Retrieve the content of a json file using shell script .Without jq /json
 

I need to get the read contents of this json file
{
"instance": {
"id": "i-061a20e8c5",
"taglist" : [
{"Key": "acp-ims-backup","Value": "sqldb"} ,
{"Key": "created by","Value": "Historia"}
],
"backup_config_clientname": "CAJ-i-061a20e8c"
}
}

I want the value of "backup_config_clientname"
I can assign it to a variable and print it.
using awk or sed .

Thanks in advance

And what's wrong with jq that you don't want to use it? I mean grep, sed, awk and friends are best suited for manipulating line-oriented data. They are an exceptionally bad fit for parsing non-line oriented formats like JSON or HTML.

More to the point - show us what efforts you have made yourself; either jq, awk, sed, ... whatever.

I have tried using jq
Well the problem in redhat is just a yum (yum install jq ) command and for suse I have to hit the internet, which I would like to avoid .
I saw these are the commands
zypper addrepo https://download.opensuse.org/reposi...SUSE:12.3.repo
zypper refresh
zypper install json-c

I have tried using jq
Well the problem in redhat is just a yum (yum install jq ) command and for suse I have to hit the internet, which I would like to avoid .
I saw these are the commands
zypper addrepo https://download.opensuse.org/reposi...SUSE:12.3.repo
zypper refresh
zypper install json-c

clientname=`jq -r '.instance.backup_config_clientname' $configfile`

echo "Backup Clinet name $clientname"

configfile is the name of the file

What version/distro of Linux are you using?? Because the repository you mention is VERY old. And why are you wanting to avoid hitting the Internet to install a package? Why is that an issue? Pretty much all distros use online repositories. Also seems fairly easy (providing your input file is consistent), to just grep for "backup_config_clientname", then use awk/sed to tidy it up.

Sorry, I don't get it. According to pkgs.org jq is present in the OpenSUSE Oss repository.

I am using it for a suse 12 machine. Probably for all Suse machines.
I am not good at awk and sed , that's why posted the question :(

believe me, much easier to install jq and use it. Especially because you posted the solution in #6.

Marking this as SOLVED

As a quick and dirty workaround
But this code is wrong on so many levels. You really, REALLY (I cannot stress it enough) should be using jq for this even if that means building it from source.

If you mean SLES 12, that's a pay-for distro, and as such, you pay for access to the online repositories; again, why are you not wanting to download and install a package easily? And if you mean openSUSE 12, that is ANCIENT and needs to be updated as soon as possible.
If you never try to do anything, you will always be 'not good' at it; given that there are thousands of easily found examples for awk, grep, and sed you could find with simple search terms (like "find a line in a text file with grep" or "split text with awk"), along with the man pages, you could have made a start at trying to do this on your own. Again, we're happy to help, but don't ask us to do it for you.