My apologies if these are known -- I'm new to R (moved over from S-Plus). All on Windows NT GUI, R1.6.1 and associated packages. 1. save(x) gives the following error message: Error in save(x) : `file' must be non-empty string According to the save() help file, the empty string is the default. 2. In the file>Display file GUI menu, if you try to display a file that is in use by another application -- or at least one particular application I use -- R crashes. 3. I also have noticed that for ** SOME ** data sets I have, the relationargument of the scales= option for xyplot (in Lattice) only works for relation='same' -- both 'free' and 'sliced' give error messages. Unfortunately, it's not handy for me to go reproduce the error at the moment, so I can't give the specifics. I'll send more details in future unless told that this is already known about. NEVERTHELESS -- I am hooked on R. I moved over from S-Plus because I was bothered by their failure to address known bugs and fix their documentation, which is appalling for the new object oriented features. Why should I pay for poor quality? As free and open source software, R has no obligation to do these things (or anything else, for that matter) better than its commercial cousin. But it does! This is just crazy! Aside from the mostly improved qualityand stability, there are lots more cool features,but that you know about.. Also, there aren't all the stupid S-Plus GUI features that I have to remove or ignore (although I did like the Object Explorer for keeping track of my objects and attached directories). Cheers, Bert Gunter Biometrics Research RY 84-16 Merck & Company P.O. Box 2000 Rahway, NJ 07065-0900 Phone: (732) 594-7765 mailto: bert_gunter@merck.com "The business of the statistician is to catalyze the scientific learning process." -- George E.P. Box ------------------------------------------------------------------------------
On Fri, 3 Jan 2003 bert_gunter@merck.com wrote:> My apologies if these are known -- I'm new to R (moved over from S-Plus). > > All on Windows NT GUI, R1.6.1 and associated packages. > > 1. save(x) gives the following error message: > Error in save(x) : `file' must be non-empty string > > According to the save() help file, the empty string is the default.Indeed it is. And the default is to generate an error. That's deliberate, I understand. The help file does say file: a connection or the name of the file where the data will be saved. Must be a file name for workspace format version 1. Since "" is none of the above, it is clearly erroneous. I am not sure what to do to improve this: do you have any suggestions? [That dump() chooses a meaningless filename and people persist in using in (e.g. in S library sections) seems counter-productive to me.]> 2. In the file>Display file GUI menu, if you try to display a file that is > in use by another application -- or at least one particular application I > use -- R crashes.That's hard to debug. Do you get any useful feedback from a Dr.Mingw or other dump? My guess is that is a problem in msvcrt.dll, but I will see if I can trigger it. What Windows version, BTW?> 3. I also have noticed that for ** SOME ** data sets I have, the relation> argument of the scales= option for xyplot (in Lattice) only works for > relation='same' -- both 'free' and 'sliced' give error messages. > Unfortunately, it's not handy for me to go reproduce the error at the > moment, so I can't give the specifics. I'll send more details in future > unless told that this is already known about.Lattice has been updated a couple of times recently. If this persists once R 1.6.2 is out (due in a week's time) we would want to see the details. -- Brian D. Ripley, ripley@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595
I know that Bert has used something like the following in his S code in the
past:
myfunction <- function(..., requiredArg = stop("requiredArg not
supplied",
...)
Wouldn't this work? This would be a problem if somehow the requiredArg is
not evaluated, but then it wouldn't be a required agrument...
We are still on NT4sp6. Scheduled to move to XP this year.
Cheers,
Andy
> From: ripley@stats.ox.ac.uk [mailto:ripley@stats.ox.ac.uk]
> Sent: Friday, January 03, 2003 12:52 PM
> To: bert_gunter@merck.com
> Cc: r-devel@stat.math.ethz.ch; R-bugs@biostat.ku.dk
> Subject: Re: [Rd] save bug (PR#2418)
>
>
> On Fri, 3 Jan 2003 bert_gunter@merck.com wrote:
>
> > My apologies if these are known -- I'm new to R (moved over
> from S-Plus).
> >
> > All on Windows NT GUI, R1.6.1 and associated packages.
> >
> > 1. save(x) gives the following error message:
> > Error in save(x) : `file' must be non-empty string
> >
> > According to the save() help file, the empty string is the default.
>
> Indeed it is. And the default is to generate an error. That's
> deliberate, I understand. The help file does say
>
> file: a connection or the name of the file where the data will be
> saved. Must be a file name for workspace format version 1.
>
> Since "" is none of the above, it is clearly erroneous.
>
> I am not sure what to do to improve this: do you have any suggestions?
>
> [That dump() chooses a meaningless filename and people
> persist in using in
> (e.g. in S library sections) seems counter-productive to me.]
>
>
> > 2. In the file>Display file GUI menu, if you try to display
> a file that is
> > in use by another application -- or at least one particular
> application I
> > use -- R crashes.
>
> That's hard to debug. Do you get any useful feedback from a Dr.Mingw
> or other dump? My guess is that is a problem in msvcrt.dll,
> but I will
> see if I can trigger it. What Windows version, BTW?
>
>
> > 3. I also have noticed that for ** SOME ** data sets I
> have, the relation> > argument of the scales= option for xyplot (in
Lattice) only
> works for
> > relation='same' -- both 'free' and 'sliced'
give error messages.
> > Unfortunately, it's not handy for me to go reproduce the
> error at the
> > moment, so I can't give the specifics. I'll send more
> details in future
> > unless told that this is already known about.
>
> Lattice has been updated a couple of times recently. If this persists
> once R 1.6.2 is out (due in a week's time) we would want to see the
> details.
>
> --
> Brian D. Ripley, ripley@stats.ox.ac.uk
> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
> University of Oxford, Tel: +44 1865 272861 (self)
> 1 South Parks Road, +44 1865 272866 (PA)
> Oxford OX1 3TG, UK Fax: +44 1865 272595
>
> ______________________________________________
> R-devel@stat.math.ethz.ch mailing list
> http://www.stat.math.ethz.ch/mailman/listinfo/r-devel
>
------------------------------------------------------------------------------
On Fri, 3 Jan 2003 18:34:49 +0100 (MET), you wrote:>2. In the file>Display file GUI menu, if you try to display a file that is >in use by another application -- or at least one particular application I >use -- R crashes.I've fixed this (it'll show up in 1.6.2). It's not in the beta build on my web page <http://www.stats.uwo.ca/faculty/murdoch/software/r-devel>, but will be soon. Check the date on the file to see when I've updated it. Duncan Murdoch