fmt.Println("Got result", resp)
        // Return the response for reuse
		instance.PutResponse(resp)
        return nil
    })
```

There are handlers for requests with:
 * No input stream: `RegisterNoInput`.
 * No initial payload: `RegisterNoPayload`.

Note that Responses sent for serialization are automatically reused for similar requests.

# Ensure bucket exists, purging if asked to
createBucket() {
	BUCKET=$1
	POLICY=$2
	PURGE=$3
	VERSIONING=$4
	OBJECTLOCKING=$5

	# Purge the bucket, if set & exists
	# Since PURGE is user input, check explicitly for `true`
	if [ $PURGE = true ]; then
		if checkBucketExists $BUCKET; then
			echo "Purging bucket '$BUCKET'."
			set +e # don't exit if this fails
			${MC} rm -r --force myminio/$BUCKET

- Fine grained filtering is available to pick relevant objects from source to copy from

MinIO batch jobs framework also provides

- Retrying a failed job automatically driven by user input
- Monitoring job progress in real-time
- Send notifications upon completion or failure to user configured target

Following YAML describes the structure of a replication job, each value is documented and self-describing.

```sh
cat > example.rego <<EOF
package httpapi.authz

import input

default allow = false

# Allow the root user to perform any action.
allow {
 input.owner == true
}

# All other users may do anything other than call PutObject
allow {
 input.action != "s3:PutObject"
 input.owner == false
}
EOF
```

Then load the policy via OPA's REST API.

```

Add the following optimal entries for `hive-site.xml` to configure Hive with **MinIO**.

```
hive.blobstore.use.blobstore.as.scratchdir=true
hive.exec.input.listing.max.threads=50
hive.load.dynamic.partitions.thread=25
hive.metastore.fshandler.threads=50
hive.mv.files.threads=40
mapreduce.input.fileinputformat.list-status.num-threads=50
```

(*) Parquet is disabled on the MinIO server by default. See below how to enable it.

## Enabling Parquet Format

Parquet is DISABLED by default since hostile crafted input can easily crash the server.

If you are in a controlled environment where it is safe to assume no hostile content can be uploaded to your cluster you can safely enable Parquet.

		object_context = event["getObjectContext"]

		# Get the presigned URL to fetch the requested
		# original object from MinIO
		s3_url = object_context["inputS3Url"]

		# Extract the route and request token from the input context
		request_route = object_context["outputRoute"]
		request_token = object_context["outputToken"]

		# Get the original S3 object using the presigned URL
		r = requests.get(s3_url)

The JSON body structure can be seen from this sample:

<details><summary>Request Body Sample</summary>

```json
{
  "input": {
    "account": "minio",
    "groups": null,
    "action": "s3:ListBucket",
    "bucket": "test",
    "conditions": {
      "Authorization": [

Such objects do not produce efficient [`LZ compression`](https://en.wikipedia.org/wiki/LZ77_and_LZ78)
which is a fitness factor for a lossless data compression.

Pre-compressed input typically compresses in excess of 2GiB/s per core,
so performance impact should be minimal even if precompressed data is re-compressed.
Decompressing incompressible data has no significant performance impact.

        if cardinality <= 0 {
                return -1
        }
        sip := siphash.New(id[:])
        sip.Write([]byte(key))
        return int(sip.Sum64() % uint64(cardinality))
}
```