R help - Sep 2004 - Warning: number of items to replace is not a multiple of replacement length

What does this warning mean precisely?
Is there any reason to care about it?
Can I Avoid it by another way of programming?
Thank you in advance.

On Wed, 29 Sep 2004, Mag. Ferri Leberl wrote:
> What does this warning mean precisely?
Usually means there isn't enough room to store your results.
> Is there any reason to care about it?
You probably should.
> Can I Avoid it by another way of programming?
Yes you can. Had you provided an example that caused the
problem, you could have received more help!

----------------------------------------------------------
SIGSIG -- signature too long (core dumped)

On Wed, 29 Sep 2004 15:03:24 +0200, "Mag. Ferri Leberl"
<ferri.leberl at gmx.at> wrote :
>What does this warning mean precisely?
You get this when you do something like this:

x <- 1:11
y <- 1:2

x[1:11] <- y

In the last line, R makes 11 assignments, reusing the values in y[1]
six times and y[2] five times.
>Is there any reason to care about it?
You rarely want recycling to work that way.  It's usually a sign that
you thought y contained a single value, and you wanted it repeated
throughout x.
>Can I Avoid it by another way of programming?
Usually correcting your code removes the message.  If you really
wanted this without the warning, you could do it in two lines:

x[1:10] <- y
x[11] <- y[1]

Duncan Murdoch

P.S. Your reply address ferri.leberl at gmx.at doesn't work; I get a
"mailbox disabled message".

Mag. Ferri Leberl wrote:
> What does this warning mean precisely?
When R replaces parts of a vector with another vector, it repeats the 
replacing value until it is as long as the number of values it needs to 
replace. For example:

  > x = 1:10
  > x[1:3]=1

   - will do, effectively, x[1:3] = rep(1,3)

  Now, if you replace with a vector that isn't an integer multiple of 
the things being replaced, you get that error:

  > x[1:3]=c(1,2)
  Warning message:
  number of items to replace is not a multiple of replacement length

  Note that if it is an integer multiple, you don't get the error:

  > x[1:6]=c(1,2)

  but it repeats the (1,2) 3 times to fill the six places:

  > x
   [1]  1  2  1  2  1  2  7  8  9 10

> Is there any reason to care about it?
  Yes, it usually means you've specified something wrong. What have you 
done?

Baz

hi, 

that means you're doing operation whith matrix or dataframe with  row or
column with different length. this can be a real reason to care about
and sometimes it could be an error and not only a warning.
you'd better adjust the length of your vector (array) to be the same in
order to avoid this warning and be sure you're doing excatly what you
want.

here's a sample of what i meant.
> toto=matrix(0,5,5)
> toto
     [,1] [,2] [,3] [,4] [,5]
[1,]    0    0    0    0    0
[2,]    0    0    0    0    0
[3,]    0    0    0    0    0
[4,]    0    0    0    0    0
[5,]    0    0    0    0    0
> tata=c(1,2,3)
> toto[,1]=tata
Error in "[<-"(*tmp*, , 1, value = tata) :
        number of items to replace is not a multiple of replacement
length

here's the correct syntax
> toto[1:length(tata)]=tata


 
Le mer 29/09/2004 ?? 15:03, Mag. Ferri Leberl a ??crit :
> What does this warning mean precisely?
> Is there any reason to care about it?
> Can I Avoid it by another way of programming?
> Thank you in advance.
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
-- 
------
Yves Magliulo <ym at climpact.com>
R&D Engineer, CLIMPACT

Tel.   : +33 (0) 1 44 27 34 31
Fax.   : +33 (0) 1 44 27 49 96 
Universite Pierre et Marie Curie
Boite 101 - Tour 45 - 5eme etage - Couloir 45/46
4 place Jussieu, 75252 Paris CEDEX 05, France

Mag. Ferri Leberl wrote:
> What does this warning mean precisely?
> Is there any reason to care about it?
> Can I Avoid it by another way of programming?
As you have already gotten most of your questions answered, here is a 
possible answer to the last. You may want to care about it, but not 
necessarily. I have had to use the recycling capability of R without knowing 
whether the length of the result would be a multiple of that of the source 
vector. Say I want to label a vector of daily observations, but I do not know 
which day the observations start or how many there are:

label.weekdays<-function(ndays,start=1) {
 daynames<-rep(c("Sun","Mon","Tue","Wed","Thu","Fri","Sat"),ndays/7+1)
 return(daynames[start:(start+ndays-1)])
}

By creating a vector that is intentionally too long, then truncating it on 
one or both ends, I can recycle the source vector and avoid the warning.

Jim