Add column from another file into specific area between quotes

rootaccess · 11-19-2021, 12:44 PM

I have a list of rows that look like this.

Code:

{data, {"data", "data-that-needs-to-be-changed"}},

The 3rd column is what I need to change. I have that 3rd column solo in a file from a previous

Code:

awk -F\" '{print $4}' file

which just prints the area defined as

Code:

data-that-needs-to-be-changed

between the quotes.

Thanks.

pan64 · 11-19-2021, 12:50 PM

is this a json file?
You need to use a tool which can handle that properly (for example python). awk is definitely unsafe in such cases.

shruggy · 11-19-2021, 02:09 PM

Field variables in awk can be assigned to

Code:

 awk -F\" -vOFS=\" '$4="new-data"' file

But as pan64 said, if this is indeed JSON then better handle it with a JSON parser. If this were part of a shell script, jq would be my tool of choice.

rootaccess · 11-19-2021, 10:28 PM

Quote:

Originally Posted by shruggy

Field variables in awk can be assigned to

Code:

 awk -F\" -vOFS=\" '$4="new-data"' file

But as pan64 said, if this is indeed JSON then better handle it with a JSON parser. If this were part of a shell script, jq would be my tool of choice.

Wasnt able to do much with that since new-data is data that is a file full of rows that needs to modify that. but i ended up just using paste with that column. i printed just the 2 columns up to the 3rd quote, then i pasted the column with the new data, then i just appended a

Code:

"}},

to the end of each row. simple as that. thank you.

Turbocapitalist · 11-19-2021, 10:44 PM

Right. But if the format of that data is JSON then there are specific tools which will be easier to work with, reduce or eliminate formatting errors, and be more flexible and less brittle.

MadeInGermany · 11-27-2021, 08:29 AM

With shell built-ins

Code:

#!/bin/sh
q=\"
while
  IFS="$q" read f1 f2 f3 f4 f5 <&3
do
  read nd <&4
  echo "$f1$q$f2$q$f3$q$nd$q$f5"
done 3<file 4<new-data