I'm trying to parse this json so that I can extract all dictionaries from the input where the "uri" field does not start with the string "/themes"
{
"version": 0,
"permissionMask": 2,
"creationDate": "2022-06-09T14:42:41",
"updateDate": "2021-11-17T06:48:52",
"label": "webPageView.css",
"uri": "/themes/default/webPageView.css",
"resourceType": "file"
},
{
"version": 0,
"permissionMask": 2,
"creationDate": "2022-06-09T14:42:43",
"updateDate": "2021-11-17T06:48:48",
"label": "workflow_icons_sprite@05x.png",
"uri": "/internal/default/images/workflow_icons_sprite@05x.png",
"resourceType": "file"
}
so the output would only contain
{
"version": 0,
"permissionMask": 2,
"creationDate": "2022-06-09T14:42:43",
"updateDate": "2021-11-17T06:48:48",
"label": "workflow_icons_sprite@05x.png",
"uri": "/internal/default/images/workflow_icons_sprite@05x.png",
"resourceType": "file"
}
I've tried
jq -r '.[] | select( .[].uri | ( startswith( "/themes" ) | not ) )'
but that doesn't seem to work. It simply dumps all the output without filtering out the dictionaries where the uri field starts with "/themes"
Assuming that the objects are in an array
.[] | select(.uri | startswith( "/themes" ) | not)
Gives
{
"version": 0,
"permissionMask": 2,
"creationDate": "2022-06-09T14:42:43",
"updateDate": "2021-11-17T06:48:48",
"label": "workflow_icons_sprite@05x.png",
"uri": "/internal/default/images/workflow_icons_sprite@05x.png",
"resourceType": "file"
}
Regarding your filter, there's no need for the .[].uri
inside select()
since the first .[]
already loops over the array, so .
inside the select()
refers to the object itself.