README

PHP client for the Internal Job Queue API (API docs).

Usage

composer require keboola/job-queue-internal-api-php-client

use Keboola\JobQueueInternalClient\Client;

$storageClientFactory = new JobFactory\StorageClientFactory('http://connetion.keboola.com/');
$objectEncryptorFactory = new ObjectEncryptorFactory('key-id', 'us-east-1', '', '');
$jobFactory = new JobFactory($storageClientFactory, $objectEncryptorFactory);
$client = new Client(
    new NullLogger(),
    $jobFactory,
    'http://internal.queue.api/',
    'testQueueToken'
);
$client->getJobData('123');
$client->postJobResult('123', 'success', ['images' => ['digests' => []]]);

Atomic job result updates

patchJobResultAtomically() performs an optimistic-locking read-modify-write of a job's result document. It reads the current result and its resultVersion, runs your mutator, and writes it back guarded by that version. The versioned write merges the returned keys into the current server-side result (the same array_merge semantics as the non-versioned write), only made atomic by the version check. On a version conflict (HTTP 409, i.e. a concurrent writer changed the result in between) it automatically re-reads and retries a bounded number of times; if the conflict still cannot be resolved within the retry budget it throws a ResultVersionConflictException rather than overwriting the concurrent change.

The mutator receives the current result array and must return a \JsonSerializable whose jsonSerialize() yields an array. The returned keys are merged into the current result — it is not a full replace, so keys you omit are preserved (you cannot delete a key this way):

use Keboola\JobQueueInternalClient\Result\GenericJobResult;

$job = $client->patchJobResultAtomically('123', function (array $current): GenericJobResult {
    $current['processedRows'] = ($current['processedRows'] ?? 0) + 1;
    return new GenericJobResult($current);
});

GenericJobResult is a trivial \JsonSerializable wrapper for callers that do not have their own result value object.

Development

Prerequisites:

• configured az and aws CLI tools (run az login and aws configure --profile keboola-dev-platform-services)
• installed GCP CLI gcloud (and run gcloud auth login or gcloud auth application-default login)
• installed terraform (https://www.terraform.io) and jq (https://stedolan.github.io/jq) to setup local env
• intalled docker and docker compose to run & develop the app

TL;DR:

export NAME_PREFIX= # your name/nickname to make your resource unique & recognizable

cat <<EOF > ./provisioning/local/terraform.tfvars
name_prefix = "${NAME_PREFIX}"
EOF

cat <<EOF > .env.local
TEST_HOSTNAME_SUFFIX=keboola.com
TEST_STORAGE_API_URL=https://connection.keboola.com
TEST_STORAGE_API_TOKEN=
TEST_STORAGE_API_TOKEN_MASTER=
EOF

terraform -chdir=./provisioning/local init -backend-config="key=job-queue-internal-api-php-client/${NAME_PREFIX}.tfstate"
terraform -chdir=./provisioning/local apply
./provisioning/local/update-env.sh azure # or aws

License

MIT licensed, see LICENSE file.