CODEWITHSUNDEEP
r
BRZ
BRZ

Reputation: 695

Extracting numbers from String

Basic question... I've a string and I'm trying to extract the numbers from a string using the below.

str2="<P>3.1 Design objectives ..... 
<Link>25 
</Link> </P>"
str2
a<-strapply(str2,"\\d+",as.numeric,simplify=TRUE)
a

This returns 3, 1, and 25 which is fine. (It's ok that 3.1 is split up)

However, if the string has no numbers at all:

pr <- "This is a test string"
b<-strapply(pr,"\\d+",as.numeric,simplify=TRUE)

b returns: 

[[1]]
NULL

How can I use the above to detect the presence of a number(s) in a string?

is.null(b) returned False as it should.

length(b) gave 1, which is the NULL Object?

So, is it right to say that the string does not contain any numbers if length(b) is 1? Is there a more elegant way of doing this?

Upvotes: 2

Views: 265

Answers (2)

agstudy
agstudy

Reputation: 121568

I would unlist the result and test its length, For example using strapplyc:

  (ll <- as.numeric(unlist(strapplyc(str2,"\\d+",simplify = TRUE))))
   numeric(0)
   length(ll)
   [1] 0

EDIT Rereading your question, I think it is better/safer to convert your XML structure to a simple text before appyling regular expression on it. For example you can do this :

library(XML)
toString(xmlToList(str2))
[1] "3.1 Design objectives ..... \n, 25,  \n"

Upvotes: 0

IRTFM
IRTFM

Reputation: 263301

require(gsubfn)
pr <- "This is a test string"
b<-strapply(pr,"\\d+",as.numeric,simplify=TRUE)
if( !length(b[[1]]) ){TRUE}
#[1] TRUE

(This is the standard way of testing whether a list element is NULL.)

Upvotes: 4

Related Questions