Mingw/Cygwin build success ?

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|

Mingw/Cygwin build success ?

Dirk Eddelbuettel

In order to build my RQuantLib binding on Windows, I need a suitable Windows
build of QuantLib -- preferably using MinGW.  

Has anybody succeeded in doing this?  I got the same failure with the 0.9.7
pre-release and the 0.9.7 release that some some later-stage link attempt
fail. I don't have the exact error message handy but could re-create it. I
did check the older HOWTOs but no luck so far.

Hints would be appreciated to make this build.

Regards, Dirk

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

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Re: Mingw/Cygwin build success ?

Luigi Ballabio

On Nov 23, 2008, at 5:38 PM, Dirk Eddelbuettel wrote:

> In order to build my RQuantLib binding on Windows, I need a suitable  
> Windows
> build of QuantLib -- preferably using MinGW.
>
> Has anybody succeeded in doing this?  I got the same failure with  
> the 0.9.7
> pre-release and the 0.9.7 release that some some later-stage link  
> attempt
> fail. I don't have the exact error message handy but could re-create  
> it. I
> did check the older HOWTOs but no luck so far.

I didn't try MinGW, but I succeeded with g++ on Windows using Dev-C++.  
Would that suit you?

Luigi


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Mingw/Cygwin build success ?

R P Herrold
In reply to this post by Dirk Eddelbuettel
On Sun, 23 Nov 2008, Dirk Eddelbuettel wrote:

> In order to build my RQuantLib binding on Windows, I need a
> suitable Windows build of QuantLib -- preferably using
> MinGW.

We also need and have on our docket exploring the MinGW build
issues, as Cygwin has lagged, for the trading-shim project

The Emerging Technologies folks at Red Hat have been working
on pushing MinGW into RH's Fedora developmental line

[ It is somewhat curious, in that the Cygnus folks (an early
purchase by RHT) are not at the tip of this spear, but as I
lurk on the Cygwin mailing list, I see much of a desire to get
closure on new releases; I have a need for getaddrinto() for 6
month's now in Cugnux but they do not yet support IPv6 and
fairly standard things on Linux ]

It may be productive to ping the following person, and run
down these links.

-- Russ herrold


Date: Fri, 5 Sep 2008 05:30:31
From: Richard W.M. Jones <[hidden email]>
Reply-To: Development discussions related to Fedora
     <[hidden email]>
To: [hidden email]
Subject: fedora-d-rh] MinGW on Fedora - an update

Dan Berrange (in particular) and myself (a little) have done
quite a bit more work on mingw in Fedora.

We now have:
  - automatic dependency generation
  - automatic stripping of binaries
  - a collection of RPM macros which simplify writing spec files
  - ~10 working libraries

Anyway, check out the development repository:
  http://hg.et.redhat.com/misc/fedora-mingw--devel/

Please read the README file first!

Rich.

--
Richard Jones, Emerging Technologies, Red Hat  http://et.redhat.com/~rjones
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://et.redhat.com/~rjones/virt-top

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

and this trailhead bug:

https://bugzilla.redhat.com/show_bug.cgi?id=454414

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

there is a mailing list as well, but I don ont find that it is
very active.


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Re: Mingw/Cygwin build success ?

Dirk Eddelbuettel
In reply to this post by Luigi Ballabio

On 23 November 2008 at 18:10, Luigi Ballabio wrote:
|
| On Nov 23, 2008, at 5:38 PM, Dirk Eddelbuettel wrote:
| > In order to build my RQuantLib binding on Windows, I need a suitable  
| > Windows
| > build of QuantLib -- preferably using MinGW.
| >
| > Has anybody succeeded in doing this?  I got the same failure with  
| > the 0.9.7
| > pre-release and the 0.9.7 release that some some later-stage link  
| > attempt
| > fail. I don't have the exact error message handy but could re-create  
| > it. I
| > did check the older HOWTOs but no luck so far.
|
| I didn't try MinGW, but I succeeded with g++ on Windows using Dev-C++.  
| Would that suit you?

I think so! I do use g++, but 'have to' use the MinGW variant as that is what
R uses to build a Windows-native beast.  I tried very basic configure
options, and a plain make. It fails in libtool-driven linking stage not
finding library files.  

Did you use a DOS command prompt window?  I am running in Msys/bash/... to
keep the shell behaviour close to my preferred habitat.

Easiest for me would be if you supplied such a lib via sf.net :)

Dirk

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

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Re: Mingw/Cygwin build success ?

Dirk Eddelbuettel
In reply to this post by R P Herrold

On 23 November 2008 at 12:56, R P Herrold wrote:
| On Sun, 23 Nov 2008, Dirk Eddelbuettel wrote:
|
| > In order to build my RQuantLib binding on Windows, I need a
| > suitable Windows build of QuantLib -- preferably using
| > MinGW.
|
| We also need and have on our docket exploring the MinGW build
| issues, as Cygwin has lagged, for the trading-shim project
|
| The Emerging Technologies folks at Red Hat have been working
| on pushing MinGW into RH's Fedora developmental line

Ah, right. I tend to forget that I have those right at my fingertips under
Debian (and Ubuntu at work):

edd@joe:~$ wajig search mingw
mingw32 - Minimalist GNU win32 (cross) compiler
mingw32-binutils - Minimalist GNU win32 (cross) binutils
mingw32-runtime - Minimalist GNU win32 (cross) runtime
edd@joe:~$

Thanks for the reminder re cross-builds. This could be useful for other
projects too. And I used this without too much trouble some years ago.

Now, you didn't say that you managed to build QL with this, did you?

Cheers, Dirk

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

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Mingw/Cygwin build success ?

R P Herrold
On Sun, 23 Nov 2008, Dirk Eddelbuettel wrote:

> | We also need and have on our docket exploring the MinGW build
> | issues, as Cygwin has lagged, for the trading-shim project

> Thanks for the reminder re cross-builds. This could be useful for other
> projects too. And I used this without too much trouble some years ago.

good to know  ;)

> Now, you didn't say that you managed to build QL with this, did you?

hmmm ... 25M of archives to download for mingw series, and
119M of space post install on my Debian Testing

not yet for QL, but I'll solve the shim, and then move on to
the other tools.  ;)  We had the shim going except for
getaddrinfo() and one other long long library some months ago,
which we 'ifdef'd' around

Now ... my Windows 2000 box's hard drive rolled over and died
some months ago.  I'll have to build anohter to test.  ;)

-- Russ herrold

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Re: Mingw/Cygwin build success ?

Dirk Eddelbuettel
In reply to this post by Dirk Eddelbuettel

On 23 November 2008 at 12:49, Dirk Eddelbuettel wrote:
| On 23 November 2008 at 18:10, Luigi Ballabio wrote:
| | On Nov 23, 2008, at 5:38 PM, Dirk Eddelbuettel wrote:
| | > In order to build my RQuantLib binding on Windows, I need a suitable  
| | > Windows
| | > build of QuantLib -- preferably using MinGW.
| |
| | I didn't try MinGW, but I succeeded with g++ on Windows using Dev-C++.  
| | Would that suit you?

That worked like a charm, thanks for the suggestion.

Dirk

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

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Re: Mingw/Cygwin build success ?

Luigi Ballabio
On Mon, 2008-11-24 at 12:29 -0600, Dirk Eddelbuettel wrote:

> On 23 November 2008 at 12:49, Dirk Eddelbuettel wrote:
> | On 23 November 2008 at 18:10, Luigi Ballabio wrote:
> | | On Nov 23, 2008, at 5:38 PM, Dirk Eddelbuettel wrote:
> | | > In order to build my RQuantLib binding on Windows, I need a suitable  
> | | > Windows
> | | > build of QuantLib -- preferably using MinGW.
> | |
> | | I didn't try MinGW, but I succeeded with g++ on Windows using Dev-C++.  
> | | Would that suit you?
>
> That worked like a charm, thanks for the suggestion.

Good. Also, it's good to know there's an alternative to the MinGW
build...

Luigi


--

The rule on staying alive as a forecaster is to give 'em a number or
give 'em a date, but never give 'em both at once.
-- Jane Bryant Quinn



-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Bond Pricing: root not bracketed

Eduardo Montoya
Hello everyone,

I am reading the Bonds.cpp example and I am trying to understand how does it work.
Could anyone give me any idea about why I am getting the following error?

root not bracketed: f[0,1] -> [nan,nan]

I would like to know if it is a runtime error or an well known application error.

thanks in advance,

Eduardo

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Reply | Threaded
Open this post in threaded view
|

Re: Bond Pricing: root not bracketed

Florent Grenier
Hi Eduardo,

This message is sent by the solver (used for the bootstrapping of the curve) when it cannot find any solution. It usually happens when there are some important jumps between two quotes (but the 'nan' values are quite weird by the way), the error message being like this one:
1st iteration: could not bootstrap the 2nd instrument, maturity March 18th, 2009: root not bracketed: f[2.22045e-16,0.975447] -> [-9.081845e+15,-3.625890e-02]

Does that error happen when executing the sample that is provided within the Quantlib package? Or did you modify it? Which version of Quantlib are you using?

Another thing, if you have some bootstrapping problems on bonds, you should consider using the FittedBondDiscountCurve class (see FittedBondCurve sample) instead, which provides more advanced fitters than the PiecewiseYieldCurve

Best regards,
Florent

2008/11/27 Eduardo Montoya <[hidden email]>
Hello everyone,

I am reading the Bonds.cpp example and I am trying to understand how does it work.
Could anyone give me any idea about why I am getting the following error?

root not bracketed: f[0,1] -> [nan,nan]

I would like to know if it is a runtime error or an well known application error.

thanks in advance,

Eduardo

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users



-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users