Problem building python module

[Gerrit Van Dyk-2]

Hi all,

I am experiencing problems building the python module of quantlib. I have
checked out the latest version of both quantlib and quantlib-python. I then
built quantlib and installed it.

I am using swig1.3.9 and python2.2 on Redhat linux 7.2. I ran
makewrappers.py to generate the swig file. This ran through successfully.

I am now trying to run "python2 setup.py build" but this returns a compile
error complaining about limits that could not be found.
The include statement is in /usr/local/include/ql/qldefines.hpp:146:22. This
include just includes <limits>.


What should I do to solve the error and get the python module built for
quantlib? Any help would be appreciated.

Regards
Gerrit van Dyk
 
-------------------------------------------------------------------------
This e-mail is intended only for the use of the individual or entity named
above and may contain information that is confidential and privileged,
proprietary to the company and protected by law. If you are not the intended
recipient, you are hereby notified that any dissemination, distribution or
copying of this e-mail is strictly prohibited. Opinions, conclusions and
other information in this message that do not relate to the official
business of our company shall be understood as neither given nor endorsed by
it.

[Luigi Ballabio-4]

At 11:02 AM 4/16/02 +0200, Gerrit Van Dyk wrote:
>I am now trying to run "python2 setup.py build" but this returns a compile
>error complaining about limits that could not be found.
>The include statement is in /usr/local/include/ql/qldefines.hpp:146:22. This
>include just includes <limits>.

Gerrit,
         I'll give it a shot. First of all, please take a look inside
/usr/local/include/ql/config.hpp or wherever the QuantLib headers got
installed and tell me whether there is a line inside reading
#define HAVE_LIMITS 1
(note: LIMITS, not CLIMITS neither LIMITS_H)
Oh, and another piece of info: which gcc version are you running?

Later,
         Luigi

[Gerrit Van Dyk-2]

Luigi wrote:

> At 11:02 AM 4/16/02 +0200, Gerrit Van Dyk wrote:
> >I am now trying to run "python2 setup.py build" but this
> returns a compile
> >error complaining about limits that could not be found.
> >The include statement is in
> /usr/local/include/ql/qldefines.hpp:146:22. This
> >include just includes <limits>.
>
> Gerrit,
>          I'll give it a shot. First of all, please take a look inside
> /usr/local/include/ql/config.hpp or wherever the QuantLib headers got
> installed and tell me whether there is a line inside reading
> #define HAVE_LIMITS 1
> (note: LIMITS, not CLIMITS neither LIMITS_H)
> Oh, and another piece of info: which gcc version are you running?
>

 

I am using gcc3(3.0.2) and g++3(3.0.2)

Yes there is a line #define HAVE_LIMITS 1, I have attached the file for
convenience

 
-------------------------------------------------------------------------
This e-mail is intended only for the use of the individual or entity named
above and may contain information that is confidential and privileged,
proprietary to the company and protected by law. If you are not the intended
recipient, you are hereby notified that any dissemination, distribution or
copying of this e-mail is strictly prohibited. Opinions, conclusions and
other information in this message that do not relate to the official
business of our company shall be understood as neither given nor endorsed by
it.
 

[Luigi Ballabio-4]

At 12:30 PM 4/16/02 +0200, Gerrit Van Dyk wrote:
>I am using gcc3(3.0.2) and g++3(3.0.2)
>
>Yes there is a line #define HAVE_LIMITS 1

Hmm. This should mean that configure *did* find <limits> during its tests.
Ok, let's track this down. Can you check where limits is located in your
system? Also, can you show me the full output of python setup.py build?

Later,
         Luigi

[Gerrit Van Dyk-2]

>Ok, let's track this down. Can you check where limits is
>located in your
>system? Also, can you show me the full output of python setup.py build?

1) limits is located in /usr/include/g++-v3/ as limits

2) Here is the output from: python2 setup.py build

[gerrit@venus1 QuantLib-Python]$ python2 setup.py build
running build
running build_py
not copying QuantLib/__init__.py (output up-to-date)
not copying QuantLib/QuantLib.py (output up-to-date)
not copying QuantLib/defaults.py (output up-to-date)
running build_ext
my_init_posix: changing gcc to g++
building 'QuantLib.QuantLibc' extension
g++ -DNDEBUG -g -O3 -Wall -Wstrict-prototypes -fPIC -I/usr/local/include
-I/usr/local/include -I/usr/include -I/usr/include/python2.2 -c
QuantLib/quantlib_wrap.cpp -o build/temp.linux-i686-2.2/quantlib_wrap.o
-I/usr/local/include -I/usr/local/include
In file included from /usr/local/include/ql/quantlib.hpp:23,
                 from QuantLib/quantlib_wrap.cpp:704:
/usr/local/include/ql/qldefines.hpp:146:22: limits: No such file or
directory
error: command 'g++' failed with exit status 1
[gerrit@venus1 QuantLib-Python]$

Regards
Gerrit
 
-------------------------------------------------------------------------
This e-mail is intended only for the use of the individual or entity named
above and may contain information that is confidential and privileged,
proprietary to the company and protected by law. If you are not the intended
recipient, you are hereby notified that any dissemination, distribution or
copying of this e-mail is strictly prohibited. Opinions, conclusions and
other information in this message that do not relate to the official
business of our company shall be understood as neither given nor endorsed by
it.

[Gerrit Van Dyk-2]

Ok, Thanks Luigi I seem to have found the missing "limits" problem. Although
I have environment variables set up for using my preferred compiler setup.py
still picks up the incorrect compiler.
I have an environment variable CC that points to gcc3 and CCC that points to
g++3. For the normal make and GNU tools this seems to be sufficient to
enable your preferred compiler. The distutils package however defaults to
gcc and g++

Looking closer at the output of "python setup.py build" I realized that it
was calling g++ and not g++3.

The command just completed successful with a minor warning. I just ran
"python2 setup.py test" this too went ok. Now I need to get into it and
start playing.

Thanks again.

Regards
Gerrit van Dyk
 
-------------------------------------------------------------------------
This e-mail is intended only for the use of the individual or entity named
above and may contain information that is confidential and privileged,
proprietary to the company and protected by law. If you are not the intended
recipient, you are hereby notified that any dissemination, distribution or
copying of this e-mail is strictly prohibited. Opinions, conclusions and
other information in this message that do not relate to the official
business of our company shall be understood as neither given nor endorsed by
it.

[Luigi Ballabio-4]

At 02:36 PM 4/16/02 +0200, Gerrit Van Dyk wrote:
>The command just completed successful with a minor warning.

Gerrit,
         glad to hear that. I assume that the warning is

In function `char* SWIG_UnpackData(char*, void*, int)':
warning: `unsigned char uu' might be used uninitialized in this function

in which case you can safely ignore it.

Have fun,
                 Luigi