Format strings to JSON with jq

Question

Might be easy enough for someone who has experience with jq. I am novice. I am learning jq with bash to create json arrays. I want to generate json array with jq in bash.

I have my file named fileForV:

V    Apple is good for you
A    Broccoli is good for you

Note: The space is a tab.

I have:

jq -R '[inputs | 
split("\t") | 
{"FruitOrVeg":.[0],"Good?":.[1]}]'<<EOF fileForV EOF

I get:

[
  {
   "FruitOrVeg": "A",
   "Good?": "Apple is good for you"
  },
  {
   "FruitOrVeg": "V",
   "Good?": "Broccoli is good for you\""
  }
]

Why do I get \"" at the end?

Should be:

[
  {
   "FruitOrVeg": "A",
   "Good?": "Apple is good for you"
  },
  {
   "FruitOrVeg": "V",
   "Good?": "Broccoli is good for you"
  }
]

Remove `inputs |`? This works for me: `jq -R '[split("\t") | {"FruitOrVeg":.[0],"Good?":.[1]}]' fileForV` — Cyrus, Aug 20 '19 at 04:35
I have inputs so I can get one time [ ]. Otherwise it will add [] after each array. — Woods, Aug 20 '19 at 04:56
Thank you! You made me realized that I had a problem with the way I am reading the file. LOL. I simple mistake. — Woods, Aug 20 '19 at 05:08

peak · Answer 1 · 2019-08-20T06:28:12.537

There's something fishy here. If you are using inputs and getting anything like the results as shown, then:

a) you would have to be using the -n command-line option;

b) you would almost certainly be using jq version 1.5 or later, but I've verified that these versions (in conjunction with bash) definitely do not produce the extraneous double-quotation mark.

Here is a typical transcript, using the input as shown but with one tab per line:

$ jq-1.5 -n -R '
[inputs | split("\t") | {"FruitOrVeg":.[0],"Good?":.[1]}]' input.tsv
[
  {
    "FruitOrVeg": "V",
    "Good?": "Apple is good for you"
  },
  {
    "FruitOrVeg": "A",
    "Good?": "Broccoli is good for you"
  }
]

Format strings to JSON with jq

1 Answers1