How do I add a comma in between JSON objects? - arrays

Question

Say I have a sample.json file that looks like this. Notice the objects are not separated by a comma. I am aware that this is not a JSON file without the commas.
[
{
"type": "message",
"user": "U024HFHU5",
"text": "hey there",
"ts": "1385407681.000003"
}
{
"type": "message",
"user": "U024HGJ4E",
"text": "right back at you",
"ts": "1385407706.000006"
}
]
How do I make it look like this (example from I got the example from https://thoughtbot.com/blog/jq-is-sed-for-json.)
[
{
"type": "message",
"user": "U024HFHU5",
"text": "hey there",
"ts": "1385407681.000003"
},
{
"type": "message",
"user": "U024HGJ4E",
"text": "right back at you",
"ts": "1385407706.000006"
}
]
I tried doing },{ as some have suggested but it does not work.
I think it's probably something like this in jq on the command line but am not sure. Here, my hope is that sample2.json outputs the desired format.
jq -r '. | join(", ")' sample.json > sample2.json

Answer 1

If the square brackets were not in the input, jq could process it like normal as a stream of objects. You could then throw them into an array or whatever.
You'll just need to invoke jq twice in this case. First read it in raw (-Rs) and trim out the leading and trailing brackets, output it raw (-r), then process in jq like normal.
$ jq -Rsr 'gsub("^\\s*\\[|\\]\\s*$"; "")' input.json | jq -n '[inputs]'

Answer 2

Using human JSON utility to recover your malformed JSON file:
hjson -j file
[
{
"type": "message",
"user": "U024HFHU5",
"text": "hey there",
"ts": "1385407681.000003"
},
{
"type": "message",
"user": "U024HGJ4E",
"text": "right back at you",
"ts": "1385407706.000006"
}
]
The option -j outputs the file with correct JSON syntax, by adding the missing comma between objects.