t3n.FlowLog

This package adds a few helper to log Flow Messages.

🔧 Still Work in Progress 🔧

ConsoleStorage

If you'd like to log your Exceptions to the console, for instance stderr or stdout, he ConsoleStorage is for you. It will log all throwables directly to the console as JSON. The JSON is formatted to be parsed by google Stackdriver.

To enable the ConsoleStorage you need to adjust your Settings.yaml like this:

Neos:
  Flow:
    log:
      throwables:
        storageClass: t3n\FlowLog\ThrowableStorage\ConsoleStorage
        optionsByImplementation:
          't3n\FlowLog\ThrowableStorage\ConsoleStorage':
            streamName: 'stderr'

The StreamName could either be stderr or stdout.

BigQueryLogger

The BigQueryLogger is an AbstractBackend so that you can use it like an normal Logger like FileBackend. If you want to log into BigQuery, here is your 3 Step manual:

• configure BigQuery (dataset, table, keyFile)
• create a BigQueryLogger via PsrLoggerFactory
• map your custom BigQueryLogger via Objects.yaml to actually use it

1) Configure BigQueryLogger

t3n:
  FlowLog:
    bigQuery:
      dataset: 't3n_flowlog'
      table: 'application_log'
      expirationMs: '7776000000' # 90 days
      keyFilePath: '/path/to/google/key.json'

ℹ️ All logs will be written into a partitioned table in BigQuery. This means that you have a "overall" table with multiple tables for each day. This day-tables can be deleted automatically via expirationMs. If you want to store your logs forever just ignore this Setting and let it blank.

2) Create your own Instance of BigQueryLogger

To actually use the BigQueryLogger you have to define your own in your Settings.yaml. The important part is loggerName ("applicationXyImportLogger") and the internal Name bigQueryLogger.

bigQueryLogger will be used in Step 3, loggerName is the actual name for the BigQueryTable (inserted for each row).

Neos:
  Flow:
    log:
      psr3:
        'Neos\Flow\Log\PsrLoggerFactory':
          bigQueryLogger:
            class: 't3n\FlowLog\Backend\BigQueryLogger'
            options:
              loggerName: 'applicationXyImportLogger'

You can create as many loggers as you want. This is really usefull if you want to "split" your logs in your BigQueryTable. E.g. one Logger for Imports, one for User-Requests, etc.

3) Map your BigQueryLogger via Objects.yaml

Last but not least you have to define your Logging-Factory to use your BigQueryLogger.

Therefore edit your Objects.yaml like this:

t3n\FlowLog\Command\ExampleCommandController: # <- adjust
  properties:
    logger:
      object:
        factoryObjectName: Neos\Flow\Log\PsrLoggerFactoryInterface
        factoryMethodName: get
        arguments:
          1:
            value: bigQueryLogger

Now you are able to use it:

/**
 * @var Psr\Log\LoggerInterface
 */
protected $logger;

....

$this->logger->log(LogLevel::INFO, 'First log entry.', ['test' => true]);

ServiceContext

ℹ️ Important if you want to log multiple applications/environments (e.g.) in BigQuery.

You should also set the ServiceContext that is used by StackDriver and BigQueryLogger:

t3n:
  FlowLog:
    serviceContext:
      service: 'flow-app'
      version: 'master'