Re: [Rd] History of seq and seq.int

2010-10-05 Thread Prof Brian Ripley
The history is open for all to see. E.g. from the NEWS file (ONEWS in pre-2.12.0) entry for R 2.4.0 o There is a new primitive seq.int() which is slightly more restricted than seq() but often very much faster, and new primitives seq_along() and seq_len() which are faster

Re: [Rd] SweaveInput + keep.source = TRUE trouble

2010-10-05 Thread Claudia Beleites
Sorry, I forgot the sessionInfos: devel: R version 2.12.0 beta (2010-10-03 r53154) Platform: x86_64-unknown-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C [3] LC_TIME=en_US.UTF-8LC_COLLATE=en_US.UTF-8 [5] LC_MONETARY=C LC_MESSAGES=en_US.UTF-8 [7]

Re: [Rd] SweaveInput + keep.source = TRUE trouble

2010-10-05 Thread Duncan Murdoch
On 05/10/2010 5:38 AM, Claudia Beleites wrote: Dear all, I have trouble with R-beta sweaving files that include definitions with \SweaveInput in combination with keep.source = TRUE Thanks for the report. I'll take a look. Duncan Murdoch Symptom: SInput is taken from too far down the

[Rd] library: loading of required packages and argument 'lib.loc'

2010-10-05 Thread Renaud Gaujoux
Hi, is there a reason why when calling the argument 'lib.loc' of library() is not used when trying to load the required packages? 0. Package A depends on package B 1. Install packages A and B in a subdirectory 'lib' of the current directory 2. Call: library(A, lib.loc='./lib') This will

Re: [Rd] SweaveInput + keep.source = TRUE trouble

2010-10-05 Thread Duncan Murdoch
This is fixed now in R-devel and the beta version. The fix isn't perfect: if you use \SweaveInput, you lose the error reporting that gives line numbers from the original file. It should be possible to have both, but I think the changes to implement that are too large to consider in a beta.

[Rd] args() and body() equivalents for S4 methods?

2010-10-05 Thread Vinh Nguyen
Dear list, Suppose some methods were defined for a generic function genF(fitobj, ...). Suppose one has signature(fitobj=lm) and signature(fitobj=glm). The ... piece of each method could have different arguments. I'm aware of getMethod(genF, lm) to get the function definition but are there

Re: [Rd] args() and body() equivalents for S4 methods?

2010-10-05 Thread Vinh Nguyen
On Tue, Oct 5, 2010 at 4:45 PM, Thomas Lumley tlum...@uw.edu wrote: You could define args4-function(generic, signature) args(getMethod(generic,signature)) and body4-function(generic, signature) body(getMethod(generic,signature)) or use selectMethod() instead of getMethod() if you want to

[Rd] How do I put ldpaths in a specific place?

2010-10-05 Thread Tena Sakai
Hi Everybody, I made R (v. 2.11.1) from source with the following config option: ./configure --prefix=/mnt/ami-centos5.5-64bit-base \ --exec-prefix=/mnt/ami-centos5.5-64bit-base \ --with-tcltk All seemed to go well, but when I tested it with: