CODEWITHSUNDEEP
unixawksedgrep
Jim Macaulay
Jim Macaulay

Reputation: 5141

Grep string in between two values Unix

I have a json data as below. Need to grep the value and export it into a new file.

{
"Data":
[
"User": [
      {"Name": "Solomon", "Age":20}, 
      {"Name": "Absolom", "Age":30}, 
   ] 
"Country": [
       {"Name" : "US", "Resident" : "Permanent"},
       {"Name" : "UK", "Resident" : "Temporary"}
]]}

Expected result,

Need to grep the value of "User" and export it to a new file User.json

User.json

"User": [
        {"Name": "Solomon", "Age":20}, 
        {"Name": "Absolom", "Age":30}, 
   ]

And again need to grep "Country" and export to Country.json

Country.json

"Country": [
       {"Name" : "US", "Resident" : "Permanent"},
       {"Name" : "UK", "Resident" : "Temporary"}
]

Was trying with sed, but am getting incorrect results

sed -e 's/.*"User"\(.*\)].*/\1/' Data.json > Users.json

Since there are many special characters ]/[/{/} , not sure how to properly grep a value and export it to a new file.

Any suggestion would be really helpful

Upvotes: 0

Views: 116

Answers (2)

nbari
nbari

Reputation: 26925

Your input is not a valid JSON, however, for parsing JSON you could use jq, for example, modifying your input a little bit, give a try to this:

 $ echo  '{
   "Data":[
      {
         "User":[
            {
               "Name":"Solomon",
               "Age":20
            },
            {
               "Name":"Absolom",
               "Age":30
            }
         ],
         "Country":[
            {
               "Name":"US",
               "Resident":"Permanent"
            },
            {
               "Name":"UK",
               "Resident":"Temporary"
            }
         ]
      }
   ]
}'| jq  -cr .'Data[] | with_entries(select(.key == "User"))';

It will output (for user):

{"User":[{"Name":"Solomon","Age":20},{"Name":"Absolom","Age":30}]}

And if you use | jq -cr .'Data[] | with_entries(select(.key == "Country"))'; (for country)

{"Country":[{"Name":"US","Resident":"Permanent"},{"Name":"UK","Resident":"Temporary"}]}

This is only one way of many possible, but I hope it can give you an idea.

Upvotes: 0

KamilCuk
KamilCuk

Reputation: 141040

With enough luck you can use GNU sed with -z option and filter the content up until a closing ] assuming there will be no ] in the strings:

sed -z 's/.*\("User":[^]]*]\).*/\1\n/'
sed -z 's/.*\("Country":[^]]*]\).*/\1\n/'

Upvotes: 1

Related Questions