Convert JSON to CSV with jq: A Step‑by‑Step Tutorial

Part 1: Extract Data

We start by reading FunTester.json and filtering the article array using the jq filter '.article[]'. This returns each object in the array as separate JSON objects rather than a single JSON array.

cat FunTester.json | jq '.article[]'
{
  "author": "tester1",
  "title": "ApiTest"
}
{
  "author": "tester2",
  "title": "performanceTest"
}

If we omit the [] and use '.article', jq returns the whole array wrapped in brackets.

cat FunTester.json | jq '.article'
[
  {
    "author": "tester1",
    "title": "ApiTest"
  },
  {
    "author": "tester2",
    "title": "performanceTest"
  }
]

Part 2: Assemble Data

Next we transform each JSON object into a two‑element array containing the author and title fields. The filter '.article[] | [.author, .title]' produces a stream of arrays.

cat FunTester.json | jq '.article[] | [.author, .title]'
[
  "tester1",
  "ApiTest"
]
[
  "tester2",
  "performanceTest"
]

The resulting arrays can later be joined into CSV rows.

Part 3: Output CSV

Finally we apply the @csv operator to format each array as a CSV line. By default jq escapes double quotes, so the output looks like "tester1","ApiTest". To obtain raw CSV without the extra escaping, we add the -r (raw output) flag.

# With default escaping
cat FunTester.json | jq '.article[] | [.author, .title] | @csv'
"\"tester1\",\"ApiTest\""
"\"tester2\",\"performanceTest\""

# Raw CSV output
cat FunTester.json | jq -r '.article[] | [.author, .title] | @csv'
"tester1","ApiTest"
"tester2","performanceTest"