Pick value from row after a column value

Shell Programming and Scripting
1

Hi,

my input

abc def "."; ghi
abc kgk mhi def "="; klm nop
abc kgg def "j"; klm mno pwd rst

I would like to get the content in the inverted commas after def. So, my output would be

.
=
j

How do I do it? Please make a note that my input field separator is space and not tab.

Thanks

2

Hi

$ cut -d\" -f2 file
.
=
j

Guru.

1 Like
3 
$ sed 's/def /def~/g' <sample10.txt | tr " " "\n" | grep "^def" | cut -c6
.
=
j
4

Hi guru,

Thanks for ur reply.

I just found out that the end portion of my file contains records that are in inverted commas.

abc def "."; ghi
abc kgk mhi def "="; klm nop
abc kgg def "j"; klm mno pwd rst
abc def "="; osr ">"; kil bil wil
fgh osr "{"; gik hik juy def "*";

my output now will be

.
=
j
=
*

Sorry for any inconvenience.

5

Do you mean quotes "
Also, sometimes referred to as double-quotes so as not to be confused with the single-quote symbol '

6

yes, double quotes.

And also, the sed script is printing hyphens.

My def column is changing across the entire file.

Thanks for ur time.

7

Not sure what is printing hyphens. See below.

$ cat sample10.txt
abc def "."; ghi
abc kgk mhi def "="; klm nop
abc kgg def "j"; klm mno pwd rst

$ cat sample11.txt
abc def "."; ghi
abc kgk mhi def "="; klm nop
abc kgg def "j"; klm mno pwd rst
abc def "="; osr ">"; kil bil wil
fgh osr "{"; gik hik juy def "*";

$ sed 's/def /def~/g' <sample10.txt | tr " " "\n" | grep "^def" | cut -c6
.
=
j


$ sed 's/def /def~/g' <sample11.txt | tr " " "\n" | grep "^def" | cut -c6
.
=
j
=
*
8

i am not good with sed or awk , but we can implement something with this regex

("[^"]")

with use of some back reference feature.

Again , if i am wrong , please correct me

---------- Post updated at 01:19 PM ---------- Previous update was at 10:25 AM ----------

[^a-z ;"]

this regex matches only the special character. But dont know how to implement it to extract only the special character. Some help , would be much appreciated

9

Hi,

my input

abc def "."; ghi
abc kgk mhi def "="; klm nop
abc kgg def "j"; klm mno pwd rst

I have changed my output content which would be 3 separate files.

The rows with def "." will go into file by name dot.txt

And, the rows with def "=" will go into a file by name equal.txt

And the rows with def "j" will go into a file by name jack.txt

The remaining records should go into a file by name remaining.txt

How do I do it? Please make a note that my input field separator is space and not tab.

Thanks in advance

10

Not pretty, but should fit your request

awk '/\"\.\"/{print>"dot.txt"};/\"=\"/{print>"equal.txt"};/\"j\"/{print>"jack.txt"};!/\"[\.=j]\"/{print>"remaining.txt"}' inputfile
1 Like
11

Hi cabrao,

Thanks for your response.

The code should consider the letters def before the quotes too.

How do I append it?