matcher = ( "test1 test2" =~ /test([0-9]+)/ )
matcher.each { fullMatch, number ->
    println "full match: $fullMatch, number: $number"
}

prints:

full match: test1, number: 1
full match: test2, number: 2

I was struggling to get groups out of a find until I read this :

“A limitation of the iterator-based methods is that they don’t give you access to the individual groups”

why doesn’t this work (why Is x a sting instead of list ) :

matcher = ( “test1 test2″ =~ /test([0-9]+)/ )

for ( x in matcher ){
println x
}

On the down side, I can’t get that song out of my head and it’s getting depressing :)

@kodeninja that is an interesting situation that I hadn’t considered previously. It looks like “=~” takes precedence over the “~” in the groovy parser.

So that in your example where you don’t have a space between any of “=~/”, the groovy parser is picking “=~” over the “~/”.

This demonstrates what’s going on:

shoutedWord = "NASA"
 
// no space between = and ~ so it treats it as a =~
// returns a matcher object and doesn't modify shoutedWord
def noSpaceCreatesMatcher() { shoutedWord =~/\b[A-Z]+\b/ }
 
// space between = and ~ so this assigns the Pattern to shoutedWord 
def spaceAssignsPattern() { shoutedWord = ~/\b[A-Z]+\b/ }
 
assert "NASA" == shoutedWord
assert java.util.regex.Matcher == noSpaceCreatesMatcher().class
assert "NASA" == shoutedWord
assert java.util.regex.Pattern == spaceAssignsPattern().class
assert java.util.regex.Pattern == shoutedWord.class
assert ( /\b[A-Z]+\b/ == shoutedWord.toString() )

I guess the moral of the story is to be explicit where you put your spaces to insure you’re getting the right intention. If you want a matcher use “=~ /” If you want a pattern, use “= ~/”.

def shoutedWord = “NASA”
def shoutedWordP = ~/\b[A-Z]+\b/ // creates a pattern
def shoutedWordM = “EUREKA” =~ /\b[A-Z]+\b/ // creates a matcher

shoutedWord =~/\b[A-Z]+\b/ // what does this create?

-kodeninja