How to export data to csv file using JQ

Question

JSON file example.

{
  "entities": [
    {
      "id": "d130",
      "name": "Abdoulaye, Aminata",
      "division": {
        "id": "d130",
        "name": "",
        "selfUri": "/api/v2/authorization/"
      },
      "chat": {
        "jabberId": "62a0e0b060"
      },
      "email": "Aminata.Abdoulaye@",
      "primaryContactInfo": [
        {
          "address": "Aminata.Abdoulaye@",
          "mediaType": "EMAIL",
          "type": "PRIMARY"
        }
      ],
      "addresses": [
        {
          "address": "Aminata.Abdoulaye@",
          "mediaType": "EMAIL",
          "type": "WORK"
        }
      ],

When converting JSON to csv, I have managed to create a csv file containing the data but there is no headers being displayed in the csv file.

gc --profile NewProfile users list -a --pageSize 100 | jq-win64 -r -c ".[1,2,3] | [.username,.addresses[].address,.id,.title]| @csv" > test.csv

I have managed to create a csv file that contains the headers but no data.

gc --profile NewProfile users list -a --pageSize 100 | jq-win64 -r ".? |(map(keys) | add | unique) as $cols | map(. as $row | $cols | map($row[.])) as $rows | $cols, $rows[] | @csv" > test.csv

Then when i try to merge the two together, I get an error saying, string ("Waled.Abdu...) has no keys

gc --profile NewProfile users list -a --pageSize 100 | jq-win64 -r ".[1,2,3] | [.username,.addresses[].address,.id,.title] |(map(keys) | add | unique) as $cols | map(. as $row | $cols | map($row[.])) as $rows | $cols, $rows[] | @csv" > test.csv

jq: error (at <stdin>:39688): string ("Waled.Abdu...) has no keys

Waled's details in JSON are shown below.

{
      "id": "b143",
      "name": "Abdulkadir, Waled",
      "division": {
        "id": "b143",
        "name": "",
        "selfUri": "/api/v2/authorization/"
      },
      "chat": {
        "jabberId": "63e6e"
      },
      "department": "Canada Solutions",
      "email": "Waled.Abdulkadir@",
      "primaryContactInfo": [
        {
          "address": "Waled.Abdulkadir@",
          "mediaType": "EMAIL",
          "type": "PRIMARY"
        }
      ],
      "addresses": [],
      "state": "active",
      "title": "Engineer",
      "username": "Waled.Abdulkadir@",
      "manager": {
        "id": "f3cc8f83-14c0",
        "selfUri": "/api/v2/users/14c0"
      },
      "version": 4,
      "acdAutoAnswer": false,
      "selfUri": "/api/v2/users/14c0"
    },

Any suggestions would be greatly appreciated. Thanks

Answer 1

I'd go with something like:

jq -r '["username", "address", "id", "title"],
   (.entities[]
    | (try .addresses[].address // null) as $a
    | [.username, $a, .id, .title] ) 
   | @csv'

To circumvent shell quoting issues, consider using the -f command-line option.