Sed inside awk

ZAMO · 02-26-2009, 02:54 AM

Hi all,

Is it possible to insert a sed inside awk?? Am trying to output the awk contents to 2 different files.

Code:

cat myfile|awk -F "*" '{print $1F >> "tfile1" ; print $NF >> "lastfile"} '

In this i need to do a sed in my first print pattern of awk . So am trying with the following and failing. Is there a way to do it.

Code:

cat myfile|awk -F "*" '{print $1F |sed 's/#//g' >> "tfile1" ; print $NF >> "lastfile"} '

I want to remove the # in file in my first awk. Is there a way to use sed inside awk or am I trying with something wrong.

colucix · 02-26-2009, 04:13 AM

Indeed you can use external commands within an awk program and share their output, but you don't really need them to achieve your task. In awk there are some useful functions to manipulate string. For example the sub and the gsub functions, act the same as sed. They alter the content of the fields, so that when you print them out, you will see the new value of the string.

Suppose you have the file

Code:

$ cat myfile
#this is a commented line
this is a line of code
#another commented line

You can use the gsub function in the following way:

Code:

$ awk '{gsub(/#/,"",$1); print}' myfile
this is a commented line
this is a line of code
another commented line

the value of the first field has been altered substituting the part of the string matching the regexp /#/, with the replacement string "". The subsequent print command, prints out the line with the altered field. Take a look at the string manipulation functions on the AWK Programming Guide for more details. Hope this helps.