JSON到CSV使用JQ

Question

Looking to convert the following json example into csv using jq. The work is being performed on a centos 6.9, i am able to add tools to the box if need. The purpose of normalizing the json into csv is for a SIEM feed.

JSON:

{
  "Key1": "Value1",
  "Key2": "Value2",
  "Key3": "Value3",
  "key4": 4,
  "key5": 5,
  "key6": "value6"
}
{
  "Key1": "Value1",
  "Key2": "Value2",
  "Key3": "Value3",
  "key4": 4,
  "key5": 5,
  "key6": "value6"
}
{
  "Key1": "Value1",
  "Key2": "Value2",
  "Key3": "Value3",
  "key4": 4,
  "key5": 5,
  "key6": "value6"
}

Output expecting:

"key1":"Value1","key2":"Value2","key3":"Value3","key4":4,"key5":5,"key6":"Value6"
"key1":"Value1","key2":"Value2","key3":"Value3","key4":4,"key5":5,"key6":"Value6"
"key1":"Value1","key2":"Value2","key3":"Value3","key4":4,"key5":5,"key6":"Value6"

Answer 1

First, that's not csv. The quotes would be interpreted as containing the single value. You have multiple sets of quotes per comma-separated value. But if that's what you want as output, you can definitely generate it, just know that it's not real csv.

$ jq -r 'to_entries | map("\(.key|tojson):\(.value|tojson)") | join(",")' input.json

Answer 2

Here is a filter which generates the expected output (assuming data in data.json ):

$ jq -Mr '[keys[] as $k | {($k):.[$k]} | tojson[1:-1]] | join(",")' data.json
"Key1":"Value1","Key2":"Value2","Key3":"Value3","key4":4,"key5":5,"key6":"value6"
"Key1":"Value1","Key2":"Value2","Key3":"Value3","key4":4,"key5":5,"key6":"value6"
"Key1":"Value1","Key2":"Value2","Key3":"Value3","key4":4,"key5":5,"key6":"value6"

Try it online at jqplay.org