jq flattening JSON arrays

Question

This

echo '{"a":[{"b":[{"c":"xxx"}]},{"b":[{"c":"yyy"},{"c":"zzz"}]}]}' |
    jq '.a[].b | map({"c": .c})'

produces this:

[
  {
    "c": "xxx"
  }
]
[
  {
    "c": "yyy"
  },
  {
    "c": "zzz"
  }
]

How do I get a single output array like:

[
  {
    "c": "xxx"
  },
  {
    "c": "yyy"
  },
  {
    "c": "zzz"
  }
]

Answer 1

$ jq [.[][][][]] <<< "$JSON"
[
  {
    "c": "xxx"
  },
  {
    "c": "yyy"
  },
  {
    "c": "zzz"
  }
]

OR:

$ jq '[.. | select(has("c")?)]' <<< "$JSON"
[
  {
    "c": "xxx"
  },
  {
    "c": "yyy"
  },
  {
    "c": "zzz"
  }
]

Answer 2

You can use this:

jq '[{"c":(.a[].b[].c)}]' file.json

Answer 3

This does it:

jq '[ .a[][][] ]'