OpenTelemetry

comon_logger_otel connects comon_logger to comon_otel.

Use it when the app already emits traces or metrics through OpenTelemetry and logs should go to the same collector or backend.

What the package does

• implements OtelLogHandler as a normal LogHandler
• converts LogRecord into comon_otel log records
• maps LogLevel to OpenTelemetry SeverityNumber
• preserves structured metadata such as layer, type, feature, and extra
• automatically attaches the active span context from OtelContext

That last point is the main value: logs written inside an active trace are exported with the same trace and span identifiers.

Install

dart pub add comon_logger_otel

For Flutter:

flutter pub add comon_logger_otel

You also need a configured comon_otel SDK instance.

Minimal setup

import 'package:comon_logger/comon_logger.dart';
import 'package:comon_logger_otel/comon_logger_otel.dart';
import 'package:comon_otel/comon_otel.dart';

Future<void> main() async {
  await Otel.init(
    serviceName: 'catalog-api',
    exporter: OtelExporter.console,
  );

  Logger.root.addHandler(ConsoleLogHandler());
  Logger.root.addHandler(OtelLogHandler());

  final log = Logger('catalog.service');
  log.info('Application started', feature: 'catalog');
}

Severity mapping

comon_logger_otel keeps the comon_logger level model but exports compatible OpenTelemetry severities.

Exported attributes

Each emitted log carries the original message body plus structured attributes.

Nested extra maps are flattened recursively so observability backends can filter on individual fields.

Filtering

Because OtelLogHandler is just another LogHandler, you can keep one noise threshold for local console output and a different threshold for exported logs.

Logger.root.addHandler(
  OtelLogHandler(
    filter: const LevelLogFilter(LogLevel.WARNING),
  ),
);

Trace correlation

If a log is emitted inside an active OpenTelemetry span, comon_logger_otel exports the matching trace and span IDs automatically.

await Otel.instance.tracer.traceAsync(
  'create-user',
  fn: () async {
    final log = Logger('user.service');
    log.info('Validating payload');
    log.info('User created', feature: 'signup');
  },
);

In the backend this means the logs can be displayed directly on the trace for create-user instead of being a disconnected stream.

Recommended pairing

The package becomes more useful when paired with other instrumented integrations:

• OpenTelemetry Dio for outbound HTTP spans
• ORM OpenTelemetry for database spans and metrics
• OpenTelemetry core for exporter and processor configuration

When to use it

Use comon_logger_otel when:

• logs should land in OTLP-compatible backends such as Grafana, Tempo, Loki pipelines, or vendor observability stacks
• trace and log correlation matters more than local-only formatting
• you already standardized on comon_otel in the service or app

Do not use it as a replacement for local developer ergonomics. Keep ConsoleLogHandler or Flutter history handlers when local inspection is still part of the workflow.