Re: [Quantlib-dev] Changes to compile Java bindings in QuantLib-SWIG module

[Joseph W. <jo...@gn...>]

Yikes.  That option problem is nasty.

I've fixed that in HEAD.  It may take me a few days to go through the other=
=20
patches.  Thanks!!!


=D4=DA Tuesday 30 January 2007 15:20=A3=AC=C4=FA=D0=B4=B5=C0=A3=BA
> Joe,
>
> In the course of playing with QuantLib we've found some problems with
> the Java compilation of the QuantLib-SWIG project.
>
> We've made some modifications to the configuration files, please see
> the attached patch.
> Would be great if this can be included into the HEAD of the
> QuantLib-SWIG project.
> The diffs are based on the QuantLib-SWIG-0.3.14 release, and i don't
> think there have been any changes to the relevant files since the
> release.
>
> The patch file expects there to be a directory QuantLib-SWIG-0.3.14
> and the changes to be made in a dir called
> QuantLib-SWIG-0.3.14_java_patch.  However, you can just run it with
> 'patch -p1 < quantlib-swig_java_ver_1.6-head.patch' in the HEAD of
> QuantLib-SWIG and it'll apply the changes.
>
> Also, it seems that there's a discrepancy with the SWIG/options.i file
> and the way the Option.Type enum is defined. I think the current file
> in the HEAD of SWiG/options.i has the values reversed - the
> Quantlib option.hpp file has it this way at
> http://quantlib.cvs.sourceforge.net/quantlib/QuantLib/ql/option.hpp?revis=
io
>n=3D1.41&view=3Dmarkup enum Type { Put =3D -1, Call =3D 1  };
> while the SWIG/options.i is
> enum Type { Call =3D -1, Put =3D 1};
> I took the liberty of flipping it to look like the option.hpp file in
> the patch as well.
>
> The patch file also contains a small test program Hello.java that does
> a basic sanity check to make sure the created Java library works.
> And it expects the variable JAVA_HOME to be declared and pointing to
> the location of the JDK, and tries to place the generated java wrapper
> library into the lib/ directory under the JRE.
>
> hope this helps. please let me know if you need me to change anything,
> or if things don't work for you.
>
> On 1/29/07, Joseph Wang <jo...@gn...> wrote:
> > The SWIG bindings usually lag the Quantlib libraries a bit, but since I
> > need to use them, I've been trying my best to sync them up.  I'm working
> > mainly with the Python and R binding, so if there is anything broken in
> > the other bindings, let me know and I'll try to fix them.

=2D-=20
=2D------------------------------------------------------------------------=
=2D-----
Joseph Wang Ph.D. - jo...@gn... =20
China Derivatives Researcher and Software Developer - QuantLib
http://en.wikiversity.org/wiki/User:Roadrunner