[SOLVED] bash- sed, awk, grep

schneidz · 01-31-2011, 06:21 PM

hi, lets say i got a file like

Code:

hi my name is simon
and i like to do drawings
are you looking at my bum
l33t h4x0rz simon chun-li
madam, i'm adam

and i want to enumerate the simon's so it looks like:

Code:

hi my name is simon-1
and i like to do drawings
are you looking at my bum
l33t h4x0rz simon-2 chun-li
madam, i'm adam

obviously, sed s/simon/simon-x/ wont help me.

can someone please give me some ideas ?
thankx,

crts · 01-31-2011, 07:04 PM

Hi,

here is an awk solution

Code:

awk '/simon/{c++; gsub("simon","simon-"c)}1' dummy

or even shorter

Code:

awk '/simon/{gsub("simon","simon-"++c)}1' dummy

kurumi · 01-31-2011, 07:22 PM

Quote:

Originally Posted by crts

Hi,

here is an awk solution

Code:

awk '/simon/{c++; gsub("simon","simon-"c)}1' dummy

or even shorter

Code:

awk '/simon/{gsub("simon","simon-"++c)}1' dummy

that don't work if there are more than 1 simon in one line

kurumi · 01-31-2011, 07:40 PM

Ruby(1.9+)

Code:

ruby -F"\s" -ane 'BEGIN{c=0};$F.map!{|x| x["simon"]? x="#{x}#{c+=1}":x=x };puts $F.join("\s")' file

crts · 01-31-2011, 07:47 PM

Quote:

Originally Posted by kurumi

that don't work if there are more than 1 simon in one line

True. But it is not obvious from the OP's example that there will be more than one Simon in a line. And from the sample data provided is also not obvious how multiple Simons in a line should be handled. Until further clarification I like to keep it as simple as possible.

grail · 01-31-2011, 07:52 PM

Quote:

Originally Posted by kurumi

that don't work if there are more than 1 simon in one line

I beg to differ seeing that the 'g' in gsub stands for global changes so all would be included.
What you might have meant to say is that subsequent simon's on the same line will have the same number at the end

Just to clarify

So an alternative would be:

Code:

awk '/simon/{for(i=1;i<=NF;i++)if($i == "simon")$i = $i"-"++c}1' file

schneidz · 01-31-2011, 08:03 PM

Quote:

Originally Posted by crts

True. But it is not obvious from the OP's example that there will be more than one Simon in a line. And from the sample data provided is also not obvious how multiple Simons in a line should be handled. Until further clarification I like to keep it as simple as possible.

only 1 simon per line... i guess multiple simons would be worth bonus points.
(reputation added)

schneidz · 01-31-2011, 08:05 PM

Quote:

Originally Posted by kurumi

Ruby(1.9+)

Code:

ruby -F"\s" -ane 'BEGIN{c=0};$F.map!{|x| x["simon"]? x="#{x}#{c+=1}":x=x };puts $F.join("\s")' file

thanks for offering your suggestion. unfortunately, my corperations installation of aix doesnt have ruby so i couldnt test it out.

kurumi · 01-31-2011, 08:33 PM

Quote:

Originally Posted by grail

What you might have meant to say is that subsequent simon's on the same line will have the same number at the end

yes, that's what I meant.

grail · 01-31-2011, 08:48 PM

Same, same, but a little bit different

Code:

awk '/simon/{gsub(/simon/,"&-"++c)}1' file

crts · 01-31-2011, 08:56 PM

Quote:

Originally Posted by grail

Same, same, but a little bit different

Code:

awk '/simon/{gsub(/simon/,"&-"++c)}1' file

I was suspecting that awk must have an equivalent to sed's backreference mechanism.