quantlib-dev — Discussion of QuantLib development

Re: [Quantlib-dev] Error with old-style implied vol

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

On Wed, Nov 26, 2003 at 10:29:04PM -0600, Dirk Eddelbuettel wrote:
> 
> In the process of updating RQuantLib to 0.3.4, I found that my wrappers
> around European and American implied vol no longer work. [ That I found that
> is a good thing; R stronly encourages a manual page for every function.
> These manual pages usually have examples, and during the 'make check' phase
> all of the available examples are run. Quite nice. ]
> 
> I do know that I'll have to convert to the new prices one day anyway, but is
> there a shortcut to keep the current alive while I prepare one based on the
> new pricing engines?  
> 
> I include a simple stand-alone below.  It builds fine, compute the NPV of an
> option (just to show that it links right etc pp) but abort on implied vol.
> 
> CHROOT edd@chibud:~/src/progs/C++$ g++ -Wall -o test_ql_impliedvol
> test_ql_impliedvol.cc -lQuantLib
> CHROOT edd@chibud:~/src/progs/C++$ ./test_ql_impliedvol
> Value:  5.81756
> Aborted
> 
> Here is the short program:
> 
> 
> // implied vol in RQuantLib-0.1.7 blows up with QL 0.3.4
> 
> #include <ql/quantlib.hpp>
> #include <iostream>
> 
> using namespace QuantLib;
> using QuantLib::Pricers::EuropeanOption;
> 
> int main(void) {
>   double underlying = 100;
>   double strike = 100;
>   double dividendYield = 0.02;
>   double riskFreeRate = 0.03;
>   double maturity = 0.5; 
>   double volatility = 0.2;
>   Option::Type optionType = Option::Call;
> 
>   EuropeanOption EO = EuropeanOption(optionType, underlying, strike,
> 				     dividendYield, riskFreeRate, maturity, 
> 				     volatility);
>   std::cout << "Value:  " << EO.value() << std::endl;
>   double impliedVol = EO.impliedVolatility(EO.value()+0.1);
>   std::cout << "Implied:" << impliedVol << std::endl;
>   exit(0);
> }
> 
> Thanks in advance for any insights.

Given the deafening silence :) as well as some curiousity on my part as to
whether I'd manage to understand the fancy-pants new-style C++, I turned the
above example into a small-ish standalone -- see below.

Now that I crossed that bridge, is there an FAQ / site I could consult to
upgrade my early-1990s understanding of C++ to learn about the handles etc?

Cheers, Dirk


// implied vol in RQuantLib-0.1.7 blows up with QL 0.3.4

#define NEW 1

#ifdef OLD
#include <ql/quantlib.hpp>
#include <iostream>

using namespace QuantLib;
using QuantLib::Pricers::EuropeanOption;

int main(void) {
  double underlying = 100;
  double strike = 100;
  double dividendYield = 0.02;
  double riskFreeRate = 0.03;
  double maturity = 0.5; 
  double volatility = 0.2;
  Option::Type optionType = Option::Call;

  EuropeanOption EO = EuropeanOption(optionType, underlying, strike,
				     dividendYield, riskFreeRate, maturity, 
				     volatility);
  double value = EO.value();
  std::cout << "Value:  " << value << std::endl;
  double impliedVol = EO.impliedVolatility(value-0.1);
  std::cout << "Implied:" << impliedVol << std::endl;
  exit(0);
}
#endif 

#include <ql/quantlib.hpp>
#include <iostream>

#include <ql/Instruments/vanillaoption.hpp>
#include <ql/PricingEngines/mceuropeanengine.hpp>
#include <ql/TermStructures/flatforward.hpp>
#include <ql/Volatilities/blackconstantvol.hpp>
#include <ql/Calendars/target.hpp>
#include <cppunit/TestSuite.h>
#include <cppunit/TestCaller.h>
#include <map>


using namespace QuantLib;
using namespace QuantLib::PricingEngines;
using namespace QuantLib::Instruments;
using namespace QuantLib::TermStructures;
using namespace QuantLib::VolTermStructures;
using namespace QuantLib::DayCounters;
using namespace QuantLib::Calendars;
using namespace QuantLib::MonteCarlo;

int main(void) {
  Option::Type type = Option::Call;
  double strike = 100;
  int length = 180;
  double underlyingprice = 100.0;
  Rate qRateval = 0.01;
  Rate rRateval = 0.05;
  double volval = 0.20;

  Date today = Date::todaysDate();
  Date reference = TARGET().advance(today,2,Days);

  Handle<SimpleMarketElement> underlying(new SimpleMarketElement(0.0));
  Handle<SimpleMarketElement> volatility(new SimpleMarketElement(0.0));
  Handle<BlackVolTermStructure> 
    volCurve(new BlackConstantVol(reference,
				  RelinkableHandle<MarketElement>(volatility),
				  Actual365()));
  Handle<SimpleMarketElement> qRate(new SimpleMarketElement(0.0));
  Handle<TermStructure> 
    divCurve(new FlatForward(today, reference,
			     RelinkableHandle<MarketElement>(qRate),
			     Actual365()));
  Handle<SimpleMarketElement> rRate(new SimpleMarketElement(0.0));
  Handle<TermStructure> 
    rfCurve(new FlatForward(today, reference,
			     RelinkableHandle<MarketElement>(rRate),
			     Actual365()));
  Handle<PricingEngine> engine(new AnalyticEuropeanEngine);

  Date exDate = today.plusDays(length);
  Handle<VanillaOption> 
    option(new VanillaOption(type, 
			     RelinkableHandle<MarketElement>(underlying), 
			     strike,
			     RelinkableHandle<TermStructure>(divCurve), 
			     RelinkableHandle<TermStructure>(rfCurve), 
			     EuropeanExercise(exDate), 
			     RelinkableHandle<BlackVolTermStructure>(volCurve),
			     engine));

  underlying->setValue(underlyingprice);
  qRate->setValue(qRateval);
  rRate->setValue(rRateval);
  volatility->setValue(volval);
  
  double value = option->NPV();
  volatility->setValue(volval*1.5); // initialize as too high
  double implVol = option->impliedVolatility(value+0.1);
	      
  std::cout << "Value:  " << value << std::endl;
  std::cout << "Implied:" << implVol << std::endl;

}


-- 
Those are my principles, and if you don't like them... well, I have others.
                                                -- Groucho Marx

[Quantlib-dev] Error with old-style implied vol

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

In the process of updating RQuantLib to 0.3.4, I found that my wrappers
around European and American implied vol no longer work. [ That I found that
is a good thing; R stronly encourages a manual page for every function.
These manual pages usually have examples, and during the 'make check' phase
all of the available examples are run. Quite nice. ]

I do know that I'll have to convert to the new prices one day anyway, but is
there a shortcut to keep the current alive while I prepare one based on the
new pricing engines?  

I include a simple stand-alone below.  It builds fine, compute the NPV of an
option (just to show that it links right etc pp) but abort on implied vol.

CHROOT edd@chibud:~/src/progs/C++$ g++ -Wall -o test_ql_impliedvol
test_ql_impliedvol.cc -lQuantLib
CHROOT edd@chibud:~/src/progs/C++$ ./test_ql_impliedvol
Value:  5.81756
Aborted

Here is the short program:


// implied vol in RQuantLib-0.1.7 blows up with QL 0.3.4

#include <ql/quantlib.hpp>
#include <iostream>

using namespace QuantLib;
using QuantLib::Pricers::EuropeanOption;

int main(void) {
  double underlying = 100;
  double strike = 100;
  double dividendYield = 0.02;
  double riskFreeRate = 0.03;
  double maturity = 0.5; 
  double volatility = 0.2;
  Option::Type optionType = Option::Call;

  EuropeanOption EO = EuropeanOption(optionType, underlying, strike,
				     dividendYield, riskFreeRate, maturity, 
				     volatility);
  std::cout << "Value:  " << EO.value() << std::endl;
  double impliedVol = EO.impliedVolatility(EO.value()+0.1);
  std::cout << "Implied:" << impliedVol << std::endl;
  exit(0);
}


Thanks in advance for any insights.

Dirk


-- 
Those are my principles, and if you don't like them... well, I have others.
                                                -- Groucho Marx

Re: [Quantlib-dev] 0.3.4 releases

[Liguo S. <Lig...@va...>]

Hi, Nando,

>> I am not talking about the version number but the name of the tar ball,
>> such as the QuantLib-${VERSION}-src.tar.gz or
>> QuantLib-${VERSION}.tar.gz. I'd propose that we choose one of the names
>> and stick to it.
> 
> 
> I do agree. Luigi, what do you prefer for the makedist output? Currently 
> I guess it is QuantLib-${VERSION}.tar.gz and it's me that in the past 
> manually renamed it to QuantLib-${VERSION}-src.tar.gz
> Liguo, you picked up QuantLib-${VERSION}.src.rpm: is the .src more common?
> 
> just let me know and 0.3.5 will be compliant
I used QuantLib-${VERSION}-src.tar.gz because it is the name of the final 
release tar ball. And in QuantLib-${VERSION}.src.rpm, src is just saying this is 
the source package against the i386 - binary package.

As for the tar ball name, I do prefer QuantLib-${VERSION}-src.tar.gz as it 
emphasize on the fact that it is the tar ball of the source code. I have the 
impression that the tar ball without src in the name could be a binary tar ball 
without the source code.

So, my vote goes to QuantLib-${VERSION}-src.tar.gz for the source code tar ball.

Thanks.


Liguo (Leo)

Re: [Quantlib-dev] 0.3.4 releases

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

On 2003.11.25 18:18, Ferdinando Ametrano wrote:
>> I am not talking about the version number but the name of the tar  
>> ball,
>> such as the QuantLib-${VERSION}-src.tar.gz or
>> QuantLib-${VERSION}.tar.gz. I'd propose that we choose one of the  
>> names
>> and stick to it.
> 
> I do agree. Luigi, what do you prefer for the makedist output?

Well, it's not that I prefer one---"make dist" gives me
QuantLib-${VERSION}.tar.gz. If you want to rename it when you release  
it, it's ok. As the "master" copy of the sources, it makes kind of  
sense to me the way it is. But that's just me.

Later,
	Luigi

Re: [Quantlib-dev] 0.3.4 releases

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

>I just put the packages on the sourceforge ftp's incoming server again.
>They are also available at my website now. Please use whichever
>convenient for you. :)

done. They are on-line now. Sorry for being late, but ill was not planned :)

>I am not talking about the version number but the name of the tar ball,
>such as the QuantLib-${VERSION}-src.tar.gz or
>QuantLib-${VERSION}.tar.gz. I'd propose that we choose one of the names
>and stick to it.

I do agree. Luigi, what do you prefer for the makedist output? Currently I 
guess it is QuantLib-${VERSION}.tar.gz and it's me that in the past 
manually renamed it to QuantLib-${VERSION}-src.tar.gz
Liguo, you picked up QuantLib-${VERSION}.src.rpm: is the .src more common?

just let me know and 0.3.5 will be compliant

ciao -- Nando 

[Quantlib-dev] Templatization (was Re: MarketElement)

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

On 2003.11.24 19:16, Ferdinando Ametrano wrote:
> Getting back to an old question of yours about the templetization of  
> trees: I was wondering if that in your opinion would make more  
> difficult the merge of the Finite Difference framework and the  
> Lattice framework, which currently includes trees.
> I would say that the merging of Finite Difference and Lattice should  
> be the top priority design on that front, and would stay away from  
> anything that might make this process harder.
> What your take on this issue?

They're two sides of the same coin. Either we give them both a  
templated implementation, or none. Adding finite differences into the  
mix shouldn't make it any more difficult than it already is :)

Later,
	Luigi

[Quantlib-dev] MarketElement

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

Hi all,
	how about renaming MarketElement to Quote? It's shorter and  
more to the point (i.e., it is the actual name for the thing.)

Later,
	Luigi

RE: [Quantlib-dev] Trouble with QuantLib-Guile

[<lui...@fa...>]

>-- Original Message --
>From: Liguo Song <Lig...@va...>
>To: QuantLib-dev <qua...@li...>
>Subject: [Quantlib-dev] Trouble with QuantLib-Guile
>Date: Thu, 20 Nov 2003 13:16:06 -0600
>
>
>When I tried to do 'guile -s setup.scm test' or 'guile -s 
>quantlib-test-suite.scm' after the installation of QuantLib-Guile, I got=

the 
>following error:
>	ERROR: no such module (ice-9 greg)

> Do I need something else to get this work?

http://www.gnu.org/software/greg/greg.html

The above works out of the box with guile 1.4.x. If you're using guile 1.=
6.x
instead, there are a number of small corrections that need to be done in
order to bring some deprecated syntax up to date. If you're any comfortab=
le
with Scheme, you can replace instances of the empty list written as () wi=
th
'(). Otherwise, I can send you the patched version Monday (I'm not at wor=
k
today.)

Later,
    Luigi

[Quantlib-dev] Trouble with QuantLib-Guile

[Liguo S. <Lig...@va...>]

When I tried to do 'guile -s setup.scm test' or 'guile -s 
quantlib-test-suite.scm' after the installation of QuantLib-Guile, I got the 
following error:
	ERROR: no such module (ice-9 greg)

Do I need something else to get this work?

Thanks.


Liguo (Leo)

Re: [Quantlib-dev] 0.3.4 releases

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

On Thu, Nov 20, 2003 at 12:23:26PM +0100, Luigi Ballabio wrote:
> 
> >Consequently 0.3.4-2 packages are being uploaded now. As I  
> >'accidentally wasted' the 0.3.4 label on the previous golden master,  
> >I decided to for once forgo the rule of shipping pristine .orig.tar. 
> >gz files and stick with the previous one. The compressed diff is  
> >small.
> 
> Dirk,
> 	I'm not sure I follow. The latest tarball (now in gm/) contains  

I probably wasn't very coherent -- sorry!

> actual bug-fixes. Were those included in your upload?

Yes.

0.3.4-1: based on the then-current 'gm' QuantLib-0.3.4.tar.gz
	 (which we rename quantlib_0.3.4.orig.tar.gz)
	 released with a .diff.gz which is essentially the debian/ dir
	 
0.3.4-2: based on Nov 18 'gm' QuantLib-0.3.4.tar.gz
	 unpacked it into a fresh QuantLib-0.3.4/ directory
	 copied debian/ from the previous one
	 build debian packages against _previous quantlib_0.3.4.orig.tar.gz_
	 hence with a .diff.gz which is the difference with respect to the 
	 0.3.4-1 release, i.e. the upstream changes as well as debian/

What I should have done is naming the first one 0.3.3.cvs.20031104-1 instead
of 0.3.4-1 which would have left 0.3.4-1 for the actual release. 

I just checked, the autobuilders already created binaries for six or so more
architectures. 

Regards,  Dirk	 


-- 
Those are my principles, and if you don't like them... well, I have others.
                                                -- Groucho Marx

[Quantlib-dev] Re: 0.3.4 releases

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

On Tue, Nov 18, 2003 at 07:10:52PM +0100, Ferdinando Ametrano wrote:
> Luigi has provided today the latest QuantLib 0.3.4 C++ tarball at 
> http://quantlib.org/gm
> This version worked fine for the Win32 build.
> Dirk and Liguo: please go ahead with you packages. Barring last minute 
> problems with Debian/RPM I would like to announce 0.3.4 next Friday, 
> November 21th.

Disaster struck yesterday and took something down on my development box. Not
quite sure what -- upon I get 'crc error -- system halted' immediately after
the uncompressing kernel step. That took up some time last night. Anyway, I
am now booting from another partition, and the chroot(8) in which I build
the packages wasn't affected at all. 

Consequently 0.3.4-2 packages are being uploaded now. As I 'accidentally
wasted' the 0.3.4 label on the previous golden master, I decided to for once
forgo the rule of shipping pristine .orig.tar.gz files and stick with the
previous one. The compressed diff is small.  And I'll try to avoid that
naming trap next time. I think I'll do Python and Ruby tomorrow.

Thanks to everybody for the hard work on this.

Dirk

-- 
Those are my principles, and if you don't like them... well, I have others.
                                                -- Groucho Marx

Re: [Quantlib-dev] 0.3.4 releases

[Liguo S. <Lig...@va...>]

Hi, Nando,

The RPM for the main packages, including devel, test-suite, and docs, are OK. I 
just noticed that I named the documentation package as QuantLib-doc, while the 
tar ball is named QuantLib-docs. I corrected the discrepancy here, and added a 
item to ChangeLog.

Python package doesn't have the spec file. I attached it here, please put it in 
the CVS or the tar ball.

Same for Ruby package.

I am still learning my way around CVS, so I am not very confident to mess with 
the CVS. :)

As Guile package doesn't have the prefix option to install in 0.3.4, I am still 
looking for a hack to get it work. I will keep you posted on the progress.

As MzScheme package, there is not MzScheme rpm package available. So it might 
take much longer to get package it. I wouldn't hold my breath for the RPM 
package. Sorry.


That's about it for today.

Best regards.


Liguo (Leo)









Ferdinando Ametrano wrote:
> Hi all
> 
> Luigi has provided today the latest QuantLib 0.3.4 C++ tarball at 
> http://quantlib.org/gm
> This version worked fine for the Win32 build.
> Dirk and Liguo: please go ahead with you packages. Barring last minute 
> problems with Debian/RPM I would like to announce 0.3.4 next Friday, 
> November 21th.
> 
> Ruby and Guile modules are also OK for me, since I won't provide Win32 
> packages.
> 
> Tomorrow I will check Python and MzScheme and will (hopefully) give a 
> green light for Debian/RPM packages
> 
> thanks everybody
> 
> ------------
> ciao -- Nando
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: SF.net Giveback Program.
> Does SourceForge.net help you be more productive?  Does it
> help you create better code?  SHARE THE LOVE, and help us help
> YOU!  Click Here: http://sourceforge.net/donate/
> _______________________________________________
> Quantlib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Quantlib-dev] 0.3.4

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

Python and MzScheme are ok on the Win32 platform. Please go ahead with 
Debian/RPM

thank you

------------
ciao -- Nando 

[Quantlib-dev] 0.3.4 releases

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

Hi all

Luigi has provided today the latest QuantLib 0.3.4 C++ tarball at 
http://quantlib.org/gm
This version worked fine for the Win32 build.
Dirk and Liguo: please go ahead with you packages. Barring last minute 
problems with Debian/RPM I would like to announce 0.3.4 next Friday, 
November 21th.

Ruby and Guile modules are also OK for me, since I won't provide Win32 
packages.

Tomorrow I will check Python and MzScheme and will (hopefully) give a green 
light for Debian/RPM packages

thanks everybody

------------
ciao -- Nando 

[Quantlib-dev] Re: [Quantlib-users] Observable Python objects

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

On 2003.11.11 06:07, Andre Louw wrote:
> Luigi,
> 
> The Observer side works well (even using inheritance), the problem
> being that Python objects can only observe QuantLib C++ objects  
> (those defined as type 'isObservable' in their SWIG interface files).  
> I'm looking at the other side where a Python object becomes  
> observable by other Python objects.
> 
> Something in the line of
> 
> class Foo(Observable):
> 	def __init__(self):
> 		Observable.__init__(self)
> 	def changeMe(self):
> 		self.notifyObservers()

Oh, I see.

Hmm, I'll have to give it some thought. Can you file this as a feature  
request?

Later,
	Luigi

[Quantlib-dev] RE: [Quantlib-users] Observable Python objects

[Andre L. <An...@de...>]

Luigi,

I probably did not express myself properly. 

The Observer side works well (even using inheritance), the problem being
that Python objects can only observe QuantLib C++ objects (those defined as
type 'isObservable' in their SWIG interface files). I'm looking at the other
side where a Python object becomes observable by other Python objects.

Something in the line of

class Foo(Observable):
	def __init__(self):
		Observable.__init__(self)
	def changeMe(self):
		self.notifyObservers()

class FooObserver:
    def __init__(self):
        self.obs = QuantLib.Observer(self.update)
    def update(self):
        print "Happens when Foo changed"
    def registerWith(self,o):
        self.obs.registerWith(o)

f = Foo()
fo = FooObserver()
fo.registerWith(f)
f.changeMe() --> should result in an update in FooObserver...

Andre

Re: [Quantlib-dev] Release?

[<ni...@bi...>]

Hi Ferdinando and others

Sorry for the late reply.

I will deliver my master thesis in less than a month, and after that I'll be
happy to incorporate the code I have developed into the QuantLib CVS. Until
then I'll concentrate on making the report part of the thesis as good as
possible.

As part of the code, I have also made a least-squares Monte Carlo routine,
that I believe is better than the one in americanoption.cpp. It is split up
in a LSMEngine and a PolynomialFit object.

Anyway, I'll check in the code in the CVS, when I have an idle moment.

Ciao... Niels

----- Original Message -----
From: "Ferdinando Ametrano" <na...@am...>
To: <lui...@fa...>
Cc: <ma...@ni...>; "QuantLib-dev" <qua...@li...>
Sent: Friday, October 31, 2003 6:20 PM
Subject: Re: [Quantlib-dev] Release?


Luigi:
>         this looks to me like a good time to branch out 0.3.4---a few
>bugs were fixed (most notably, the library works again with Borland)
>and a few features added.

I would have liked to ask Niels Elken Sønderbys (in CC here) if he would
merge his
SVJD code (http://www.nielses.dk/quantlib/nesquant/) into QuantLib before
the next release. I don't remember if he is already a quantlib-dev
subscriber: if I didn't invite you before Niels, please accept my apologies
and consider this an open invitation.
To add the code should be easy enough, but it would make sense to add a
unit test too.
Of course we might just add SVJD for 0.3.5, especially if the merge+test
take longer than 1 week.
BTW Niels: please consider to actively work on the current CVS, I grant you
we will make it clear what your (master thesis) work consists ;-)

Another thing I would like to have fixed is the broken Borland
documentation. Luigi: I'm sure it will take few minutes of your time ;-)

A 0.3.4 related issue: I would like to remove my personal address
fer...@am... (which I'm going to suspend for a few months due to
excessive spam > 300 messages per day) from the "You should have received a
copy of the license along with this program; if not, please email
fer...@am..." file header. Suggestion about which address could
be used instead? What about qua...@li... ?

>  The current code would nicely qualify as a
>semi-stable, bug-fix release which people could rely on while we go
>ahead and break the world (or start using boost, or flatten namespaces,
>or generally start some new development.)

since you were planning to be backward compatible for one release on
namespace flattening, why don't we do it for 0.3.4?

My 0.2$

ciao -- Nando

[Quantlib-dev] Re: [Quantlib-users] Observable Python objects

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

On 2003.11.10 17:23, Andre Louw wrote:
> I have been trying to use (abuse?) the observer swig interface to get
> my Python objects to be 'Observable', not so simple! In fact, no  
> luck!

Andre',
	I'd go for containment instead of inheritance, as the latter  
does not happily cross the Python/C boundaries. Defining a Python
class like:

import QuantLib

class MyObserver:
    def __init__(self):
        self.obs = QuantLib.Observer(self.update)
    def update(self):
        print "Override me!"
    def registerWith(self,o):
        self.obs.registerWith(o)

and inheriting from MyObserver should work, but beware---I didn't test  
it. As a matter of fact, I didn't even tried to run it through the  
Python interpreter...

Later,
	Luigi

[Quantlib-dev] Observable Python objects

[Andre L. <An...@de...>]

I have been trying to use (abuse?) the observer swig interface to get my
Python objects to be 'Observable', not so simple! In fact, no luck!

I thought it would be as simple as defining an Observable class of type
'IsObservable' (all this in the SWIG interface file). The Python class would
then inherit from it, initialize itself and the Observable class and VOILA!
But alas, after much ranting and raving no luck?

Any suggestions?

Re: [Quantlib-dev] Re: Debian test-suite

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

On Fri, Nov 07, 2003 at 04:20:25PM +0100, Luigi Ballabio wrote:
> On 2003.11.07 15:58, Dirk Eddelbuettel wrote:
> >On Fri, Nov 07, 2003 at 03:41:07PM +0100, Ferdinando Ametrano wrote:
> >> I took a look at the Debian build logs and noticed that the test- 
> >suite is not executed, because the library cannot be linked. Luigi  
> >confirmed me that this is a problem he also has on any box which  
> >doesn't have a previous QuantLib installed.
> >
> >> Is there a way to have the C++ test-suite executed on all those
> >Debian platform?
> >
> >It generally works. AFAIK this is my only where package where I  
> >invoke it... and nothing happens. It must be some LD_LIBRARY /  
> >libtool / ... magic that is beyond. If Luigi can fix it, great. If  
> >not, well, .... I guess it won;t be fixed.  I do not suspect a  
> >problem at Debian's end.
> 
> I don't suspect it either---it should be some libtool glitch (which I  
> fix on my box by running ldconfig, but one has to be root, hasn't one?)

Both manual and autobuilders do run as root -- commonly via the fakeroot
utility -- as they e.g. have to create files with root.root ownership etc pp.

That wouldn't be a constraint.

> I guess it stays this way for this release.

Yep.  Dirk

> 
> Later,
> 	Luigi
> 

-- 
Those are my principles, and if you don't like them... well, I have others.
                                                -- Groucho Marx

Re: [Quantlib-dev] QuantLib-SWIG

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

At 04:22 PM 11/7/2003, Luigi Ballabio wrote:
>>QuantLib-Ruby doesn't even build.
>
>Can you elaborate?

the same problem I reported for 0.3.3
See the output below.

ciao -- Nando

==========================


C:\QuantLib-Ruby-0.3.4>ruby setup.rb build
Building extension...
creating Makefile

Microsoft (R) Program Maintenance Utility   Version 6.00.8168.0
Copyright (C) Microsoft Corp 1988-1998. All rights reserved.

         cl -MD -Zi -O2b2xg- -G6 -I. 
-Ic:/programs/ruby/lib/ruby/1.8/i386-mswin32
  -Ic:/programs/ruby/lib/ruby/1.8/i386-mswin32 -I.  -I. -I./.. 
-I./../missing /MD
  /GR /GX /DNOMINMAX /IC:\Programs\QuantLib -c -Tpquantlib_wrap.cpp
Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 12.00.8804 for 80x86
Copyright (C) Microsoft Corp 1984-1998. All rights reserved.

quantlib_wrap.cpp
C:\Programs\QuantLib\ql/cashflow.hpp(56) : warning C4003: not enough actual 
para
meters for macro 'accept'
C:\Programs\QuantLib\ql/cashflow.hpp(56) : error C2059: syntax error : ','
C:\Programs\QuantLib\ql/cashflow.hpp(63) : warning C4003: not enough actual 
para
meters for macro 'accept'
C:\Programs\QuantLib\ql/cashflow.hpp(63) : error C2059: syntax error : ','
C:\Programs\QuantLib\ql/voltermstructure.hpp(100) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(100) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(126) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(126) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(149) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(149) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(189) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(189) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(201) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(201) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(212) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(212) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(219) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(219) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(223) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(223) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(230) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(230) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/voltermstructure.hpp(233) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/voltermstructure.hpp(233) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/CashFlows/coupon.hpp(71) : warning C4003: not 
enough act
ual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/coupon.hpp(71) : error C2059: syntax 
error : '
,'
C:\Programs\QuantLib\ql/CashFlows/coupon.hpp(118) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/coupon.hpp(118) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/CashFlows/coupon.hpp(124) : warning C4003: not 
enough ac
tual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/coupon.hpp(124) : error C2059: syntax error :
','
C:\Programs\QuantLib\ql/CashFlows/fixedratecoupon.hpp(59) : warning C4003: 
not e
nough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/fixedratecoupon.hpp(59) : error C2059: syntax
error : ','
C:\Programs\QuantLib\ql/CashFlows/fixedratecoupon.hpp(93) : warning C4003: 
not e
nough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/fixedratecoupon.hpp(93) : error C2059: syntax
error : ','
C:\Programs\QuantLib\ql/CashFlows/fixedratecoupon.hpp(100) : warning C4003: not
enough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/fixedratecoupon.hpp(100) : error C2059: 
syntax
  error : ','
C:\Programs\QuantLib\ql/CashFlows/floatingratecoupon.hpp(56) : warning 
C4003: no
t enough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/floatingratecoupon.hpp(56) : error C2059: 
synt
ax error : ','
C:\Programs\QuantLib\ql/CashFlows/floatingratecoupon.hpp(95) : warning 
C4003: no
t enough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/floatingratecoupon.hpp(95) : error C2059: 
synt
ax error : ','
C:\Programs\QuantLib\ql/CashFlows/floatingratecoupon.hpp(102) : warning 
C4003: n
ot enough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/floatingratecoupon.hpp(102) : error 
C2059: syn
tax error : ','
C:\Programs\QuantLib\ql/CashFlows/indexedcoupon.hpp(72) : warning C4003: 
not eno
ugh actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/indexedcoupon.hpp(72) : error C2059: 
syntax er
ror : ','
C:\Programs\QuantLib\ql/CashFlows/indexedcoupon.hpp(119) : warning C4003: 
not en
ough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/indexedcoupon.hpp(119) : error C2059: 
syntax e
rror : ','
C:\Programs\QuantLib\ql/CashFlows/indexedcoupon.hpp(126) : warning C4003: 
not en
ough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/indexedcoupon.hpp(126) : error C2059: 
syntax e
rror : ','
C:\Programs\QuantLib\ql/CashFlows/inarrearindexedcoupon.hpp(56) : warning 
C4003:
  not enough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/inarrearindexedcoupon.hpp(56) : error 
C2059: s
yntax error : ','
C:\Programs\QuantLib\ql/CashFlows/inarrearindexedcoupon.hpp(71) : warning 
C4003:
  not enough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/inarrearindexedcoupon.hpp(71) : error 
C2059: s
yntax error : ','
C:\Programs\QuantLib\ql/CashFlows/inarrearindexedcoupon.hpp(78) : warning 
C4003:
  not enough actual parameters for macro 'accept'
C:\Programs\QuantLib\ql/CashFlows/inarrearindexedcoupon.hpp(78) : error 
C2059: s
yntax error : ','
C:\Programs\QuantLib\ql/CashFlows/shortindexedcoupon.hpp(41) : error C2653: 
'Ind
exes' : is not a class or namespace name
         C:\Programs\QuantLib\ql/CashFlows/shortindexedcoupon.hpp(62) : see 
refer
ence to class template instantiation 'CashFlows::Short<IndexedCouponType>' 
being
  compiled
C:\Programs\QuantLib\ql/CashFlows/shortindexedcoupon.hpp(39) : error C2629: 
unex
pected 'class CashFlows::Short<IndexedCouponType> ('
         C:\Programs\QuantLib\ql/CashFlows/shortindexedcoupon.hpp(62) : see 
refer
ence to class template instantiation 'CashFlows::Short<IndexedCouponType>' 
being
  compiled
C:\Programs\QuantLib\ql/CashFlows/shortindexedcoupon.hpp(39) : error C2334: 
unex
pected token(s) preceding ':'; skipping apparent function body
         C:\Programs\QuantLib\ql/CashFlows/shortindexedcoupon.hpp(62) : see 
refer
ence to class template instantiation 'CashFlows::Short<IndexedCouponType>' 
being
  compiled
C:\Programs\QuantLib\ql/CashFlows/shortindexedcoupon.hpp(66) : fatal error 
C1506
: unrecoverable block scoping error
NMAKE : fatal error U1077: 'cl' : return code '0x2'
Stop. 

Re: [Quantlib-dev] QuantLib-SWIG

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

On 2003.11.07 16:06, Ferdinando Ametrano wrote:
> QuantLib-MzScheme succesfully builds, tests itself, and install on  
> Win32
> Unfortunately 'mzscheme -r setup.scm sdist' fails because 'couldn't  
> make stderr distinct from stdout'

I can't make head nor tail of this one. Oh well, just convert my  
tarball to zip format.

> QuantLib-Ruby doesn't even build.

Can you elaborate?

Later,
	Luigi

Re: [Quantlib-dev] Re: Debian test-suite

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

On 2003.11.07 15:58, Dirk Eddelbuettel wrote:
> On Fri, Nov 07, 2003 at 03:41:07PM +0100, Ferdinando Ametrano wrote:
> > I took a look at the Debian build logs and noticed that the test- 
> suite is not executed, because the library cannot be linked. Luigi  
> confirmed me that this is a problem he also has on any box which  
> doesn't have a previous QuantLib installed.
> 
> > Is there a way to have the C++ test-suite executed on all those
> Debian platform?
> 
> It generally works. AFAIK this is my only where package where I  
> invoke it... and nothing happens. It must be some LD_LIBRARY /  
> libtool / ... magic that is beyond. If Luigi can fix it, great. If  
> not, well, .... I guess it won;t be fixed.  I do not suspect a  
> problem at Debian's end.

I don't suspect it either---it should be some libtool glitch (which I  
fix on my box by running ldconfig, but one has to be root, hasn't one?)

I guess it stays this way for this release.

Later,
	Luigi

[Quantlib-dev] QuantLib-SWIG

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

QuantLib-MzScheme succesfully builds, tests itself, and install on Win32
Unfortunately 'mzscheme -r setup.scm sdist' fails because 'couldn't make 
stderr distinct from stdout'

QuantLib-Ruby doesn't even build.

QuantLib-Python has no problems at all.

ciao -- Nando

[Quantlib-dev] Re: Debian test-suite

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

On Fri, Nov 07, 2003 at 03:41:07PM +0100, Ferdinando Ametrano wrote:
> Dirk,
> 
> I took a look at the Debian build logs and noticed that the test-suite is 
> not executed, because the library cannot be linked. Luigi confirmed me that 
> this is a problem he also has on any box which doesn't have a previous 
> QuantLib installed.
> 
> This was not a big issue for 0.3.3 since the test-suite was executed by 
> QuantLib-Python. Unfortunately for 0.3.4 all QuantLib-SWIG 0.3.4 packages 
> do not include a full test-suite anymore, but only test their own 
> language-specific features, relying on the QuantLib C++ test suite for the 
> base library features.

Right. I just enabled that in ql-ruby, and saw that the tests where quick.

> Is there a way to have the C++ test-suite executed on all those Debian 
> platform?

It generally works. AFAIK this is my only where package where I invoke it
... and nothing happens. It must be some LD_LIBRARY / libtool / ... magic
that is beyond. If Luigi can fix it, great. If not, well, .... I guess it
won;t be fixed.  I do not suspect a problem at Debian's end.

As a sledgehammer solution, I could always purge prior or other QL
installations in the chroot in which I am building the packages. But that is
somewhat brute ...

Dirk

-- 
Those are my principles, and if you don't like them... well, I have others.
                                                -- Groucho Marx