[R] shifted window of string

[Charles C. Berry]

On Mon, 14 Jun 2010, david hilton shanabrook wrote:

> basically I need to create a sliding window in a string.  a way to explain this is:
>
>> v <- c("a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r","s","t","u","v","w","x","y")
>> window <- 5
>> shift <- 2
>
> I want a matrix of characters with "window" columns filled with "v" by filling a row, then shifting over "shift" and continuing to the next row until "v" is exhausted.  You can assume "v" will evenly fit "m"
>
> so the result needs to look like this matrix where each row is shifted 2 (in this case):
>
>> m
>      [,1] [,2] [,3] [,4] [,5]
> [1,] "a"  "b"  "c"  "d"  "e"
> [2,] "c"  "d"  "e"  "f"  "g"
> [3,] "e"  "f"  "g"  "h"  "i"
> [4,] "g"  "h"  "i"  "j"  "k"
> [5,] "i"  "j"  "k"  "l"  "m"
> [6,] "k"  "l"  "m"  "n"  "o"
> [7,] "m"  "n"  "o"  "p"  "q"
> [8,] "o"  "p"  "q"  "r"  "s"
> [9,] "q"  "r"  "s"  "t"  "u"
> [10,] "s"  "t"  "u"  "v"  "w"
> [11,] "t"  "u"  "v"  "w"  "x"

The last row is wrong. It is only shifted by 1.

Try

[11,] "u"  "v"  "w"  "x"  "y"

>
> This needs to be very efficient as my data is large, loops would be too slow.

And you know this because you tried it out and timed it??

Please see

 	?system.time

If window is not too big,

 	for( j in 1:window ) m[ , j ] <- v[ j + v.seq ]

for a suitable v.seq will be pretty quick.

For a vectorized solution, some combination of these pieces

 	seq( window, length(v), by=shift )
 	(1-window):0
 	outer( ... , '+') or rep(...) + rep( ... )
 	v[ ... ]
 	matrix( ... )

will give you the answer.

If the problem is so big that neither of these approaches suffices, this 
is a good candidate for the inline package - one line of C code is all 
that is needed.

HTH,

Chuck

Any ideas?  It could also be done in a string and then put into the matrix but I don't think this would be easier.
>
> I will want to put this in a function:
>
> shiftedMatrix <- function(v, window=5, shift=2){...
>
> return(m)}
>
> thanks
>
> dhs
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Charles C. Berry                            (858) 534-2098
                                             Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	            UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego 92093-0901