On Wed, 5 Mar 2003, Paul Gilbert wrote:
> versions, and I don't think require() checks versions. By putting the packages
> in a bundle I am relatively sure a user will get versions of the packages that
> are suppose to work together. Of course, this is a bigger problem than just my
> packages, so a
On Thu, 5 Jun 2003, John Chambers wrote:
> The current code passes check-all here. Please let me know of any
> problems. I'll be away the rest of today, but back tomorrow.
Running the test tha I had been doing shows amazing speed gains. It is
about 5-6 times faster then my May 23 version of R
Hello ...
I've noticed that some of our Bioconductor code was running drastically
slower under current R-devel vs. current R-patched - one example is below
using a ttest. I have the following snippet of code that demonstrates the
problem while avoiding "real" code that takes an extremely long tim
On Wed, 4 Jun 2003, John Chambers wrote:
> There was a bug leading to non-caching of methods, in the r-patched code
> from a few days ago.
> A branch update would have put the changes into R-devel also.
> The problem has been partly fixed in the current code committed to
> R-patched (and some fur
On Wed, 18 Jun 2003, Girish Zambre wrote:
> BTW, a real stupid question : is there a line command that can change the
> file directory ( unix analogue would be "cd") in Rgui.exe ? I change
> directories using the item under the "File" menu right now, but that's a
> real pain.
A while back I whip
I'm having difficulty building R-devel on Solaris 8 (current version),
while trying to update from my previous version of R-devel (2003-09-02).
When building I get this error:
---
make[4]: Leaving directory
`/misc/homes/madman/R-devel/src/library/methods/src'
make[4]: Entering director
> Looking over the Debian changelogs reveals nothing -- just maintainer
> stuff.
Just to provide more (potentially useful) info, I'm having the same
problem on a Solaris 8 machine (with the same workaround on the
Blue/blue/Red/red thing) and the teTeX installed has definitely not been
changed fo
Hello ...
With a new checkout of R-devel (last update was 2003-09-11) we are having
a problem (it seems to be happening to all of us here on a few different
machines) where during install/check/etc when the 'save image' happens (in
packages using 'save image'):
** save image
Error: couldn't find
> Did you make sure that you have
> require(methods)
> early in your R sources for the package?
No, and that was exactly the problem (although specifically I put the
'require(methods)' into the R_PROFILE.R file). I was under the impression
that methods was always loaded by default when R started
I have encountered an issue while preparing some of the Bioconductor
packages for our upcoming release, and Duncan Murdoch suggested that I
bring one of the related points up here.
The background is that we are building our packages under Windows with
"Rcmd install --build" which will flag the zi
ul to the community as well as feedback &
input.
Thanks
Jeff Gentry
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-devel
Hello ...
On both RH7.2 and Solaris 8 I'm encountering this problem - in both cases
this is from a fresh rsync of R-devel into a brand new directory (to
prevent it from me needing to 'make distclean' or the like):
building package 'ts'
mkdir -p -- ../../../library/ts/R
make[3]: Leaving directory
Out of curiosity, what happened to this?
With 1.8.1 it returns:
> getOption("repositories")
CRAN
"http://cran.r-project.org/src/contrib";
BIOC
"http://www.bioconductor.org/src/contrib";
With R-devel it returns:
> get
> On what OS?
This happened on both RH7.2 linux and Solaris 8.
> and was due to typo in src/library/Makefile.in (rm -r should be rm -f),
> and should be correct now (but rsync lags a bit behind cvs).
It doesn't seem to have pushed out to rsync yet, but applying your change
by hand to my src/li
> It isn't needed to be set any more by default: it merely duplicated the
> information in options CRAN and BIOC.
IMO, it is more useful then that, however. The nice thing about
"repositories" is that it provided one spot which could have 1->n
repositories listed. Having a catch all "repositorie
Hello ...
With a new checkout of R-devel, I'm getting the following error on
startup:
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for a HTML browser interface to help.
Type 'q()' to quit R.
Error in open.connection(con, "rb") : unable to open connection
In addition
> There has been a problem with (someone else's!) update to package stats.
> As with yesterday, it should already be fixed in cvs and will get fixed in
> rsync. A clean build from the cvs works.
Yeap - it is definitely working now. I must have just grabbed an R-devel
from before the fix when I
> Try installing into a clean directory. That is, remove the contents of
> /home/jgenty/R-1.9.0 before installing.
Right ... this is something that I try to do before speaking up that
R-devel isn't working for me (tho sometimes I forget ...).
In this case though it sounds as if I caught my rsync
Hello ...
Are there any known problems or even gotchas to look out for when using a
gzfile connection in read.csv/read.table in Windows?
In the package PROcess, available at
www.bioconductor.org/repository/devel/package/html/PROcess.html
there are two files in the PROcess/inst/Test directory whic
Hello ...
Using Win2K (and reportedly WinXP), when the length of the 'url' string
>= 280 characters, a segmentation fault occurs.
This doesn't seem to be affecting unix machines.
Thanks
-J
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.
> With what settings of argument `browser'?
In this case, none.
getOption("browser") returns NULL. When browseURL() is working form e
under that configuration it pulls up IE5
> If you absolutely must have such long URLs, try specifying a browser
> (although you will probably find limits that w
> > Well, the toy example I was using to first verify that it was coming from
> > browseURL in general was just to do this:
> > z <- rep("z", 300)
> > z <- paste(z, collapse="")
> > browseURL(z)
> That's not a URL at all, and I get nothing (as I should). If I put http://
> in front it works (as a
Hello ...
I noticed this last night when I updated R-devel, and then figured that I
had some 'make clean' or related issues and would look at it today. This
morning I did a rsync of r-devel into a fresh directory and am now having
the same issue.
Basically, whenever R is trying to install a pack
> I believe you are finding the Windows FIND.EXE not the find.exe in my set
> of tools. Have you cross-checked the path list in the INSTALL file?
That's exactly what it was. The 'tools' were ahead of all of the other
software I have installed to build packages and such, but was after the
Win32
On Wed, 21 Jul 2004, Dirk Eddelbuettel wrote:
> Precisely. Let's have one source repo but _let us label any and all binary
> restrictions_ more clearly so that I for one can skip over stuff that may
> build for you [ Windoze ] but won't for me [ Linux, preferable on all all
> ten to twelve hardware
> This is really an R-devel question -- it is about an unreleased version of
> R.
Right you are - sorry about that.
> Yes. Seems a make peculiarity, not even shared by Solaris make.
> To get you going, replace $^ by
> $(top_builddir)/library/$(pkg)/R/$(pkg).rdb. However, I am working right
>
I'm getting this error when installing packages w/ a checkout of R from a
a little bit ago (23-Aug-2004, approx 1pm eastern US time):
** help
Bareword found where operator expected at
/usr/home/jgentry/R/share/perl/R/Rdconv.pm line 2331, near "$latexout
latex_link_trans0"
(Missing operator
Here's another issue (that might well be operator error):
> install.packages("lattice")
...
...
** save image
Loading required package: grid
Error in importIntoEnv(impenv, impnames, ns, impvars) :
object(s) 'dev.list', 'cm.colors', 'gray', 'heat.colors' are not
exported by 'namespace:gra
> What version of perl?
Ack, didn't realize it was this ancient. Version 5.005_03, which is what
comes with FreeBSD 4.9 apparently. I did install the /usr/ports version
of perl is 5.8.2, although there seems to be other problems here (which
are most likely related to my system, will track that d
> Yes, the released lattice version is no longer compatible with r-devel after a
> split of the graphics package, and install.packages doesn't know to look in
> http://cran.us.r-project.org/src/contrib/2.0.0/Recommended/. Doing
> a ./tools/rsync-recommended should get you the correct version.
T
> Does anyone have any opinions as to the merits of this idea? Would there
> be any interest in seeing such a class in R?
Have you looked at the 'externalVector' package in Bioconductor? I'm
admittedly not super familiar with it, although my understanding of how it
works and what it does seems
> I don't have the `R Language Manual'. Is it explained somewhere
> in the docs that come with R?
See: /doc/manual/
__
[EMAIL PROTECTED] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
> My question is, what else is necessary for a minimal package? "Writing R
> Extensions" states:
>"The `Package', `Version', `License', `Description', `Title', `Author',
> and `Maintainer' fields are mandatory..."
> So should I really add all these (superfluous) fields, to avoid future
> r
> The current r-devel (aka R 1.7.0) now attaches the package "methods" by
> default at startup.
There's either a problem here or something is wrong on my end as I just
did a rsync to R-devel and the build bombed out with:
gcc -shared -L/usr/local/lib -o methods.so do_substitute_direct.o
methods_l
On Fri, 17 Jan 2003 [EMAIL PROTECTED] wrote:
> > There's either a problem here or something is wrong on my end as I just
> > did a rsync to R-devel and the build bombed out with:
> Try make clean and/or removing library/methods?
It actually turned out to be the latter - removing R/library/methods
Hello ...
Using a relatively recent rsync of R-devel (it started at least after
1/16/03 as I have a copy of R-devel from then and this doesn't happen),
I've noticed a bit more chatter to the console when accessing URL
connections.
> z <- url("http://www.bioconductor.org/main.html";)
> readLines(z
y difficulty with this package.
Thanks
-Jeff Gentry
__
[EMAIL PROTECTED] mailing list
http://www.stat.math.ethz.ch/mailman/listinfo/r-devel
Sorry for the self-followup, but just a clarification ...
> It's purpose is to generate UUID values, and has a simple R interface
> (currently one function: getuuid(), with multiple options).
For those that are not familiar, a UUID is a "Universally Unique ID",
a value generated that is guarante
> maintainers to generate the PACKAGES file, and to generate PACKAGES.html
> from PACKAGES. Are those something that could be made available? Or
> do people think that I should bite the bullet and submit my packages to
> CRAN instead?
You might want to take a peak at the 'reposTools' package
> Yes, par(family="mono") would work, except that I get R segfaults from this
> sequence:
> > plot.new()
> > par(family="mono")
> > par(cex=8)
> > strheight("foo")
> Process R segmentation fault (core dumped) at Mon Dec 20 16:07:56 2004
> on R 2.0.1 (2004-11-15), Red Hat Enterprise Linux AS release
> I prob. missed it but I would like to make string
> concating. Is there any string concating function in
> R?
> A <- "abc";
> B <- ".jpeg"
>
> C <- c(A,B);
> it does not do abc.jpeg in string format..?
?paste
__
R-devel@stat.math.ethz.ch mailing list
On Wed, 9 Feb 2005, Achim Zeileis wrote:
> A *library* is a directory in which you can find R *packages* (just as
> in real life you can find books in a library) and with
> library("foo", lib.loc = "/path/to/bar")
> you want to get the package (book) "foo" from the library "bar" located
> at "/p
On Thu, 21 Apr 2005, Ali - wrote:
> >However, in S3 you can create a "generic" generic function by not
> >specifying
> >arguments but only '...' - this way any methods can take any arguments (and
> >you don't force your argument names onto other developer's).
> So why did they go a step backward
On Thu, 21 Apr 2005, Ali - wrote:
> I guess the context says you cannot have something like this in S4:
> >setGeneric("testFun", function(...)
> + standardGeneric("testFun"))
But what is wrong with:
setGeneric("testFun", function(object, ...))
Keeping in mind that 'object' is the actual instant
On Thu, 21 Apr 2005, Henrik Bengtsson wrote:
> > But what is wrong with:
> > setGeneric("testFun", function(object, ...))
> Naming conflicts, may be the problem. You have that generic in your package,
> but I might, without know about yours, and I write
> setGeneric("testFun", function(x, ...))
45 matches
Mail list logo
