[R] Funny result from rep(...) procedure

Eik Vettorazzi E.Vettorazzi at uke.uni-hamburg.de
Thu Feb 18 19:24:06 CET 2010


Sorry, not reproducible. This works for me (as expected):

branches<-c(5,6)
iInd = 1
  for(i in 1:length(branches)) {
    print((1:branches[i])+iInd-1)   # iInd is a position shift of the index
    ni = branches[i]
    print(i)
    print(ni)
    print(c(ni,rep(i,times=ni)))
# ... some interesting other stuff for my project that gets wrecked 
because of this issue
    iInd = iInd + branches[i]
}


dkStevens schrieb:
> I'm observing odd behavior of the rep(...) procedure when using variables as
> parameters in a loop. Here's a simple loop on a vector 'branches' that is
> c(5,6,5,5,5). The statement in question is           
> print(c(ni,rep(i,times=ni)))
>
> that works properly first time through the loop but the second time, when
> branches[2] = 6, only prints 5 values of i.
>
> Any ideas, anyone?
>
>   iInd = 1
>   for(i in 1:length(branches)) {
>     print((1:branches[i])+iInd-1)   # iInd is a position shift of the index
>     ni = branches[i]
>     print(i)
>     print(ni)
>     print(c(ni,rep(i,times=ni)))
> # ... some interesting other stuff for my project that gets wrecked because
> of this issue
>     iInd = iInd + branches[i]
> }
>
> # first pass through loop
> [1] 1 2 3 4 5             # branches[1] + iInd - 1 content
> [1] 1                        # i value to repeat 5 times
> [1] 5                        # ni = 5 on 1st pass
> [1] 5 1 1 1 1 1           # five values - 1 1 1 1 1 - OK
>
> # second pass through loop
> [1]  6  7  8  9 10 11   # branches[2] + iInd - 1 content
> [1] 2                       # i value to repeat 6 times
> [1] 6                       # ni = 6 on 2nd pass
> [1] 6 2 2 2 2 2          # 6 'twos' but only shows 5 'twos' 
> print(c(ni,rep(i,times=ni))) - why not 6?
>
>
>   

-- 
Eik Vettorazzi
Institut für Medizinische Biometrie und Epidemiologie
Universitätsklinikum Hamburg-Eppendorf

Martinistr. 52
20246 Hamburg

T ++49/40/7410-58243
F ++49/40/7410-57790



More information about the R-help mailing list