[R] RQuantlib ?Windows Binary?

Dirk Eddelbuettel edd at debian.org
Mon May 24 23:22:18 CEST 2004


Jason,

On Fri, May 21, 2004 at 11:48:16AM -0500, Jason.L.Higbee at stls.frb.org wrote:
> R:
> 
> Is there a reason why there isn't a Windows Binary version of RQuantlib on 
> CRAN?  Usually when there is no binary, I just source the source code, but

Yes, there is. 

On windows, QuantLib wants to be built with Visual C++ whereas R wants to be
built with MinGW.  When I first published RQuantLib, I thought it to be
impossible to be built on Windows ... only to receive an email from Tsvetan
Stoyanov who detailed how to build both under win2000.  I did try that at
one point an succeeded. The recipe is still on my page for RQuantLib at 

    http://dirk.eddelbuettel.com/code/rquantlib.html
    
The recipe may well need updating.  If you try it, would you mind sharing
your experience, maybe in a private email, so that I can update the page?

> this one appears to have various calls and methods and things like that so 
> I'm hesitant to do so.  I know there has been a big discussion on why 
> Rmetrics doesn't have source for unix/linux, but that isn't on CRAN. 
> Through that Rmetrics thread, I think I read that all CRAN packages have 
> source (except for one), but shouldn't all CRAN packages also have binary? 
>  Think they should.

Not that easy as Uwe mentioned.

Hope this helps,  Dirk

-- 
The relationship between the computed price and reality is as yet unknown.  
                                             -- From the pac(8) manual page




More information about the R-help mailing list