[SOLVED] Help with sed/awk

maddyfreaks · 10-02-2015, 09:43 PM

===File a.log
----------------------

Code:

Rep N1
Forever
Limited
2015

Rep N2
Limited
2016

----------------------
I want to search for "Forever" After each line of Rep and if i find Forever print "Forever Yours" if i dont fine "Forever" insert line "Healthy Wealthy"

for above file the output should look like this...

Code:

Rep N1
Forever Yours
Limited
2015

Rep N2
Healthy Wealthy
Limited
2016

syg00 · 10-02-2015, 10:00 PM

Yet another thread where you apparently expect someone else to do all your work.
Show us you've made some effort yourself.

maddyfreaks · 10-02-2015, 10:17 PM

there is a file which i made some scripting to get to this level.

Now that am struk here. i know i cannot ask all my work here. So was asking for help.

maddyfreaks · 10-02-2015, 10:29 PM

I know with sed i can replace but i am not sure how to seach for word if matches replace and if doesnt after specific word by inserting new text

maddyfreaks · 10-02-2015, 11:01 PM

Any ideas are appreciated.

HMW · 10-03-2015, 03:04 AM

Quote:

Originally Posted by maddyfreaks

Any ideas are appreciated.

[sigh] You have to realize that, just like in your other post, nobody is going to write this for you (well, I just did). But we are willing to help you figure it out.

I have solved this by writing a small script:

Code:

./readMaddy.sh 
Rep N1
Forever yours
Limited
2015

Rep N2
Healthy Wealty
Limited
2016

And I will be more than happy to share at least parts of my solution and the logic I used, once you have shown us some of your own efforts. Mind you, my solution might not be the best/fastest/smartest, but I have at least solved this problem. Using what knowledge and talent I have. I am 100% sure that not only I, but other users would feel inclined to help you out with this, but it all boils down to, and I am sorry if I repeat myself, helping you help yourself.

So... please show us what you have done this far.

With best regards,
HMW

maddyfreaks · 10-03-2015, 08:15 AM

i tried ti make them as single line but not sure how to add some text after certain column

Code:

$ cat x.log |xargs -n 5
Rep N1 Forever Limited 2015
Rep N2 Limited 2016

i want some text after column 2

maddyfreaks · 10-03-2015, 08:45 AM

tried this way but am not getting whole file.

Code:

$ cat x.log |xargs -n 5|awk '{ if ( $3 !~ /Forever/ ) {printf( "%s\n%s\n", "Healthy", $3 );} else {print $3;}}'
Forever
Healthy
Limited

grail · 10-03-2015, 10:37 AM

Are you confusing questions? Your other question was about making all into one line, which this one does not mention, whereas this question was about updating / adding text.

You may wish to keep everything on the same topic or restate your question as this will confuse those trying to assist.

maddyfreaks · 10-03-2015, 12:43 PM

I want the multiple lines in file to be starting with Rep and ending with year.
So the file i made it to singel line. But the num of feilds are not same . I want the missing feild to be inserted as some text.

for 2nd line am missing 3rd columns so i want to fill it with some dummy value.

either make or add missing 3rd col in straight line

(or)

make or add 3rd line for the file

HMW · 10-03-2015, 01:00 PM

Quote:

Originally Posted by maddyfreaks

I want the multiple lines in file to be starting with Rep and ending with year.
So the file i made it to singel line. But the num of feilds are not same . I want the missing feild to be inserted as some text.

for 2nd line am missing 3rd columns so i want to fill it with some dummy value.

either make or add missing 3rd col in straight line

(or)

make or add 3rd line for the file

WHAT!?!

This is hopeless. Do even you know, for sure, what your desired output is supposed to be. Or is that defined by some entity whispering to you!?!

Sorry if I sound slightly vexed, but this is the second thread where you alter either the input or the desired output and/or change the criteria for your problem.
I am not trying to be mean or disrespectful - but this is, quite frankly, aggravating.

Tell you what, I am going to give you a solution, to your ORIGINAL problem, as specified by you here. You'll get this for free, as in "gratis", and after that I will stay the **** out of your threads.

Here it comes:

Code:

cat maddyfreaks.txt | tr '\n' '+' | sed 's/\(Rep N[0-9]+Forever\)/\1 Yours/;s/\(Rep N[0-9]\)+\([A-EG-Z]\)/\1+Healthy Wealthy+\2/' | tr '+' '\n'
Rep N1
Forever Yours
Limited
2015

Rep N2
Healthy Wealthy
Limited
2016

Take care and all that. I'm out.

HMW

grail · 10-03-2015, 03:43 PM

Yes it is particularly exasperating, so I also will make note to avoid posts with no attempts in them:

Code:

awk '{if(p && /Forever/)$2 = $2 "Yours";else if(p) print "Healthy Wealthy";p = /Rep/?1:0}1' file