R help - Apr 2008 - Regular Expressions Help

I am having some trouble learning regular expressions. Let me describe
the general problem I am dealing with. Consider the following setup:

Joe<- c(1,2,3)
Bob<- c(2,4,6)
Alice <- c(9,8,7)

Matrix <- cbind(Joe, Bob, Alice)
St <- c("Bob", "Alice", "Alice:Bob")

Now I want to make a new matrix having only the column's listed in St
that were in Matrix (which I can do). However, In addition for
elements of the form "Alice:Bob" I want to make a new column in my new
matrix with the product of these columns. I am not sure if a semicolon
is a valid symbol for a column name, if not we can replace it with
something like qqq.

I have been reading over various post on regular expressions, but
really haven't made any progress. As far as I can tell there aren't
standard string functions in R. (Also, as an aside, is there a
wildcard character in R? I'd want something so if x="Bob" a
statment
of the form x== "B*b" would evaluate true where * is the wildcard.)

 Any help would be appreciated!

(also, thanks again to those who helped on my last question!)

On 19.04.2008, at 06:46, maud wrote:
> I am having some trouble learning regular expressions. Let me describe
> the general problem I am dealing with. Consider the following setup:
>
> Joe<- c(1,2,3)
> Bob<- c(2,4,6)
> Alice <- c(9,8,7)
>
> Matrix <- cbind(Joe, Bob, Alice)
> St <- c("Bob", "Alice", "Alice:Bob")
> [...]
> I have been reading over various post on regular expressions, but
> really haven't made any progress. As far as I can tell there aren't
> standard string functions in R. (Also, as an aside, is there a
> wildcard character in R? I'd want something so if x="Bob" a
statment
> of the form x== "B*b" would evaluate true where * is the
wildcard.)

I'm not really sure if I understood you correctly.
If you're looking for a way to match St against something like
"B*b"
you can make usage of normal regular expressions (like grep()).

For details begin with ?regexp and ?grep

Then you can try for instance:

grep("B.b", St)

to get
[1] 1 3

(the indices of the vector St)

or directly

St[grep("B.b", St)]

Cheers,
--Hans