[Rcpp-devel] Parameter Naming (was: Speed gain assistance)

[Dirk Eddelbuettel]

On 18 August 2011 at 08:44, Dirk Eddelbuettel wrote:
| 
| On 18 August 2011 at 16:24, Darren Cook wrote:
| | I'd like to shamelessly hijack Christopher Wray's question to ask if
| | there is a convention for parameter naming with Rcpp and inline. Well,
| | actually I'll propose one, to then see what others prefer.
| | 
| | First I would propose to use long variable names, instead of
| | abbreviations, in both the CPP and R sides [1]. I already know enough R
| | to know this will be immediately rejected ;-) [2]
| | 
| | Anyway, while trying to understand the posted code I found myself
| | jumping back and forth between the below two blocks of code, muttering
| | to myself, "so the bottleneck is proportional to 2nd parameter which is
| | SamSize, which is then called SS_C, ..."
| | 
| | > src<-'
| | > NumericVector RetIn(RetVec);
| | > int SS_C=as<int>(SamSize); 
| | > double St_C=as<double>(StM);
| | > int ES_C=as<int>(ExpSize);
| | > ...
| | > sscpp2<-cfunction(
| | > signature(RetVec="numericVector",SamSize="numeric",StM="numeric",ExpSize="numeric"),
| | > src,inc, Rcpp=TRUE,cppargs="",convention=".Call")
| | 
| | So, my proposal would be to use the same names in the signature that you
| | intend to use in the C++ code, but with underline appended. I.e.
| 
| You must do that anyway or you get linking errors.

I was sloppy: compile-time errors. I.e. this:

  errorDemo <- cxxfunction(signature(badName="integer"), plugin='Rcpp', body='
     Rcpp::IntegerVector x(otherName);
     return 2*x;
  ')

gives you 

  Error in compileCode(f, code, language = language, verbose = verbose) : 
    Compilation ERROR, function(s)/method(s) not created! file181a5f5c.cpp: In function ‘SEXPREC* file181a5f5c(SEXPREC*)’:
  file181a5f5c.cpp:32:26: error: ‘otherName’ was not declared in this scope
  file181a5f5c.cpp:33:13: error: cannot convert ‘Rcpp::sugar::Times_Vector_Primitive<13, true, Rcpp::Vector<13> >’ to ‘SEXPREC*’ in return
  make: *** [file181a5f5c.o] Error 1
  In addition: Warning message:
  running command '/usr/lib64/R/bin/R CMD SHLIB file181a5f5c.cpp 2> file181a5f5c.cpp.err.txt' had status 1 
  R> 

So as I indicated, you have no choice but to make those match.

Dirk
 
| The rest is stylistic, and we do not impose a style (besides maybe the
| default indentation I get from Emacs :). If you need a style, I suggest to
| follow R conventions as layed out in the R Internals manual. Personally, I
| also like this one:
| 
|    http://quantlib.org/style.shtml
| 
| Descriptive names are good, comments are good, using separate names for
| members vars is good and I still mostly prepend m_. 
| 
| Dirk
| 
| 
| | signature(RetIn_="numericVector",SS_C_="numeric",St_C_="numeric",ES_C_="numeric")
| | 
| | The C++ code then starts:
| |   NumericVector RetIn(RetIn_);
| |   int SS_C=as<int>(SS_C_);
| |   double St_C=as<double>(St_C_);
| |   int ES_C=as<int>(ES_C);
| | 
| | What conventions are other people using?
| | 
| | Darren
| | 
| | [1]: I am not alone in this. E.g. see "Rename Method" in Martin Fowler's
| | Refactoring; "It's All Writing" in The Pragmatic Programmer (ch.8); etc.
| | 
| | [2]: I realize many R functions are direct translations of mathematical
| | functions, so calling a parameter "p" is not as meaningless as it
| | appears at first glance. I still think it should be "probabilities" not
| | "p" (yes, "qunif", I'm looking at you), but I realize I'm in a minority
| | when surrounded byq mathematicians.
| | 
| | 
| | 
| | 
| | -- 
| | Darren Cook, Software Researcher/Developer
| | 
| | http://dcook.org/work/ (About me and my work)
| | http://dcook.org/blogs.html (My blogs and articles)
| | _______________________________________________
| | 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
| 
| -- 
| Two new Rcpp master classes for R and C++ integration scheduled for 
| New York (Sep 24) and San Francisco (Oct 8), more details are at
| http://dirk.eddelbuettel.com/blog/2011/08/04#rcpp_classes_2011-09_and_2011-10
| http://www.revolutionanalytics.com/products/training/public/rcpp-master-class.php
| _______________________________________________
| 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

-- 
Two new Rcpp master classes for R and C++ integration scheduled for 
New York (Sep 24) and San Francisco (Oct 8), more details are at
http://dirk.eddelbuettel.com/blog/2011/08/04#rcpp_classes_2011-09_and_2011-10
http://www.revolutionanalytics.com/products/training/public/rcpp-master-class.php