONETARY = "en_US",
	LC_COLLATE = "en_US",
	LANG = "C"
    are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
CosNaming.cc:352: redefinition of `void 
_orbitcpp::stub::CosNaming::NamingContext::bind (const Name &, 
CORBA::Object *)'
CosNaming.cc:66: `void _orbitcpp::stub::CosNaming::NamingContext::bind 
(const Name &, CORBA::Object *)' previously defined here
CosNaming.cc:384: redefinition of `void 
_orbitcpp::stub::CosNaming::NamingContext::rebind (const Name &, 
CORBA::Object *)'
CosNaming.cc:98: `void 
_orbitcpp::stub::CosNaming::NamingContext::rebind (const Name &, 
CORBA::Object *)' previously defined here
CosNaming.cc:411: redefinition of `void 
_orbitcpp::stub::CosNaming::NamingContext::bind_context (const Name &, 
_orbitcpp::stub::CosNaming::NamingContext *)'


--
- Antony Suter  (an...@mi...)  "Exner"  openpgp:7916EE67
- "...to condense fact from the vapor of nuance."

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

[orbitcpp-list] ORBit-C++ 0.30.1 released

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

The ORBit-C++ 0.30.1 bugfix release is out.

Release Notes:
23 May 2001:
	* Release version 0.30.1

	Bugfix release:
	- Fix int/size_t mixup that broke 64 bit systems
	- Squished some memory leaks
----

I'm not sure what else has been changed since the 0.30 release, as I haven't
gone trawling through the CVS logs yet.
--=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

Re: [orbitcpp-list] 2 questions

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

Antony Suter <an...@mi...> wrote:
>=20
> 1) Is there a FAQ for this mailing list - or for orbit-cpp?

Not that I know of. There are mailing list archives available.

If you have a question, ask it.
If you're volunteering to maintain a FAQ, let us know. :)

> 2) Are there RPMs of orbit-cpp available?

The distribution comes with a SPEC file, which should be enough for you to
build an RPM from the source. I don't know how up to date it is though. If
it needs modification, post the patches here.
--=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