[Rd] Generated HTML tags not all lowercase

2006-01-24 Thread Paul Roebuck
Noticed while fixing problem in my package documentation that some Rd code referencing an environment variable as \env{PATH} produced the HTML as PATH. "/share/perl/R/Rdconv.pm#785-800": shows there are others as well. Although R seems to currently target HTML 4.0 (or older) and not the stricter

Re: [Rd] unexported symbols in libR.so of r-devel

2006-01-24 Thread Prof Brian Ripley
On Tue, 24 Jan 2006, Na Li wrote: > > Thanks to Matthias Burger, I came to know that in r-devel, a lot of symbols > are no longer exported in libR.so, which breaks the package rpvm since it > calls these functions in serialize.c. They are hidden. Since they are in Rinternals.h, this was unintent

Re: [Rd] A patch for do_sample: check replace arg

2006-01-24 Thread Robert Gentleman
should be there now Seth Falcon wrote: > A colleague sent me the following: > > If you specify probabilities in the 'sample' function and forget > to type 'prob=...', then you get nonsense. E.g. > > sample(1:10,1,c(0,0,0,0,1,0,0,0,0,0)) > > does not filter '5', while >

[Rd] useR! 2006 - Submission Deadline 2006-02-28

2006-01-24 Thread Torsten Hothorn
The submission deadline for `useR! 2006', the second R user conference to be held in Vienna June 15-17 2006, is only four weeks ahead. Now is the perfect time to submit abstracts for user-contributed sessions! The sessions will be a platform to bring together R users, contributers, package maint

[Rd] unexported symbols in libR.so of r-devel

2006-01-24 Thread Na Li
Thanks to Matthias Burger, I came to know that in r-devel, a lot of symbols are no longer exported in libR.so, which breaks the package rpvm since it calls these functions in serialize.c. R_InitInPStream R_InitOutPStream Since the change is fairly recent (earlier this month), I guess the core te

Re: [Rd] Master's project to coerce linux nvidia drivers to run generalised linear models

2006-01-24 Thread Gabor Grothendieck
On 1/24/06, Douglas Bates <[EMAIL PROTECTED]> wrote: > On 1/23/06, Gabor Grothendieck <[EMAIL PROTECTED]> wrote: > > I wonder if it would make more sense to get a relatively > > low level package to run on it so that all packages that > > used that low level package would benefit. The Matrix > > p

Re: [Rd] Master's project to coerce linux nvidia drivers to run generalised linear models

2006-01-24 Thread Douglas Bates
On 1/23/06, Gabor Grothendieck <[EMAIL PROTECTED]> wrote: > I wonder if it would make more sense to get a relatively > low level package to run on it so that all packages that > used that low level package would benefit. The Matrix > package and the functions runmean and sum.exact in > package caT

Re: [Rd] generate Rd file from R file

2006-01-24 Thread Philippe Grosjean
Hi, There is something like that (not so sophiscticated) defined in the green book, but it is not implemented in R. You should look at the mvbutils package, ?dochelp, ?flatdoc, ?doc3Rd, to get such features. Best, Philippe Grosjean Romain Francois wrote: > Hi, > > Is there something similar t

[Rd] generate Rd file from R file

2006-01-24 Thread Romain Francois
Hi, Is there something similar to javadoc or doxygen for R, ie ability to create a first draft of an Rd file reading the comments that are in a script file? For example, for the weighted.mean function in base, one could write the source as : weighted.mean <- function( x, #@ a nume