Step Function: Edit map output

Question

I have a step function which iterates over several data, at the end of the map I got a single json with all iterations and then uploads as json file to S3 bucket. This file on S3 I will need to use it later to run a Copy command on redshift, but I am having the issue that the output file comes as an array and redshift does not recognize the format
For example this is the input for the step function:

```
[
    {
      "data": { ... }
    },
    {
      "data": { ... }
},
    {
      "data": { ... }
}
]
```
 Then I include some data and this is to output being uploaded to S3

```
[
    {
      "executiontime":"....", 
       "id":"...", 
      "data": { ... }
    },
    {
        "executiontime":"....", 
       "id":"...", 
      "data": { ... }
},
    {
       "executiontime":"....", 
       "id":"...", 
      "data": { ... }
}
]
```
However for redshift I need an output like this, without the square brackets and the comma between items, I have tried different ways but I always get the same output as an array

```
    {
      "executiontime":"....", 
       "id":"...", 
      "data": { ... }
    }
    {
        "executiontime":"....", 
       "id":"...", 
      "data": { ... }
}
    {
       "executiontime":"....", 
       "id":"...", 
      "data": { ... }
}
```

Osvaldo Marte · Answer

If your data is in the format of a JSON array, as shown in your example, you'll need to transform it into a line-delimited JSON format (JSONL) before using the `COPY` command with the `json 'auto'` option in Redshift. This is because Redshift expects each line in the file to be a separate JSON object.

To transform your JSON array into a line-delimited JSON format, you can use a text editor or a command-line tool like `jq`. Here's how you can do it using `jq`:

1. Save your JSON array to a file, for example, `input.json`.

2. Run the following command to convert the JSON array into a line-delimited JSON format:

```bash
   jq -c '.[]' input.json > output.jsonl
   ```

The transformed output should be in the following JSON Lines (JSONL) format:
```json
{"executiontime":"....", "id":"...", "data": {...}}
{"executiontime":"....", "id":"...", "data": {...}}
{"executiontime":"....", "id":"...", "data": {...}}

```

3. Use the `output.jsonl` file with the Redshift `COPY` command:

```sql
   COPY your_table_name
   FROM 's3://your-bucket-name/output.jsonl'
   IAM_ROLE 'your-iam-role'
   JSON 'auto';
   ```

In this example, `jq -c '.[]' input.json` reads the JSON array from `input.json`, and for each element in the array, it outputs a compact JSON representation (`-c` flag) on a new line. The result is redirected to `output.jsonl`, which is then used in the Redshift `COPY` command.

> You can integrate the following steps into your Step Function to automatically generate the `output.jsonl` file

Step Function: Edit map output

Relevant content