[Rcpp-devel] Fwd: Extending <Rcpp/module/Module_generated_class_constructor.h>

Oops, I replied to the wrong email...

Tama Ma

Begin forwarded message:
From: Tama Ma <pingnang at phys.ethz.ch>
Date: December 3, 2010 10:04:51 PM GMT+01:00
To: Romain Francois <romain at r-enthusiasts.com>
Cc: rcpp-devel-request at lists.r-forge.r-project.org
Subject: Re: [Rcpp-devel] Extending	<Rcpp/module/Module_generated_class_constructor.h>

Dear Romain,

	It happens that <Module_generated_ctor_signature.h> has to be modified as well. Hmm, this tricked me off somehow by not showing up as a compilation error.
	Anyway, this is the corrected patch.

	Thank you very much.

Best regards,
Tama Ma

On Dec 3, 2010, at 9:02 PM, Romain Francois wrote:

Sure. The pattern is quite easy to extend.

You need to update this file and also Module_generated_Constructor.h

send us a patch against the current svn tree.

$ svn checkout svn://svn.r-forge.r-project.org/svnroot/rcpp
$ # make the changes
$ svn diff > patch.txt

And then you send the patch.txt file.

We look forward for the patch.

Romain

Le 03/12/10 20:55, Tama Ma a ?crit :
Dear Rcpp developers,

	I want to export a class with a constructor with 7 argments. However, in<Rcpp/module/Module_generated_class_constructor.h>, we have only support up to 6 arguments, eg.

   template<
       typename U0,
       typename U1,
       typename U2,
       typename U3,
       typename U4,
       typename U5

   self&  constructor( const char* docstring = 0, ValidConstructor valid =&yes_arity<6>  ){
       AddConstructor( new Constructor_6<Class,U0,U1,U2,U3,U4,U5>  , valid, docstring ) ;
       return *this ;
   }

	Can this be extended? And in what way I can help with it?

	Thank you very much.

Best regards,
Tama Ma

ETH Zurich
Tama Ma
Instit?t f?r Theoretische Physik
HIT K 31.3
Wolfgang-Pauli-Strasse 27
ETH H?nggerberg
8093 Z?rich
Switzerland

pingnang at itp.phys.ethz.ch
www.phys.ethz.ch/~pingnang

+41 44 633 76 55  (Office Phone)
+41 44 633 11 15  (Office Fax)

-- 
Romain Francois
Professional R Enthusiast
+33(0) 6 28 91 30 30
http://romainfrancois.blog.free.fr
|- http://bit.ly/gpCSpH : Evolution of Rcpp code size
|- http://bit.ly/hovakS : RcppGSL initial release
`- http://bit.ly/iaxTdO : parser 0.0-12

_______________________________________________
Rcpp-devel mailing list
Rcpp-devel at lists.r-forge.r-project.org
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rcpp-devel/attachments/20101203/b569f27b/attachment-0002.htm>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: patch-r2695.txt
URL: <http://lists.r-forge.r-project.org/pipermail/rcpp-devel/attachments/20101203/b569f27b/attachment-0001.txt>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rcpp-devel/attachments/20101203/b569f27b/attachment-0003.htm>
Tama,
| Oops, I replied to the wrong email...
| 
| Tama Ma
| 
| 
| Begin forwarded message:
| 
| 
|     From: Tama Ma <pingnang at phys.ethz.ch>
|     Date: December 3, 2010 10:04:51 PM GMT+01:00
|     To: Romain Francois <romain at r-enthusiasts.com>
|     Cc: rcpp-devel-request at lists.r-forge.r-project.org
|     Subject: Re: [Rcpp-devel] Extending <Rcpp/module/
|     Module_generated_class_constructor.h>
| 
|     Dear Romain,
| 
|     It happens that <Module_generated_ctor_signature.h> has to be modified as
|     well. Hmm, this tricked me off somehow by not showing up as a compilation
|     error.
|     Anyway, this is the corrected patch.

I still get a seg fault in 'R CMD check' for current SVN plus your two
patches.  Please do run 'R CMD check'.

Thanks, Dirk
Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com
Tama,

BTW r2697 now has both your patches so feel free to work from those.

Dirk
Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com
Tama,
| I still get a seg fault in 'R CMD check' for current SVN plus your two
| patches.  Please do run 'R CMD check'.

Sorry -- that was a combination of my being one checkout behind on the inline
package and what looks like an actual typo in one of the unit tests.  

So no error of yours, and no worries.  Thanks again for the contribution.

Cheers, Dirk
Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com
Dear Dirk,

	Thank you very much for updating the svn so fast. Well, I must admit that I only started to learn R and Rcpp yesterday, so I am very new to this business. 

	To r2697, I first do a  cleanup  followed by  R CMD INSTALL Rcpp -I /path/to/my/Rcpp/directory  . It runs okay, and my own program compiles as well. 

	However, following what you have suggested, I find an error as well, given as follows:

Tama-Mas-MacBook-Pro:pkg tamama$ pwd
/opt/local/share/Rcpp-svn/pkg
Tama-Mas-MacBook-Pro:pkg tamama$ sudo R CMD check Rcpp
* using log directory ?/opt/local/share/Rcpp-svn/pkg/Rcpp.Rcheck?
* using R version 2.12.0 (2010-10-15)
* using platform: x86_64-apple-darwin10.5.0 (64-bit)
* using session charset: UTF-8
* checking for file ?Rcpp/DESCRIPTION? ... OK
* this is package ?Rcpp? version ?0.8.9.3?
* checking package name space information ... OK
* checking package dependencies ... ERROR
Packages required but not available:
  RUnit inline rbenchmark

See the information on DESCRIPTION files in the chapter 'Creating R
packages' of the 'Writing R Extensions' manual.
Tama-Mas-MacBook-Pro:pkg tamama$ 

	I don't know what this RUnit is, anyway I will try to locate where the possible error is.

	Thank you very much.

Best regards,
Tama Ma

Tama,

BTW r2697 now has both your patches so feel free to work from those.

Dirk

-- 
Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com

Dear Dirk,

	Good to hear this!

Best regards,
Tama Ma

Tama,

On 3 December 2010 at 15:16, Dirk Eddelbuettel wrote:
| I still get a seg fault in 'R CMD check' for current SVN plus your two
| patches.  Please do run 'R CMD check'.

Sorry -- that was a combination of my being one checkout behind on the inline
package and what looks like an actual typo in one of the unit tests.  

So no error of yours, and no worries.  Thanks again for the contribution.

Cheers, Dirk

-- 
Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com

Hi Tama,
| Dear Dirk,
| 
| 	Thank you very much for updating the svn so fast. Well, I must admit that I only started to learn R and Rcpp yesterday, so I am very new to this business. 
| 
| 	To r2697, I first do a  cleanup  followed by  R CMD INSTALL Rcpp -I /path/to/my/Rcpp/directory  . It runs okay, and my own program compiles as well. 
| 
| 	However, following what you have suggested, I find an error as well, given as follows:
| 
| Tama-Mas-MacBook-Pro:pkg tamama$ pwd
| /opt/local/share/Rcpp-svn/pkg
| Tama-Mas-MacBook-Pro:pkg tamama$ sudo R CMD check Rcpp
| * using log directory ?/opt/local/share/Rcpp-svn/pkg/Rcpp.Rcheck?
| * using R version 2.12.0 (2010-10-15)
| * using platform: x86_64-apple-darwin10.5.0 (64-bit)
| * using session charset: UTF-8
| * checking for file ?Rcpp/DESCRIPTION? ... OK
| * this is package ?Rcpp? version ?0.8.9.3?
| * checking package name space information ... OK
| * checking package dependencies ... ERROR
| Packages required but not available:
|   RUnit inline rbenchmark

Yes, R tells you that you'd need packages RUnit, inline and rbenchmark from
CRAN. Feel free to install them and try again -- but as I found out the hard
way you also need rev41 of inline from its SVN on R-Forge.

But with that and rev2699 of Rcpp, you should be good and see that your patch
applied cleanly :)

Cheers, Dirk
Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com