R help - Mar 2009 - Retrieving the context

Dear list,

I have a general problem that I really don't know how to solve efficiently
in R. Lets say we have a sequence of things, like for instance a string of
words, that is stored in a file. We need all the words in a table format, so
therefore we create an id for the word, that links the word to a file and
the position of the word within the file, like:

#In this case a very short file
> strsplit("This is a text string, wich is stored in the file
myfile","
")[[1]] -> mystring
#Now, store in a data.frame
> mydf <- data.frame(strings=mystring,
word_id=paste("myfile",1:length(mystring),sep="_"))
> mydf
   strings   word_id
1     This  myfile_1
2       is  myfile_2
3        a  myfile_3
4     text  myfile_4
5  string,  myfile_5
6     wich  myfile_6
7       is  myfile_7
8   stored  myfile_8
9       in  myfile_9
10     the myfile_10
11    file myfile_11
12  myfile myfile_12

Now, I would like to see all the words 'is' in a user defined window: so
see_context("is",mydf,1) would give
This is a
wich is stored

and see_context("is",mydf,2) would show two words before and after..
and so
on.

Any ideas on how to solve this kind of problem in R?

/Fredrik


-- 
"Life is like a trumpet - if you don't put anything into it, you
don't get
anything out of it."

	[[alternative HTML version deleted]]

Example:

see_context <- function(word, data, context) {
     dat <- data$strings
     temp <- grep(paste("^", word, "$",
sep=""), dat)
     if(length(temp))
         index <- lapply(temp, function(x) seq(max(0, x - context), 
min(x + context, length(dat))))
     sapply(index, function(x) paste(dat[x], collapse=" "))
}
see_context("is", mydf, 1)
see_context("is", mydf, 2)


Uwe Ligges


Fredrik Karlsson wrote:
> Dear list,
> 
> I have a general problem that I really don't know how to solve
efficiently
> in R. Lets say we have a sequence of things, like for instance a string of
> words, that is stored in a file. We need all the words in a table format,
so
> therefore we create an id for the word, that links the word to a file and
> the position of the word within the file, like:
> 
> #In this case a very short file
>> strsplit("This is a text string, wich is stored in the file
myfile","
> ")[[1]] -> mystring
> #Now, store in a data.frame
>> mydf <- data.frame(strings=mystring,
> word_id=paste("myfile",1:length(mystring),sep="_"))
>> mydf
>    strings   word_id
> 1     This  myfile_1
> 2       is  myfile_2
> 3        a  myfile_3
> 4     text  myfile_4
> 5  string,  myfile_5
> 6     wich  myfile_6
> 7       is  myfile_7
> 8   stored  myfile_8
> 9       in  myfile_9
> 10     the myfile_10
> 11    file myfile_11
> 12  myfile myfile_12
> 
> Now, I would like to see all the words 'is' in a user defined
window: so
> see_context("is",mydf,1) would give
> This is a
> wich is stored
> 
> and see_context("is",mydf,2) would show two words before and
after.. and so
> on.
> 
> Any ideas on how to solve this kind of problem in R?
> 
> /Fredrik
> 
>