quantlib-dev — Discussion of QuantLib development

Re: [Quantlib-dev] Other tentative 0.9.9 tarballs and partial branch freeze

[Dirk E. <ed...@de...>]

On 4 November 2009 at 18:15, Luigi Ballabio wrote:
| Report here any problems you may have.  As usual, I'd particularly
| appreciate if you tried the library on cygwin or mingw, as I don't have
| a test environment for those platforms.

MinGW success using Dev-C++ (which I last used to build QL 0.9.7); also built
one example (FittedBondCurve) which runs. Looks good.

Also updating the Debian package to 0.9.9~20091104-1.

Dirk

-- 
Three out of two people have difficulties with fractions.

[Quantlib-dev] Other tentative 0.9.9 tarballs and partial branch freeze

[Luigi B. <lui...@gm...>]

Hi all,
	as it turns out, the latest tarballs weren't the final ones after all.
So, if you have any more cycles to spare, please download and try out
the new tarballs at <http://quantlib.org/prerelease/>.  If nobody
reports any showstoppers by next Wednesday, these ones are going to be
released.

Report here any problems you may have.  As usual, I'd particularly
appreciate if you tried the library on cygwin or mingw, as I don't have
a test environment for those platforms.

Also, as we're close to release, I'm partially freezing the 0.9.9
branch; please don't commit anything in its QuantLib and QuantLib-SWIG
modules without checking with me first.

Thanks,
	Luigi


-- 

There are two ways of constructing a software design. One way is to 
make it so simple that there are obviously no deficiencies. And the 
other way is to make it so complicated that there are no obvious 
deficiencies. 
-- C. A. R. Hoare 

[Quantlib-dev] QuantLib.xla improvement

[Alexander L. <ale...@go...>]

Ciao Nando,

I had some problems with QuantLib.xla and did som improvements.

First of all: 

Module -> ObjectBuilder -> Function ListFilesInFolder

...
If FileItem.Type = "XML Document" - doesn't work for the German version :-(
I extended it with "XML-Dokument"
...

Maybe there is a better idea how to check the file. For example using its
extension.

Second:

Module -> ObjectBuilder -> Function buildConstructor

...
values(1,1) = "=ohRangeRetrieveError(R[-1]C)" doesn't work for the older
excel version (2002)
...

I exchanged it with values(1,1) = "=ohRangeRetrieveError(" &
dest.offset(0,1).Cells.Address & ")"
...

Hope it will be helpfull. I've spent some time trying to figure out the
problem.

Ciao,

Alexander
-- 
View this message in context: http://old.nabble.com/QuantLib.xla-improvement-tp26199309p26199309.html
Sent from the quantlib-dev mailing list archive at Nabble.com.

Re: [Quantlib-dev] QuantLibXL.xla Problem with QuantLib 0.9.9

[Ferdinando A. <na...@am...>]

On Tue, Nov 3, 2009 at 8:41 PM, Alexander Lotter
<ale...@go...> wrote:
> I've just renamed this file (no file - no problem as written in your comment
> in svn) and it worked. So please take a look at it if you have time and
> update the svn.

I've fixed the file on the 0.9.9 branch, so now it should work with
the xml config file too

ciao -- Nando

Re: [Quantlib-dev] Tentative 0.9.9 tarballs

[Chris K. <chr...@ya...>]

Hi,

on OS X 10.6.1 (Snow Leopard) + boost 1.40.0
build OK
tests pass, but fails on exiting with:

*** No errors detected
quantlib-test-suite(15731) malloc: *** error for object 0x3000102504200: pointer being freed was not allocated
*** set a breakpoint in malloc_error_break to debug
Abort trap


Best regards,
Chris

Re: [Quantlib-dev] Linking Error on QuantLibXL_full

[Ferdinando A. <na...@am...>]

On Tue, Nov 3, 2009 at 8:33 PM, Alexander Lotter
<ale...@go...> wrote:
> Should I use the bug tracker for such small things?

I've fixed it on the 0.9.9 branch

thank you

ciao -- Nando

Re: [Quantlib-dev] QuantLibXL.xla Problem with QuantLib 0.9.9

[Alexander L. <ale...@go...>]

I've just renamed this file (no file - no problem as written in your 
comment in svn) and it worked. So please take a look at it if you have 
time and update the svn.

Re: [Quantlib-dev] Linking Error on QuantLibXL_full

[Alexander L. <ale...@go...>]

No problem for me. Just wanted to let you know.

Should I use the bug tracker for such small things?


Ferdinando Ametrano wrote:
> On Tue, Nov 3, 2009 at 2:49 PM, Alexander Lotter
> <ale...@go...> wrote:
>   
>> there is no file like register_overnighindexedswap.cpp in
>> QuantLibXLStatic-Project.
>>
>> In QuantLibAllDynamic it is included, so there is no error compiling the
>> project.
>>
>> Does the generation tool includes the files to the project automatically?
>>     
>
> the files are automatically generated and manually added to the
> project. So it's just that it has been added to QuantLibAllDynamic and
> not to QuantLibXLStatic.
> It will be fixed on the trunk, in the meantime please add it manually
> to your version
>
> ciao -- Nando
>
>   

Re: [Quantlib-dev] QuantLibXL.xla Problem with QuantLib 0.9.9

[Alexander L. <ale...@go...>]

Hello Ferdinando,

that is what I get using vc8 and vc9 libraries. I think it is not 
depending on some configuration and lib. I've just checked out a trunk 
version again and there is still a problem. I can't find this tag 
FeedUse in the xml and I don't understand why this is not an optional 
node/attribute.

<Environment name="Local">
  <FrameworkVersion>10</FrameworkVersion>
  <Workbooks>C:\Projects\DevEnv\trunk\QuantLibXL\Workbooks</Workbooks>
  <HelpFile>C:\Projects\DevEnv\trunk\QuantLibXL\Docs</HelpFile>
  
<FunctionMetadata>C:\Projects\DevEnv\trunk\QuantLibAddin\gensrc\metadata</FunctionMetadata>
  <AddinList>
    <Addin>
      
<Path>C:\Projects\DevEnv\trunk\ObjectHandler\xll\ObjectHandler-xll-vc80-mt-0_9_8.xll</Path>
      <DeliveredByLauncher>False</DeliveredByLauncher>
    </Addin>
    <Addin>
      
<Path>C:\Projects\DevEnv\trunk\QuantLibXL\xll\QuantLibXLDynamic-vc80-mt-0_9_8.xll</Path>
      <DeliveredByLauncher>False</DeliveredByLauncher>
    </Addin>
  </AddinList>
  <StartupActions>
    <SetEvaluationDate>False</SetEvaluationDate>
    <EvaluationDate>39462</EvaluationDate>
    <YieldCurveBootstrap>False</YieldCurveBootstrap>
    <CapVolBootstrap>False</CapVolBootstrap>
    <SwapSmileBootstrap>False</SwapSmileBootstrap>
    <CalibrateCMS>False</CalibrateCMS>
    <FitCMS>False</FitCMS>
    <IndexesTimeSeries>False</IndexesTimeSeries>
    <LoadBonds>False</LoadBonds>
    <MainChecks>True</MainChecks>
    <StaticData>False</StaticData>
    <InitSource>Excel</InitSource>
  </StartupActions>
  <Variables />
</Environment>

Take care,

Alexander

Ferdinando Ametrano wrote:
> On Tue, Nov 3, 2009 at 3:32 PM, Alexander Lotter
> <ale...@go...> wrote:
>   
>> I got the following error loading XLA:
>>
>> Error initializing the QuantLibXL framework: there was an error processing
>> QuantLibXL.launch.xml
>>
>> expected node with tag Environment to ahve exactly one child node with tag
>> FeedUse.
>>
>> Are there some problems with the QuantLibXL.launch.xml in trunk?
>>
>> The only thing I did with the file is - changing pathes to *.xll...
>>     
>
> chances are you did more than changing paths... the error is about
> invalid xml file. Have you checked that you only have exactly one
> child node with tag FeedUse ?
>
> ciao -- Nando
>
>   

Re: [Quantlib-dev] QuantLibXL.xla Problem with QuantLib 0.9.9

[Ferdinando A. <na...@am...>]

On Tue, Nov 3, 2009 at 3:32 PM, Alexander Lotter
<ale...@go...> wrote:
> I got the following error loading XLA:
>
> Error initializing the QuantLibXL framework: there was an error processing
> QuantLibXL.launch.xml
>
> expected node with tag Environment to ahve exactly one child node with tag
> FeedUse.
>
> Are there some problems with the QuantLibXL.launch.xml in trunk?
>
> The only thing I did with the file is - changing pathes to *.xll...

chances are you did more than changing paths... the error is about
invalid xml file. Have you checked that you only have exactly one
child node with tag FeedUse ?

ciao -- Nando

Re: [Quantlib-dev] Linking Error on QuantLibXL_full

[Ferdinando A. <na...@am...>]

On Tue, Nov 3, 2009 at 2:49 PM, Alexander Lotter
<ale...@go...> wrote:
> there is no file like register_overnighindexedswap.cpp in
> QuantLibXLStatic-Project.
>
> In QuantLibAllDynamic it is included, so there is no error compiling the
> project.
>
> Does the generation tool includes the files to the project automatically?

the files are automatically generated and manually added to the
project. So it's just that it has been added to QuantLibAllDynamic and
not to QuantLibXLStatic.
It will be fixed on the trunk, in the meantime please add it manually
to your version

ciao -- Nando

[Quantlib-dev] QuantLibXL.xla Problem with QuantLib 0.9.9

[Alexander L. <ale...@go...>]

Does anybody expirienced a problem loading QuantLibXL.xla from the actual
trunk?

I got the following error loading XLA: 

Error initializing the QuantLibXL framework: there was an error processing
QuantLibXL.launch.xml

expected node with tag Environment to ahve exactly one child node with tag
FeedUse.

Are there some problems with the QuantLibXL.launch.xml in trunk?

The only thing I did with the file is - changing paths to *.xll...

Cheers,

Alexander
-- 
View this message in context: http://old.nabble.com/QuantLibXL.xla-Problem-with-QuantLib-0.9.9-tp26160162p26160162.html
Sent from the quantlib-dev mailing list archive at Nabble.com.

[Quantlib-dev] QuantLibXL.xla Problem with QuantLib 0.9.9

[Alexander L. <ale...@go...>]

Does anybody expirienced a problem loading QuantLibXL.xla from the actual
trunk?

I got the following error loading XLA: 

Error initializing the QuantLibXL framework: there was an error processing
QuantLibXL.launch.xml

expected node with tag Environment to ahve exactly one child node with tag
FeedUse.

Are there some problems with the QuantLibXL.launch.xml in trunk?

The only thing I did with the file is - changing pathes to *.xll...

Cheers,

Alexander
-- 
View this message in context: http://old.nabble.com/QuantLibXL.xla-Problem-with-QuantLib-0.9.9-tp26160160p26160160.html
Sent from the quantlib-dev mailing list archive at Nabble.com.

[Quantlib-dev] Linking Error on QuantLibXL_full

[Alexander L. <ale...@go...>]

Dear all,

I tried to build QuantLibXLStatic and get linking errors: 

register_all.obj: LNK2019: void __cdecl registerOvernightindexedswap... and
unregisterOvernightindexedswap...

QuantLib 0.9.9
QuantLibXL_full_vc8
MS VS 2005
Boost 1_40
Debug
Mixed Platforms

QuantLibAllDynamic_vc8.sln - works fine.

This functions look to me same as the other functions that defined as extern
and doesn't produce any linking error. Does anybody have an idea?

Cheers

Alexander
-- 
View this message in context: http://old.nabble.com/Linking-Error-on-QuantLibXL_full-tp26160118p26160118.html
Sent from the quantlib-dev mailing list archive at Nabble.com.

Re: [Quantlib-dev] Linking Error on QuantLibXL_full

[Alexander L. <ale...@go...>]

Update:

This function was defined in register_all.cpp - which is generated
automaticall but there is no file like register_overnighindexedswap.cpp in
QuantLibXLStatic-Project.

In QuantLibAllDynamic it is included, so there is no error compiling the
project.

Does the generation tool includes the files to the project automatically? In
case of QuantLibXLStatic it was not done for some reason only for this one
file.

Does anybody had a similar problem?
-- 
View this message in context: http://old.nabble.com/Linking-Error-on-QuantLibXL_full-tp26160118p26160127.html
Sent from the quantlib-dev mailing list archive at Nabble.com.

Re: [Quantlib-dev] QuantLib-SWING: is it possible to use polymorphism?

[Luigi B. <lui...@gm...>]

On Thu, 2009-10-29 at 17:59 +0000, Andrea wrote:
> > There would be quite a bit of work to do.  You would have to:
> > - add a constructor to boost::shared_ptr in common.i so that it takes a
> > pointer;
> 
> I've got a big doubt here.
> Is it safe to stick the raw pointer into a shared_ptr? Is it already owned by some other swig mechanism?

You're right, the shared_ptr would take ownership against SWIG.
I don't see any easy solution to this one...

Another possibility would be to write a C++ payoff that adapts a Java
payoff. Its constructor would take a Java instance and store it, and its
operator() would use the foreign interface to call the corresponding
method in the Java instance. Kind of a home-made director. I don't know
if it would be possible to hide the mechanism so that it seems that the
Java class inherits from Payoff, though...

Luigi


-- 

There's no sense in being precise when you don't even know what you're 
talking about. 
-- John von Neumann 

Re: [Quantlib-dev] Tentative 0.9.9 tarballs

[Roland L. <rol...@go...>]

>
> Hi all,
>        if you have any cycles to spare during the weekend, please download
> and
> try out the tarballs at <http://quantlib.org/prerelease/>.  If nobody
> reports any showstoppers by the middle of next week, they're going to be
> the final ones.
>
> Report here any problems you may have.  I'd particularly appreciate if
> you tried the library on cygwin or mingw, as I don't have a test
> environment for those platforms.
>

Hi Luigi,

compiled and test suite completed without errors on
- Mac OS X 10.5.7 with boost 1.39
- MinGW under Parallels/Vista with boost 1.34.1

Best regards,
Roland

Re: [Quantlib-dev] Problems with QL 0.9.7 and Cubic interpolation for curves

[Dirk E. <ed...@de...>]

On 29 October 2009 at 15:25, Luigi Ballabio wrote:
| On Sat, 2009-10-17 at 10:35 -0500, Dirk Eddelbuettel wrote:
| > While working on RQuantLib [1], Khanh and I discovered a number of numerical
| > issues with Cubic interpolation on yield curves.  The RQuantLib side of the
| > code that triggers it is actually pretty old and used to run so I am confused
| > as to when this changed.  
| > 
| > A relatively simple way to trigger the same issue is to modify the
| > swapvaluation.cpp example as follows:
| > 
| > edd@ron:~/svn/quantlib/Examples/Swap$ diff -u swapvaluation.cpp.orig swapvaluation.cpp
| > --- swapvaluation.cpp.orig      2008-08-01 03:55:43.000000000 -0500
| > +++ swapvaluation.cpp   2009-10-17 10:08:42.000000000 -0500
| > @@ -316,7 +316,7 @@
| >          depoSwapInstruments.push_back(s10y);
| >          depoSwapInstruments.push_back(s15y);
| >          boost::shared_ptr<YieldTermStructure> depoSwapTermStructure(
| > -            new PiecewiseYieldCurve<Discount,LogLinear>(
| > +            new PiecewiseYieldCurve<Discount,Cubic>(
| >                                            settlementDate, depoSwapInstruments,
| >                                            termStructureDayCounter,
| >                                            std::vector<Handle<Quote> >(),
| > 
| > and you get to trigger the bug as easily as 
| > 
| > edd@ron:~/svn/quantlib/Examples/Swap$ g++ -o newswap swapvaluation.cpp -lQuantLib -lm && ./newswap
| 
| Dirk,
| 	as of today, I couldn't reproduce the error on the 0.9.9 release branch
| (branches/R000909-branch in Subversion.)  The swap example runs
| successfully after the change.  Does it work for you? Of course it's not
| picking up some other library version you installed earlier, is it?

Confirmed!  I (finally) got a chance to build QL 0.9.9 (using 20091023
tarballs) and things work now.  Very good news.  

I only needed a few touchups to get RQuantLib to build with it too.

So I put 0.9.9~20091023 into Debian -- and may follow up with the Nov 2
variant you released today.  The full build and regression test takes well
over an hour but seemingly passed each and every test.  

Looking forward to QL 1.0!!

Dirk

-- 
Three out of two people have difficulties with fractions.

[Quantlib-dev] Tentative 0.9.9 tarballs

[Luigi B. <lui...@gm...>]

Hi all,
	if you have any cycles to spare during the weekend, please download and
try out the tarballs at <http://quantlib.org/prerelease/>.  If nobody
reports any showstoppers by the middle of next week, they're going to be
the final ones.

Report here any problems you may have.  I'd particularly appreciate if
you tried the library on cygwin or mingw, as I don't have a test
environment for those platforms.

Thanks,
	Luigi


-- 

There are two ways of constructing a software design. One way is to 
make it so simple that there are obviously no deficiencies. And the 
other way is to make it so complicated that there are no obvious 
deficiencies. 
-- C. A. R. Hoare 

Re: [Quantlib-dev] QuantLib-SWING: is it possible to use polymorphism?

[Andrea <mar...@go...>]

On 31/10/09 14:15, Tito Ingargiola wrote:
> 
> 
> Hi Andrea,
> 
> Another approach you may consider is illustrated in the DiscreteHedging.java example. 
> Here I wanted to use Quantlib's MC capabilities but wanted the pricer to be implemented in java.
> By pulling the MonteCarloModel impl into java I was able to use a pathpricer implemented in java.


When you say "pulling", you mean you have rewritten it in Java, mirroring the PathPricer->JPathPricer.

What I am trying to do is to have the JPathPricer automatically generated by swig.
This is actually easy since JPathPricer only returns a double.

My problem is the class "Option" that usually returns a shared_ptr to a "Payoff" and here I have
problems.


> Not sure if this is applicable to your problem, but it's an approach I've used in a number of cases to integrate quantlib and java. 
> You may also implement your Payoff in C++ and then expose it yourself via swig to your java....

This is certanly possible, but I would like to avoid writing any C++ and only use Java.

> 
> Best,
> 
>      Tito.

Re: [Quantlib-dev] QuantLib-SWING: is it possible to use polymorphism?

[Tito I. <tit...@ya...>]

Hi Andrea,

Another approach you may consider is illustrated in the DiscreteHedging.java example.  Here I wanted to use Quantlib's MC capabilities but wanted the pricer to be implemented in java.  By pulling the MonteCarloModel impl into java I was able to use a pathpricer implemented in java.  

Not sure if this is applicable to your problem, but it's an approach I've used in a number of cases to integrate quantlib and java.  You may also implement your Payoff in C++ and then expose it yourself via swig to your java....

Best,

     Tito.


----- Original Message ----
From: Andrea <mar...@go...>
To: qua...@li...
Sent: Thu, October 29, 2009 1:59:42 PM
Subject: Re: [Quantlib-dev] QuantLib-SWING: is it possible to use polymorphism?

On 29/10/09 16:03, Luigi Ballabio wrote:
> On Wed, 2009-10-21 at 17:58 +0100, Andrea wrote:
>> On 21/10/09 15:55, Luigi Ballabio wrote:
>>>     unfortunately, polymorphism doesn't currently work across
>> languages.
>>>
>> Indeed.
>> I've tried to use directors and it seems to work well if there are no
>> shared_ptr involved.
>>
>> My problem now is [...] I don't seem to be able to create a shared_ptr
>> in the target language.
> 
> There would be quite a bit of work to do.  You would have to:
> - add a constructor to boost::shared_ptr in common.i so that it takes a
> pointer;

I've got a big doubt here.
Is it safe to stick the raw pointer into a shared_ptr? Is it already owned by some other swig mechanism?

> - modify the exported payoff classes so that they no longer hide the
> shared_ptr (e.g., export the actual PlainVanillaPayoff, not
> PlainVanillaPayoffPtr) and enable directors;
> - export the shared_ptr<Payoff> as a separate class;
> - fix the other interfaces so that they work with the new classes.
> 
> Let me know if you get stuck in any of the steps.
> 
> Luigi
> 
> 



------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay 
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
QuantLib-dev mailing list
Qua...@li...
https://lists.sourceforge.net/lists/listinfo/quantlib-dev

Re: [Quantlib-dev] ql/experimental/inflation/yoycapfloortermpricesurface.hpp: bug in intersect() ? (raffaele.pellicani@zkb.ch)

[Chris K. <chr...@ya...>]

Hi,

yes, this was a bug (thanks Raffaele). The fix is now in the 0.99 branch as:

if ( floorPrice_(t,fStrikes_.back()) > capPrice_(t,fStrikes_.back()) ) {

logic as before.  

Changing to fStrikes_.back() on both sides of the comparison ensures that the same strike is used (also as suggested), otherwise strange effects are possible.    

Best regards,
Chris

Re: [Quantlib-dev] QuantLib-SWING: is it possible to use polymorphism?

[Andrea <mar...@go...>]

On 29/10/09 16:03, Luigi Ballabio wrote:
> On Wed, 2009-10-21 at 17:58 +0100, Andrea wrote:
>> On 21/10/09 15:55, Luigi Ballabio wrote:
>>> 	unfortunately, polymorphism doesn't currently work across
>> languages.
>>>
>> Indeed.
>> I've tried to use directors and it seems to work well if there are no
>> shared_ptr involved.
>>
>> My problem now is [...] I don't seem to be able to create a shared_ptr
>> in the target language.
> 
> There would be quite a bit of work to do.  You would have to:
> - add a constructor to boost::shared_ptr in common.i so that it takes a
> pointer;

I've got a big doubt here.
Is it safe to stick the raw pointer into a shared_ptr? Is it already owned by some other swig mechanism?

> - modify the exported payoff classes so that they no longer hide the
> shared_ptr (e.g., export the actual PlainVanillaPayoff, not
> PlainVanillaPayoffPtr) and enable directors;
> - export the shared_ptr<Payoff> as a separate class;
> - fix the other interfaces so that they work with the new classes.
> 
> Let me know if you get stuck in any of the steps.
> 
> Luigi
> 
> 

Re: [Quantlib-dev] QuantLib-SWING: is it possible to use polymorphism?

[Luigi B. <lui...@gm...>]

On Wed, 2009-10-21 at 17:58 +0100, Andrea wrote:
> On 21/10/09 15:55, Luigi Ballabio wrote:
> > 	unfortunately, polymorphism doesn't currently work across
> languages.
> > 
> Indeed.
> I've tried to use directors and it seems to work well if there are no
> shared_ptr involved.
> 
> My problem now is [...] I don't seem to be able to create a shared_ptr
> in the target language.

There would be quite a bit of work to do.  You would have to:
- add a constructor to boost::shared_ptr in common.i so that it takes a
pointer;
- modify the exported payoff classes so that they no longer hide the
shared_ptr (e.g., export the actual PlainVanillaPayoff, not
PlainVanillaPayoffPtr) and enable directors;
- export the shared_ptr<Payoff> as a separate class;
- fix the other interfaces so that they work with the new classes.

Let me know if you get stuck in any of the steps.

Luigi


-- 

Green's Law of Debate: 
Anything is possible if you don't know what you're talking about. 

Re: [Quantlib-dev] Problems with QL 0.9.7 and Cubic interpolation for curves

[Luigi B. <lui...@gm...>]

On Sat, 2009-10-17 at 10:35 -0500, Dirk Eddelbuettel wrote:
> While working on RQuantLib [1], Khanh and I discovered a number of numerical
> issues with Cubic interpolation on yield curves.  The RQuantLib side of the
> code that triggers it is actually pretty old and used to run so I am confused
> as to when this changed.  
> 
> A relatively simple way to trigger the same issue is to modify the
> swapvaluation.cpp example as follows:
> 
> edd@ron:~/svn/quantlib/Examples/Swap$ diff -u swapvaluation.cpp.orig swapvaluation.cpp
> --- swapvaluation.cpp.orig      2008-08-01 03:55:43.000000000 -0500
> +++ swapvaluation.cpp   2009-10-17 10:08:42.000000000 -0500
> @@ -316,7 +316,7 @@
>          depoSwapInstruments.push_back(s10y);
>          depoSwapInstruments.push_back(s15y);
>          boost::shared_ptr<YieldTermStructure> depoSwapTermStructure(
> -            new PiecewiseYieldCurve<Discount,LogLinear>(
> +            new PiecewiseYieldCurve<Discount,Cubic>(
>                                            settlementDate, depoSwapInstruments,
>                                            termStructureDayCounter,
>                                            std::vector<Handle<Quote> >(),
> 
> and you get to trigger the bug as easily as 
> 
> edd@ron:~/svn/quantlib/Examples/Swap$ g++ -o newswap swapvaluation.cpp -lQuantLib -lm && ./newswap

Dirk,
	as of today, I couldn't reproduce the error on the 0.9.9 release branch
(branches/R000909-branch in Subversion.)  The swap example runs
successfully after the change.  Does it work for you? Of course it's not
picking up some other library version you installed earlier, is it?

Later,
	Luigi


-- 

It is better to know some of the questions than all of the answers. 
-- James Thurber