AWK - easy: get first column that matches ...

Author

easy: get first column that matches ...

Gernot Frisch

2005-02-01, 3:56 pm

Hmmm.. this must be easy:

Assume a lines:

TEXT "Some Text" ID_SOME
#define ID_CS 12312

I'd like to get the tokens that start with "ID" - what's the easiest
way? split?

--
-Gernot
int main(int argc, char** argv) {printf
("%silto%c%cf%cgl%ssic%ccom%c", "ma", 58, 'g', 64, "ba", 46, 10);}

Stephane CHAZELAS

2005-02-01, 3:56 pm

2005-02-1, 14:49(+01), Gernot Frisch:
> Hmmm.. this must be easy:
>
> Assume a lines:
>
> TEXT "Some Text" ID_SOME
> #define ID_CS 12312
>
> I'd like to get the tokens that start with "ID" - what's the easiest
> way? split?
[...]

That's not a awk solution, but I would do it this way:

tr -s ' \t' '\n\n' | grep '^ID'
or
tr -cs 'a-zA-Z0-9-' '[\n*]' | grep '^ID'

to answer the question in the subject:

{
for (i = 1; i <= NF; i++) {
if ($i ~ /^ID/) {
print $i
break
}
}
}

--
Stéphane

William James

2005-02-01, 8:55 pm

Instead of split, use shatter.

z {
z shatter( $0, a, "(^|[^a-zA-Z_0-9])ID[a-zA-Z_0-9]*" )
z for (i=2;i in a;i+=2)
z { token = a[i]; sub(/^[^a-zA-Z_0-9]/, "", token)
z print token
z }
z }
z
z # Produces array of nonmatching and matching
z # substrings. The size of the array will
z # always be an odd number. The first and the
z # last item will always be nonmatching.
z function shatter( s, array, re )
z { gsub( re, "\1&\1", s )
z return split( s, array, "\1" )
z }

Ed Morton

2005-02-02, 3:56 am

Gernot Frisch wrote:

> Hmmm.. this must be easy:
>
> Assume a lines:
>
> TEXT "Some Text" ID_SOME
> #define ID_CS 12312
>
> I'd like to get the tokens that start with "ID" - what's the easiest
> way? split?

No, just redefine your RS so every "field" gets treated as a separate
record, then look for records that start with "ID":

gawk -vRS="[[:space:]]" '/^ID/'

Regards,

Ed.

Sponsored Links