Re: Sed, awk?

[John Haxby <jch@xxxxxxxxxxxxxxx>]

Jake McHenry wrote:

> Thanks! Would there be a way to get them all on one line instead of
> many, seperated by a space for input to another command? Being they
> are all on their own lines, it's kind of messy for input to another
> command.
>  
   echo $(oneliner | grep 'whatever' | cut -f1 -d:)
or
   echo $(oneliner | awk -F: '/whatever/' {print $1}')
or
   echo $(oneliner | sed -n '/whatever/s/:.*//p')
or
   echo $(oneliner | perl -ne 'if (/whatever) {split(/:/); print @_[0] 
. "\n"}')
or
   echo $(for i in $(oneliner | grep '/whatever/'); do echo "${i%:*}"; 
done)

(This last one has problems when "oneliner" produces lines that match 
the RE that contain spaces, but that's fixable.)


Replace "echo" with the command of your choice.

How many different ways would you like?   This is Unix :-) I got bored 
thinking up different ways.


Actually, I've just noticed you said:

   oneliner | grep 'whatever' *

In this case grep looks for whatever in all the files in the current 
directory.   What's the purpose of "oneliner"?  Or is  the "*" an oversight.

jch

>  


--
Shrike-list mailing list
Shrike-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/shrike-list