Can awk use groups in regular expressions ?

[majkinetor]

I want to translate this line from SED to AWK:

s/(g1)(g2)/\1\n/g

is this possible to do in AWK ?
I didn't find anybody using \N in it.


If not,is there any variant of awk supporting this ?

thx.

 

[feherke]

Hi

Yes, [tt]gawk[/tt], the GNU version of [tt]awk[/tt]. But only in the [tt]gensub()[/tt] function.

[blue]master #[/blue]echo "hello world" | awk '{$0=gensub(/(^| )(.)/,"\\1[\\2]","g")}1'
[h]ello [w]orld

Feherke.

http://rootshell.be/~feherke/

 

[majkinetor]

AH thank you very much.

 

[majkinetor]

Its very strange though that program doesn't work at windows console ?!

I used this script:

{
      $0 = gensub(/(.+) (.+)/, "\\2 \\1", "g")
      print
}

When I create the file and use command

 echo | gawk -f script.awk

it works nice

but if I set the script in "..." it reports (not with this script:

gawk "{$0 = gensub(/(.+) (.+)/, "\\2 \\1", "g");print}" out

gawk: cmd. line:1: {$0 = gensub(/(.+) (.+)/, \\2
gawk: cmd. line:1: ^ backslash not last character on line

 

[majkinetor]

sorry, the above command is

echo abc def | gawk -f script.awk

 

[feherke]

Hi

Use single quotes ( ' ) around the code :

gawk [red][b]'[/b][/red]{$0 = gensub(/(.+) (.+)/, "\\2 \\1", "g");print}[red][b]'[/b][/red] out

Feherke.

http://rootshell.be/~feherke/

 

[majkinetor]

Ah, stupid me....

... I work in 2 many languages...



Thank you