I need help to use a C++ library in QNX. The library I have is NURBS++. (A
Nurbs library which ic compatible with linux,unix and even windows! and
available via http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/). gcc
2.95 in linux works fine with with the library but I couldn’t manage to
compile it in QNX.
I need help to use a C++ library in QNX. The library I have is NURBS++. (A
Nurbs library which ic compatible with linux,unix and even windows! and
available via > http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ). gcc
2.95 in linux works fine with with the library but I couldn’t manage to
compile it in QNX.
I need help to use a C++ library in QNX. The library I have is NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even windows! and
available via > http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t manage to
compile it in QNX.
I need help to use a C++ library in QNX. The library I have is NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even windows! and
available via > http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t manage to
compile it in QNX.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even windows!
and
available via > http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t manage
to
compile it in QNX.
I don’t really know Dean, I was just trying to give a hand to Mr “please
help”
From the documentation : "Non-Uniform Rational B-Splines (NURBS) curves and
surface are parametric functions which can represent any type of curves or
surfaces. This C++ library hides the basic mathematics of NURBS. This allows
the user to focus on the more challenging parts of their projects.
Mario Charest wrote:
I gave a try at compiling it under QNX6. It seems the following
statement
cause
a problem.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even windows!
and
available via http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t manage
to
compile it in QNX.
First of all sorry for cross posting and not being specific enough. I just
wanted to get an answer as soon as possible.
I appriciate any help.
Let me redefine the problem. I have QNX 6.0 and need to compile the
NURBS++
6.0? Ok then that’s different because the C and C++ library/header are
different
then with 6.1. I was using 6.1. I don’t have access to a 6.0 machine at
this
time to help you out sorry.
library so that later I can link it to my program. Being a GNU software
and
presumably platform independent, this library always needs some work to
compile.
By the way, for those who are interested in, the NURBS++ library helps you
to mathematically create your own surfaces (or curves), so that you can
create your own graphics. You can save results as VRML97 or .ps and even
view them with some GNU viewers. Many commertial complex CAD/CAM software
use a NURBS library, and this one is a free and a very powerful one.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even
windows!
and
available via http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t
manage
to
compile it in QNX.
First of all sorry for cross posting and not being specific enough. I just
wanted to get an answer as soon as possible.
I appriciate any help.
Let me redefine the problem. I have QNX 6.0 and need to compile the NURBS++
library so that later I can link it to my program. Being a GNU software and
presumably platform independent, this library always needs some work to
compile. http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/archive/threads.html#00
270 is an unactive newsgroup which has may tips/tricks to compile it under
different platforms. I manage to compile it under linux and windows myself,
but being a newbee not under QNX. I use gcc 2.95 compiler with QNX6.0.
By the way, for those who are interested in, the NURBS++ library helps you
to mathematically create your own surfaces (or curves), so that you can
create your own graphics. You can save results as VRML97 or .ps and even
view them with some GNU viewers. Many commertial complex CAD/CAM software
use a NURBS library, and this one is a free and a very powerful one.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even windows!
and
available via http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t manage
to
compile it in QNX.
Hi,
Maybe you can still explain me how you can run the configure without
problems. I guess you try to install the library as it is defined in http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/download.shtml. I am
not able to define a legal host type and default (guess) option doen’t work.
By the way I just learned that my QNX is Nuetrino 6.0. (if it makes any
difference.)
“please help” <> whatever160@hotmail.com> > wrote in message
news:9kul77$k6u$> 1@inn.qnx.com> …
First of all sorry for cross posting and not being specific enough. I
just
wanted to get an answer as soon as possible.
I appriciate any help.
Let me redefine the problem. I have QNX 6.0 and need to compile the
NURBS++
6.0? Ok then that’s different because the C and C++ library/header are
different
then with 6.1. I was using 6.1. I don’t have access to a 6.0 machine at
this
time to help you out sorry.
library so that later I can link it to my program. Being a GNU software
and
presumably platform independent, this library always needs some work to
compile.
By the way, for those who are interested in, the NURBS++ library helps
you
to mathematically create your own surfaces (or curves), so that you can
create your own graphics. You can save results as VRML97 or .ps and even
view them with some GNU viewers. Many commertial complex CAD/CAM
software
use a NURBS library, and this one is a free and a very powerful one.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even
windows!
and
available via http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t
manage
to
compile it in QNX.
Hi,
Maybe you can still explain me how you can run the configure without
problems. I guess you try to install the library as it is defined in http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/download.shtml> . I am
not able to define a legal host type and default (guess) option doen’t
work.
By the way I just learned that my QNX is Nuetrino 6.0. (if it makes any
difference.)
“please help” <> whatever160@hotmail.com> > wrote in message
news:9kul77$k6u$> 1@inn.qnx.com> …
First of all sorry for cross posting and not being specific enough. I
just
wanted to get an answer as soon as possible.
I appriciate any help.
Let me redefine the problem. I have QNX 6.0 and need to compile the
NURBS++
6.0? Ok then that’s different because the C and C++ library/header are
different
then with 6.1. I was using 6.1. I don’t have access to a 6.0 machine
at
this
time to help you out sorry.
library so that later I can link it to my program. Being a GNU
software
and
presumably platform independent, this library always needs some work
to
compile.
By the way, for those who are interested in, the NURBS++ library helps
you
to mathematically create your own surfaces (or curves), so that you
can
create your own graphics. You can save results as VRML97 or .ps and
even
view them with some GNU viewers. Many commertial complex CAD/CAM
software
use a NURBS library, and this one is a free and a very powerful one.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even
windows!
and
available via http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t
manage
to
compile it in QNX.
Since ostrstream is a class, when declared as such the error disappear
typedef class ostrstream ErrorStream;
I messed up, forget what I said ;-( The problem seems related to
namespace, and this I’m not familiar with. Sorry
The above won’t compile with the new ISO-compliant Dinkumware libraries which
make proper use of the std namespace.
You need either (preferably):
#include
typedef std::ostrstream ErrorStream;
or #include
using namespace std; // you may need this in every header!
typedef ostrstream ErrorStream;
Note that strstream is deprecated, so you should instead use sstream in new
code (although there are things strstream can do that sstream can’t, so I
don’t think it should be deprecated).
You may be able to configure gcc to use the old libraries rather than
Dinkumware (check QDN?), and this may be the easiest solution for now.
Tom
I don’t know why it works under other OS. I have look for other type
of compile error either.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even windows!
and
available via > http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t manage
to
compile it in QNX.
Thanks for helping me out. I made big progress and rigth now I have the
following compile error. Any more suggestions?
…/configure works fine. Then while making:
make
Making all in include
make[1]: Entering directory /root/nurbs/nurbs++-3.0.10/include' make[1]: Leaving directory /root/nurbs/nurbs+±3.0.10/include’
Making all in matrix
make[1]: Entering directory /root/nurbs/nurbs++-3.0.10/matrix' /bin/sh ../libtool --mode=compile g++ -DHAVE_CONFIG_H -I. -I. -I../include -I../ include -g -O2 -fno-implicit-templates -c cvector.cc rm -f .libs/cvector.lo g++ -DHAVE_CONFIG_H -I. -I. -I../include -I../include -g -O2 -fno-implicit-t empl ates -Wp,-MD,.deps/cvector.pp -c cvector.cc -fPIC -DPIC -o .libs/cvector.lo In file included from cvector.cc:27: cvector.hh:50: macro index’ used with just one arg
cvector.hh:51: macro index' used with just one arg cvector.hh:52: macro index’ used with just one arg
cvector.hh:53: macro index' used with just one arg cvector.hh:54: macro index’ used with just one arg
make[1]: *** [cvector.lo] Error 1
make[1]: Leaving directory `/root/nurbs/nurbs+±3.0.10/matrix’
make: *** [all-recursive] Error 1
or #include <strstream
using namespace std; // you may need this in every header!
typedef ostrstream ErrorStream;
Note that strstream is deprecated, so you should instead use sstream in
new
code (although there are things strstream can do that sstream can’t, so I
don’t think it should be deprecated).
You may be able to configure gcc to use the old libraries rather than
Dinkumware (check QDN?), and this may be the easiest solution for now.
Tom
I don’t know why it works under other OS. I have look for other type
of compile error either.
I need help to use a C++ library in QNX. The library I have is
NURBS++.
(A
Nurbs library which ic compatible with linux,unix and even
windows!
and
available via http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/> ).
gcc
2.95 in linux works fine with with the library but I couldn’t
manage
to
compile it in QNX.
I believe the problem occurs during linking with gcc 2.95.2. Here is what I
do:
tar -zxvf nurbs+±3.0.10.tar.gz
cd nurbs+±3.0.10
mkdir x86-pc-nto-qnx
(I add new config.sub and config.guess from ftp.gnu.org pub/gnu/config)
cd x86-pc-nto-qnx
…/configure
Everthing seems to work fine till here. Then
gcc --version
2.95.2
make
Making all in include
make[1]: Entering directory /root/NV/nurbs++-3.0.10/x86-pc-nto-qnx/include' make[1]: Leaving directory /root/NV/nurbs+±3.0.10/x86-pc-nto-qnx/include’
Making all in matrix
make[1]: Entering directory /root/NV/nurbs++-3.0.10/x86-pc-nto-qnx/matrix' /bin/sh ../libtool --mode=compile g++ -DHAVE_CONFIG_H -I. -I../../matrix -I../in clude -I../include -g -O2 -fno-implicit-templates -c .../../matrix/cvector.cc g++ -DHAVE_CONFIG_H -I. -I../../matrix -I../include -I../include -g -O2 -fno -imp licit-templates -c ../../matrix/cvector.cc -o cvector.o In file included from ../../matrix/cvector.cc:27: .../../matrix/cvector.hh:50: macro index’ used with just one arg
…/…/matrix/cvector.hh:51: macro index' used with just one arg .../../matrix/cvector.hh:52: macro index’ used with just one arg
…/…/matrix/cvector.hh:53: macro index' used with just one arg .../../matrix/cvector.hh:54: macro index’ used with just one arg
make[1]: *** [cvector.lo] Error 1
make[1]: Leaving directory `/root/NV/nurbs+±3.0.10/x86-pc-nto-qnx/matrix’
make: *** [all-recursive] Error 1
: I repost my problem since I haven’t recieved any further response. This time
: I try to be more clear. You are my only hope )
: I still need help to compile the Nurbs++ libray in QNX Neutrino 6.0. (The
: NURBS library is available via http://telin.rug.ac.be/pub/ftp/sid/Nurbs/,
: version 3.0.10)
…
: g++ -DHAVE_CONFIG_H -I. -I…/…/matrix -I…/include -I…/include -g -O2 -fno
: -imp
: licit-templates -c …/…/matrix/cvector.cc -o cvector.o
: In file included from …/…/matrix/cvector.cc:27:
: …/…/matrix/cvector.hh:50: macro index' used with just one arg : ../../matrix/cvector.hh:51: macro index’ used with just one arg
: …/…/matrix/cvector.hh:52: macro index' used with just one arg : ../../matrix/cvector.hh:53: macro index’ used with just one arg
: …/…/matrix/cvector.hh:54: macro index' used with just one arg : make[1]: *** [cvector.lo] Error 1 : make[1]: Leaving directory /root/NV/nurbs+±3.0.10/x86-pc-nto-qnx/matrix’
: make: *** [all-recursive] Error 1
: #
Sorry, I know longer have a 6.0 machine, but this seems simple
enough. The error message is clear. All you have to do is
edit …/…/matrix/cvector.hh and put
#undef index
somewhere at the beginning.
index() was a macro in the old <strings.h> #define index(s,c) (strchr(s,c))
: I believe the problem is due to gcc using wrong (old or maybe new)
: libraries. The NURBS library is said to be ISO compilant and it works under
: Linux with gcc 2.95.1
: I repost my problem since I haven’t recieved any further response. This time
: I try to be more clear. You are my only hope )
: I still need help to compile the Nurbs++ libray in QNX Neutrino 6.0. (The
: NURBS library is available via http://telin.rug.ac.be/pub/ftp/sid/Nurbs/,
: version 3.0.10)
…
: Please please please, help me…
There is something so … hmmm…
I decided to see what the fuss is all about, I salvage an old
6.0 that was not blown away, yet:
gzip -dc nurbs+±3.0.10.tar.gz |tar xvf -
cd nurbs+±3.0.10
cd matrix
vi cvector.hh
// edit the file and put
// #undef index
// after the #include
// save
cd …
// I was to lazy to get config.sub and config.guess from
// ftp.gnu.org pub/gnu/config)
Why would you suspect libraries? You’re only at the compile stage, not the
link stage… If you look in /usr/include/strings.h, you’ll find a macro
index(s,c) which takes two arguments. This is a case of a name collision so
you’ll have to either not include strings.h, do something like “#undef
index” or rename the index function in the Nurbs library.
I repost my problem since I haven’t recieved any further response. This time
I try to be more clear. You are my only hope > > )
I still need help to compile the Nurbs++ libray in QNX Neutrino 6.0. (The
NURBS library is available via > http://telin.rug.ac.be/pub/ftp/sid/Nurbs/> ,
version 3.0.10)
I believe the problem occurs during linking with gcc 2.95.2. Here is what I
do:
tar -zxvf nurbs+±3.0.10.tar.gz
cd nurbs+±3.0.10
mkdir x86-pc-nto-qnx
(I add new config.sub and config.guess from ftp.gnu.org pub/gnu/config)
cd x86-pc-nto-qnx
…/configure
Everthing seems to work fine till here. Then
gcc --version
2.95.2
make
Making all in include
make[1]: Entering directory /root/NV/nurbs++-3.0.10/x86-pc-nto-qnx/include' make[1]: Leaving directory /root/NV/nurbs+±3.0.10/x86-pc-nto-qnx/include’
Making all in matrix
make[1]: Entering directory /root/NV/nurbs++-3.0.10/x86-pc-nto-qnx/matrix' /bin/sh ../libtool --mode=compile g++ -DHAVE_CONFIG_H -I. -I../../matrix -I../in clude -I../include -g -O2 -fno-implicit-templates -c ../../matrix/cvector.cc g++ -DHAVE_CONFIG_H -I. -I../../matrix -I../include -I../include -g -O2 -fno -imp licit-templates -c ../../matrix/cvector.cc -o cvector.o In file included from ../../matrix/cvector.cc:27: ../../matrix/cvector.hh:50: macro index’ used with just one arg
…/…/matrix/cvector.hh:51: macro index' used with just one arg ../../matrix/cvector.hh:52: macro index’ used with just one arg
…/…/matrix/cvector.hh:53: macro index' used with just one arg ../../matrix/cvector.hh:54: macro index’ used with just one arg
make[1]: *** [cvector.lo] Error 1
make[1]: Leaving directory `/root/NV/nurbs+±3.0.10/x86-pc-nto-qnx/matrix’
make: *** [all-recursive] Error 1
I believe the problem is due to gcc using wrong (old or maybe new)
libraries. The NURBS library is said to be ISO compilant and it works under
Linux with gcc 2.95.1
(> http://yukon.genie.uottawa.ca/~lavoie/software/nurbs/download.shtml> ). I
have it working under VC6.0.
Please please please, help me…
Thanks.
–
Kris Warkentin kewarken@qnx.com
(613)591-0836 x9368
One of the main causes of the fall of the Roman Empire was that, lacking zero,
they had no way to indicate sucessful termination of their C programs.
: I repost my problem since I haven’t recieved any further response. This
time
: I try to be more clear. You are my only hope > > )
: I still need help to compile the Nurbs++ libray in QNX Neutrino 6.0.
(The
: NURBS library is available via http://telin.rug.ac.be/pub/ftp/sid/Nurbs/> ,
: version 3.0.10)
…
: Please please please, help me…
There is something so … hmmm…
I decided to see what the fuss is all about, I salvage an old
6.0 that was not blown away, yet:
gzip -dc nurbs+±3.0.10.tar.gz |tar xvf -
cd nurbs+±3.0.10
cd matrix
vi cvector.hh
// edit the file and put
// #undef index
// after the #include
// save
cd …
// I was to lazy to get config.sub and config.guess from
// ftp.gnu.org pub/gnu/config)