0

I was trying to write a query in linux for alternate a json file and write it in a new file.

My json file is formatted like so:

{"intents": [
  {
    "patterns": "'For the last 8 years of his life, Galileo was under house arrest for espousing this man's theory'",
    "responses": "Copernicus"
  },
  {
    "patterns": "'No. 2: 1912 Olympian; football star at Carlisle Indian School; 6 MLB seasons with the Reds, Giants & Braves'",
    "responses": "Jim Thorpe"
  },

for about 200k entries like the above.

The command that I execute is bellow:

jq --argjson r \
  "$( jo tag = "$(curl -Ss "https://www.random.org/integers/?num=1&min=0&max=1000000&col=1&base=10&format=plain&rnd=new")")" \
  '.intents[] += $r' \
< intents7.json > intents_super.json

i was wanted to add a new keyname in the list as a tag name and i wanted to fill the key(the tag) for every entry with a random number. The command was executed but i am waiting like 30 mins so far and nothing is outputting in the file intents_super.json.

Note: The cpu is on constant 100% also in terminal i was getting those 2 lines, the command still running though..:

Argument `tag' is neither k=v nor k@v
Argument `17208' is neither k=v nor k@v

Does the command do what I want?

2 Answers 2

0

Assuming you want to add a new key, tag, to each of the elements of the top-level intents array with a random number as value, then you may do this like so:

some-command | jq -n 'input | .intents |= map(.tag = input)' intents7.json -

... where some-command is a command that generates an endless stream of random numbers, one per line (for example, shuf -i 0-100 -r or jot -r 0 or similar).

The jq command is used with its -n option to turn off the normal reading of the input. Instead, input is used to fetch the next object.

The first object is read from the file intents7.json, and it contains the intents array that we want to modify. We modify each element of the array by using map(). The expression that we map onto each element adds a tag key with the value read using input. Each call to input (after the very initial one) reads from the standard input stream of jq. On that stream, there are random numbers available, and these are used for the values of the new tag keys.

Example run (using a corrected variant of the data in the question):

$ shuf -i 0-1000 -r | jq -n 'input | .intents |= map(.tag = input)' intents7.json -
{
  "intents": [
    {
      "patterns": "'For the last 8 years of his life, Galileo was under house arrest for espousing this man's theory'",
      "responses": "Copernicus",
      "tag": 517
    },
    {
      "patterns": "'No. 2: 1912 Olympian; football star at Carlisle Indian School; 6 MLB seasons with the Reds, Giants & Braves'",
      "responses": "Jim Thorpe",
      "tag": 955
    }
  ]
}
1
  • thats actually better!! Thanks! and yes i used the dev random Jun 22 at 11:21
0

I fixed my problem by another way:

od -t x -An /dev/random | tr -d " " | fold -w 8 | jq -nRc --slurpfile stream intents7.json '$stream[] .intents[] | .tag=input' > intensfinalllllll.json
1

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.