how get multiple line starting with and format them - awk

Question

I Have a file which contains multiple lines.I want only few things from that output.Below is the output i am getting from server.
Output:
"az": "nova",
"cloud": "envvars",
"config_drive": "",
"created": "2016-08-19T17:21:24Z",
"flavor": {
"id": "4",
"name": "m1.large"
},
"hostId": "f714baee5967dc17e7d36c7b72eb92a4f1ab68d9782fa90a968ceae5",
"human_id": "dsc-test-3",
"id": "3f0a1188-c151-4e5e-9930-969d0423601b",
"image": {
"id": "7f4ad1f4-6fab-4978-b65a-ec4b9a407c5c",
"name": "mitel-dsc-7.9-9.15_3nic"
},
"interface_ip": "172.16.17.15",
"key_name": "key1",
"metadata": {},
"name": "dsc-test-3",
"networks": {
"dsc-InterInstance": [
"172.16.18.15"
],
"dsc-OAM": [
"172.16.16.20"
],
"dsc-sig": [
"172.16.17.15",
"10.10.72.15"
]
},
My intention is to get below things
Required:
"networks": {
"dsc-InterInstance": [
"172.16.18.15"
],
"dsc-OAM": [
"172.16.16.20"
],
"dsc-sig": [
"172.16.17.15",
"10.10.72.15"
]
}

Answer 1

Try this method
sed -n '/networks.*{/,/}/p' fileName
Outputs:
"networks": {
"dsc-InterInstance": [
"172.16.18.15"
],
"dsc-OAM": [
"172.16.16.20"
],
"dsc-sig": [
"172.16.17.15",
"10.10.72.15"
]
}

Answer 2

Awk solution:
awk '/"networks"/,/}/' file.txt
This gives the output
"networks": {
"dsc-InterInstance": [
"172.16.18.15"
],
"dsc-OAM": [
"172.16.16.20"
],
"dsc-sig": [
"172.16.17.15",
"10.10.72.15"
]
},
The syntax used here is /start_regex/,/stop_regex/, which starts matching when a line matches start_regex and stops matching after a line matches stop_regex (so all the lines in between also get matched). Since no action is specified, the default {print} action is used.
Compared to your requirement, this outputs an extra , on the last line, since that comma is present in the input. If that is unacceptable, you could get rid of it using the action {sub("},","}");print}. Or by using sed as in the other answer.

Answer 3

... and the tagged grep solution as well:
$ cat file|tr '\n' _| grep -o \[^_\]*\"networks\"\[^}\]*}|tr _ '\n'
"networks": {
"dsc-InterInstance": [
"172.16.18.15"
],
"dsc-OAM": [
"172.16.16.20"
],
"dsc-sig": [
"172.16.17.15",
"10.10.72.15"
]
}
ie. change every \n to _, grep out the requested block and put the \n back. _ may not be the best choice for a substitute but works in this particular case.

Answer 4

With GNU awk for multi-char RS and RT you can just use a regexp to describe the string you're looking for:
$ awk -v RS='"networks": {[^}]+}' 'RT{print RT}' file
"networks": {
"dsc-InterInstance": [
"172.16.18.15"
],
"dsc-OAM": [
"172.16.16.20"
],
"dsc-sig": [
"172.16.17.15",
"10.10.72.15"
]
}