How can I flatten this json array and write out to a csv file using jq so that each row contains all the fields i.e.
I'm a bit of a novice with jq but started with this, which does not capture the inner nested array
jq -r '(.[0] | keys_unsorted) as $keys | $keys, map([.[ $keys[] ]])[]'
[
{
"name":"PM2.5",
"unit":"ug/m3",
"values":[
{
"timestamp":"2020-11-16T13:10:21.135Z",
"value":3.696
},
{
"timestamp":"2020-11-16T13:10:31.649Z",
"value":3.696
},
{
"timestamp":"2020-11-16T13:10:42.35Z",
"value":3.696
},
{
"timestamp":"2020-11-16T13:10:52.953Z",
"value":3.696
}
]
},
{
"name":"PM10",
"unit":"ug/m3",
"values":[
{
"timestamp":"2020-11-16T13:10:21.135Z",
"value":7.596
},
{
"timestamp":"2020-11-16T13:10:31.649Z",
"value":7.596
},
{
"timestamp":"2020-11-16T13:10:42.35Z",
"value":7.596
},
{
"timestamp":"2020-11-16T13:10:52.953Z",
"value":7.596
}
]
}
]
To a csv as follows:
"name", "unit", "timestamp", "value"
"PM2.5", "ug/m3", "2020-11-16T13:10:21.135Z", 3.696
"PM2.5", "ug/m3", "2020-11-16T13:10:31.649Z", 3.696
"PM2.5", "ug/m3", "2020-11-16T13:10:42.35Z", 3.696
"PM2.5", "ug/m3", "2020-11-16T13:10:52.953Z", 3.696
"PM10", "ug/m3", "2020-11-16T13:10:21.135Z", 7.596
"PM10", "ug/m3", "2020-11-16T13:10:31.649Z", 7.596
"PM10", "ug/m3", "2020-11-16T13:10:42.35Z", 7.596
"PM10", "ug/m3", "2020-11-16T13:10:52.953Z", 7.596
$ jq -r '
[["name","unit","timestamp","value"]],
[ .[]
| [ .name, .unit ] +
(.values[] | [.timestamp,.value]) ]
| .[]
| @csv' input.json
"name","unit","timestamp","value"
"PM2.5","ug/m3","2020-11-16T13:10:21.135Z",3.696
"PM2.5","ug/m3","2020-11-16T13:10:31.649Z",3.696
"PM2.5","ug/m3","2020-11-16T13:10:42.35Z",3.696
"PM2.5","ug/m3","2020-11-16T13:10:52.953Z",3.696
"PM10","ug/m3","2020-11-16T13:10:21.135Z",7.596
"PM10","ug/m3","2020-11-16T13:10:31.649Z",7.596
"PM10","ug/m3","2020-11-16T13:10:42.35Z",7.596
"PM10","ug/m3","2020-11-16T13:10:52.953Z",7.596
Builds an array of arrays, one per row including a header line, and then formats them all as CSV records.