jq - Extra space when outputting to CSV after jq filter

Question

Facing an issue with a simple requirement.

1. Make a CURL request and store in a variable
2. Parse JSON from variable using JQ and output into CSV file.

When I try to do it, I am getting extra leading space in the file from the second line for some reason. Is there a way to configure jq to not do this ?

Code :

output=`curl -v -k -H 'Content-Type: application/json' '<<URL>>' -d '<<PARAMS>>'`
csvOutput=`echo $output | jq --raw-output '.result .items[]|[.name,.key,.created,.updated]|@csv'`
echo $csvOutput >> out.csv

The following code (all in one line) does not have the issue. But I need the curl output in a variable to perform some checks ,so this option does not help.

echo "$(curl -v -k -H 'Content-Type: application/json' '<<URL>>' -d '<<PARAMS>>" | jq -r '.result .items[]|[.name,.key,.created,.updated]|@csv' >> out.csv

Strangely, when I read the json output (taken from curl response) from a file and then apply jq filters, I don't see the issue.

cat response.json | jq -r '.result .items[]|[.name,.key,.created,.updated]|@csv' > out.csv

response.json (formatted) :

{
  "result": {
    "total_items": 22,
    "total_pages": 1,
    "items_per_page": 1000,
    "current_page": 1,
    "items": [
      {
        "key": "1",
        "name": "Name 1",
        "created": "2016-12-20T08:51:13Z",
        "updated": "2016-12-20T09:29:08Z"
      },
      {
        "key": "2",
        "name": "Name 2",
        "created": "2016-12-20T08:51:13Z",
        "updated": "2016-12-20T09:29:08Z"
      },
      {
        "key": "3",
        "name": "Name 3",
        "created": "2016-12-20T08:51:13Z",
        "updated": "2016-12-20T09:29:08Z"
      }
    ]
  },
  "id": 1
}

Answer 1

When you write your result to your csv file your are using echo $csvOutput >> out.csv instead of :

echo "$csvOutput" >> out.csv 

The double quotes will preserve newlines, see this post