How to insert a new line with content in front of each line in json?

Question

I think sed should be the command to do this, but I haven't figured out the proper command yet.

My json file looks like this :

{"LAST_MODIFIED_BY":"david","LAST_MODIFIED_DATE":"2018-06-26 12:02:03.0","CLASS_NAME":"/SC/Trade/HTS_CA/1234abcd","DECISION":"AGREE","TASK_TYPE_ID":"abcd1234-832b-43b6-afa6-361253ffe1d5","NAME":"something"}
{"LAST_MODIFIED_BY":"sarah","LAST_MODIFIED_DATE":"2018-08-26 12:02:03.0","CLASS_NAME":"/SC/Import/HTS_US/9876abcd","DECISION":"DISAGREE","TASK_TYPE_ID":"abcd1234-832b-43b6-afa6-361253ffe1d5","NAME":"nicename"}
... more rows to follow

what I wanted to achieve is a json document with below contents:

{"index":{}}
{"LAST_MODIFIED_BY":"david","LAST_MODIFIED_DATE":"2018-06-26 12:02:03.0","CLASS_NAME":"/SC/Trade/HTS_CA/1234abcd","DECISION":"AGREE","TASK_TYPE_ID":"abcd1234-832b-43b6-afa6-361253ffe1d5","NAME":"something"}
{"index":{}}
{"LAST_MODIFIED_BY":"sarah","LAST_MODIFIED_DATE":"2018-08-26 12:02:03.0","CLASS_NAME":"/SC/Import/HTS_US/9876abcd","DECISION":"DISAGREE","TASK_TYPE_ID":"abcd1234-832b-43b6-afa6-361253ffe1d5","NAME":"nicename"}
... more rows to follow

so that I could run bulk load API against Elasticsearch.

The closest one is this one: Elasticsearch Bulk JSON Data, but it split my json file into broken items instead of my desired format.

Any ideas how I can achieve this would be greatly appreciated!

peak · Accepted Answer · 2021-05-23T09:50:21.053

2

Using `sed`:

     sed 's/^/{"index":{}}\
/'

The trick here is the \.

Alternatively, if your shell supports it:

sed $'s/^/{"index":{}}\n/'

or (as per @sundeep's suggestion):

sed $'i\\\n{"index":{}}\n'

Using jq:

jq -nc 'inputs | {"index":{}}, .'

Here, the key is the -c option to produce JSONLines.

Using awk:

awk '{print "{\"index\":{}}"; print;}'

Etc.

edited May 23 '21 at 09:50

answered May 22 '21 at 14:34

peak

105,803
17
152
177

thanks but I didn't completely follow, say my input is this file: `data1.json`, my output is this name: `data2.json`, how will this command look like? thanks – Fisher Coder May 22 '21 at 14:38
Both sed and jq accept an input file as argument, so you could use the pattern: `COMMAND data1.json > data2.json` – peak May 22 '21 at 14:40
Could also use the `i` command instead of `s` command – Sundeep May 22 '21 at 14:43
perfect, this command worked: `jq -nc 'inputs | {"index":{}}, .' data2.json > data3.json` thanks a ton! – Fisher Coder May 22 '21 at 14:47

score 1 · Answer 2 · answered May 22 '21 at 16:46

1

This might work for you (GNU sed):

sed 'i{"index":{}}' file

Insert {"index":{}} before each line.

answered May 22 '21 at 16:46

potong

55,640
6
51
83

How to insert a new line with content in front of each line in json?

2 Answers2

Using sed:

Using jq:

Using awk:

Using `sed`: