Send data from Flutter app

This page explains how to send data from Flutter apps to Axiom.

This page explains how to send structured logs from Flutter apps to Axiom using a custom logging library built with the Dio HTTP client.

Prerequisites

Create an Axiom account.
Create a dataset in Axiom where you send your data.
Create an API token in Axiom with permissions to ingest data to the dataset you have created.

Install Flutter SDK version 3.0.0 or higher.

Install required dependencies

To install the required Flutter dependencies, add these lines to your pubspec.yaml file:

YAML

dependencies:
  dio: ^5.4.0
  intl: ^0.19.0

Then run the following code in your terminal to install dependencies:

shell

flutter pub get

dio: A powerful HTTP client for Dart that handles network requests to send logs to Axiom.
intl: Provides internationalization and date/time formatting utilities for creating properly formatted timestamps.

Create axiom_logger.dart file

Create a lib/axiom_logger.dart file with the following content. This file defines the logger configuration, log levels, and the main logging functionality that sends structured logs to Axiom.

The logger implementation below includes the following key features:

Log Levels: Five severity levels (debug, info, warning, error, critical) for categorizing log entries.
Batching: Logs are buffered and sent in batches to reduce network overhead and improve performance.
Immediate Sending: Critical logs are sent immediately to ensure important events are captured right away.
Metadata Support: Attach custom metadata to logs for richer context and easier filtering.
Automatic Timestamps: Logs are automatically timestamped in ISO 8601 format.

dart

import 'dart:convert';
import 'package:dio/dio.dart';
import 'package:intl/intl.dart';

/// Log level enumeration
enum LogLevel {
  debug,
  info,
  warning,
  error,
  critical,
}

/// Extension to convert LogLevel to string
extension LogLevelExtension on LogLevel {
  String get name {
    switch (this) {
      case LogLevel.debug:
        return 'DEBUG';
      case LogLevel.info:
        return 'INFO';
      case LogLevel.warning:
        return 'WARNING';
      case LogLevel.error:
        return 'ERROR';
      case LogLevel.critical:
        return 'CRITICAL';
    }
  }
}

/// Configuration for the Axiom Logger
class AxiomLoggerConfig {
  final String domain;
  final String dataset;
  final String apiToken;
  final Duration timeout;
  final bool enableDebugLogs;

  AxiomLoggerConfig({
    required this.domain,
    required this.dataset,
    required this.apiToken,
    this.timeout = const Duration(seconds: 10),
    this.enableDebugLogs = false,
  });

  String get ingestUrl => '$domain/v1/ingest/$dataset';
}

/// Main Axiom Logger class
class AxiomLogger {
  final AxiomLoggerConfig config;
  late final Dio _dio;
  final List<Map<String, dynamic>> _logBuffer = [];
  final int _batchSize;
  final Duration _flushInterval;
  bool _isInitialized = false;

  AxiomLogger({
    required this.config,
    int batchSize = 10,
    Duration flushInterval = const Duration(seconds: 5),
  })  : _batchSize = batchSize,
        _flushInterval = flushInterval {
    _initializeDio();
  }

  void _initializeDio() {
    _dio = Dio(
      BaseOptions(
        baseUrl: config.domain,
        connectTimeout: config.timeout,
        receiveTimeout: config.timeout,
        headers: {
          'Authorization': 'Bearer ${config.apiToken}',
          'Content-Type': 'application/json',
        },
      ),
    );

    if (config.enableDebugLogs) {
      _dio.interceptors.add(
        LogInterceptor(
          requestBody: true,
          responseBody: true,
          error: true,
          requestHeader: true,
          responseHeader: false,
        ),
      );
    }

    _isInitialized = true;
  }

  /// Log a message with specified level
  Future<void> log(
    LogLevel level,
    String message, {
    Map<String, dynamic>? metadata,
    bool sendImmediately = false,
  }) async {
    if (!_isInitialized) {
      print('AxiomLogger: Logger not initialized');
      return;
    }

    final logEntry = _createLogEntry(level, message, metadata);

    if (sendImmediately) {
      await _sendLogs([logEntry]);
    } else {
      _logBuffer.add(logEntry);
      if (_logBuffer.length >= _batchSize) {
        await flush();
      }
    }
  }

  /// Create a structured log entry
  Map<String, dynamic> _createLogEntry(
    LogLevel level,
    String message,
    Map<String, dynamic>? metadata,
  ) {
    final now = DateTime.now().toUtc();
    final timestamp = DateFormat("yyyy-MM-dd'T'HH:mm:ss.SSS'Z'").format(now);

    return {
      '_time': timestamp,
      'level': level.name,
      'message': message,
      if (metadata != null) ...metadata,
    };
  }

  /// Flush all buffered logs to Axiom
  Future<bool> flush() async {
    if (_logBuffer.isEmpty) {
      return true;
    }

    final logsToSend = List<Map<String, dynamic>>.from(_logBuffer);
    _logBuffer.clear();

    return await _sendLogs(logsToSend);
  }

  /// Send logs to Axiom
  Future<bool> _sendLogs(List<Map<String, dynamic>> logs) async {
    try {
      final response = await _dio.post(
        config.ingestUrl,
        data: jsonEncode(logs),
      );

      if (response.statusCode == 200 || response.statusCode == 204) {
        if (config.enableDebugLogs) {
          print('AxiomLogger: Successfully sent ${logs.length} log(s) to Axiom');
        }
        return true;
      } else {
        print('AxiomLogger: Failed to send logs. Status: ${response.statusCode}');
        return false;
      }
    } catch (e) {
      print('AxiomLogger: Error sending logs to Axiom: $e');
      return false;
    }
  }

  /// Convenience methods for different log levels
  Future<void> debug(String message, {Map<String, dynamic>? metadata}) async {
    await log(LogLevel.debug, message, metadata: metadata);
  }

  Future<void> info(String message, {Map<String, dynamic>? metadata}) async {
    await log(LogLevel.info, message, metadata: metadata);
  }

  Future<void> warning(String message, {Map<String, dynamic>? metadata}) async {
    await log(LogLevel.warning, message, metadata: metadata);
  }

  Future<void> error(String message, {Map<String, dynamic>? metadata}) async {
    await log(LogLevel.error, message, metadata: metadata);
  }

  Future<void> critical(String message, {Map<String, dynamic>? metadata}) async {
    await log(LogLevel.critical, message, metadata: metadata, sendImmediately: true);
  }

  /// Dispose resources
  Future<void> dispose() async {
    await flush();
    _dio.close();
  }
}

Create main.dart file

Create an example/main.dart file with the following content. This file demonstrates how to use the Axiom Logger with different log levels and metadata.

dart

import 'package:flutter_logging/axiom_logger.dart';

/// Example usage of the Axiom Logger
Future<void> main() async {
  print('=== Flutter Axiom Logger Test ===\n');

  // Initialize the logger with your Axiom configuration
  final config = AxiomLoggerConfig(
    domain: 'AXIOM_DOMAIN',
    dataset: 'DATASET_NAME',
    apiToken: 'API_TOKEN',
    enableDebugLogs: true, // Enable to see HTTP requests/responses
  );

  final logger = AxiomLogger(
    config: config,
    batchSize: 5, // Send logs in batches of 5
    flushInterval: Duration(seconds: 3),
  );

  print('Logger initialized. Sending test logs to Axiom...\n');

  try {
    // Test 1: Simple info log
    print('Test 1: Sending INFO log...');
    await logger.info(
      'Application started successfully',
      metadata: {
        'app_name': 'Flutter Logging Demo',
        'version': '1.0.0',
        'environment': 'development',
      },
    );

    // Test 2: Debug log with metadata
    print('Test 2: Sending DEBUG log with metadata...');
    await logger.debug(
      'User authentication flow initiated',
      metadata: {
        'user_id': 'user_12345',
        'session_id': 'session_abc123',
        'ip_address': '192.168.1.100',
      },
    );

    // Test 3: Warning log
    print('Test 3: Sending WARNING log...');
    await logger.warning(
      'High memory usage detected',
      metadata: {
        'memory_mb': 512,
        'threshold_mb': 400,
        'process': 'main_app',
      },
    );

    // Test 4: Error log
    print('Test 4: Sending ERROR log...');
    await logger.error(
      'Failed to connect to database',
      metadata: {
        'error_code': 'DB_CONN_001',
        'database': 'production_db',
        'retry_count': 3,
        'last_error': 'Connection timeout after 30s',
      },
    );

    // Test 5: Multiple logs to test batching
    print('Test 5: Sending multiple logs to test batching...');
    for (int i = 1; i <= 3; i++) {
      await logger.info(
        'Processing batch item $i',
        metadata: {
          'batch_id': 'batch_001',
          'item_number': i,
          'status': 'processing',
        },
      );
    }

    // Test 6: Critical log (sends immediately)
    print('Test 6: Sending CRITICAL log (immediate send)...');
    await logger.critical(
      'System failure: Out of memory',
      metadata: {
        'available_memory_mb': 10,
        'required_memory_mb': 500,
        'action_taken': 'emergency_shutdown',
      },
    );

    // Flush any remaining buffered logs
    print('\nFlushing remaining logs...');
    await logger.flush();

    print('\n✅ All logs sent successfully!');
    print('\nYou can now check your Axiom dashboard at:');
    print('https://app.axiom.co/DATASET_NAME');

  } catch (e) {
    print('❌ Error during logging: $e');
  } finally {
    // Clean up resources
    await logger.dispose();
    print('\nLogger disposed. Test complete.');
  }
}

Run the app and observe logs in Axiom

Run the following code in your terminal to run the Flutter example:
shell
```
dart run example/main.dart
```
The app sends logs with different severity levels to Axiom, demonstrating batching, immediate sending for critical logs, and the use of custom metadata.
In Axiom, go to the Stream tab, and then click your dataset. This page displays the logs sent to Axiom and enables you to monitor and analyze your app's behavior and performance.

Send data from an existing Flutter project

Basic integration

To add Axiom logging to your existing Flutter app, follow these steps:

Add the Axiom Logger to your project by copying the axiom_logger.dart file to your lib directory.

Initialize the logger early in your app's lifecycle, typically in your main() function.

dart

import 'package:your_app/axiom_logger.dart';

void main() async {
  // Initialize logger
  final config = AxiomLoggerConfig(
    domain: 'https://your-axiom-domain.com',
    dataset: 'your-dataset',
    apiToken: 'your-api-token',
    enableDebugLogs: false, // Set to true for development
  );

  final logger = AxiomLogger(config: config);

  // Log app startup
  await logger.info('App started', metadata: {
    'version': '1.0.0',
    'platform': 'Flutter',
  });

  runApp(MyApp(logger: logger));
}

Use the logger throughout your app to capture important events, errors, and debug information. For example:

dart

// Log user actions
await logger.info('User logged in', metadata: {
  'user_id': userId,
  'login_method': 'email',
});

// Log errors with context
try {
  await someOperation();
} catch (e, stackTrace) {
  await logger.error('Operation failed', metadata: {
    'error': e.toString(),
    'stack_trace': stackTrace.toString(),
    'operation': 'someOperation',
  });
}

Integration with Flutter error handling

Capture Flutter framework errors and send them to Axiom:

dart

void main() async {
  final config = AxiomLoggerConfig(
    domain: 'https://your-axiom-domain.com',
    dataset: 'your-dataset',
    apiToken: 'your-api-token',
  );

  final logger = AxiomLogger(config: config);

  // Capture Flutter framework errors
  FlutterError.onError = (FlutterErrorDetails details) async {
    await logger.error(
      'Flutter framework error',
      metadata: {
        'exception': details.exception.toString(),
        'stack_trace': details.stack.toString(),
        'library': details.library ?? 'unknown',
        'context': details.context?.toString(),
      },
    );
  };

  // Capture async errors
  PlatformDispatcher.instance.onError = (error, stack) {
    logger.error(
      'Uncaught async error',
      metadata: {
        'error': error.toString(),
        'stack_trace': stack.toString(),
      },
    );
    return true;
  };

  runApp(MyApp(logger: logger));
}

Logging user interactions

Track user behavior and navigation patterns:

dart

class MyHomePage extends StatefulWidget {
  final AxiomLogger logger;

  const MyHomePage({required this.logger});

  @override
  State<MyHomePage> createState() => _MyHomePageState();
}

class _MyHomePageState extends State<MyHomePage> {
  @override
  void initState() {
    super.initState();
    widget.logger.info('User navigated to home page', metadata: {
      'timestamp': DateTime.now().toIso8601String(),
      'screen': 'home',
    });
  }

  Future<void> _handleButtonPress() async {
    await widget.logger.debug('Button pressed', metadata: {
      'button': 'submit',
      'screen': 'home',
    });

    // Your button logic here
  }

  @override
  Widget build(BuildContext context) {
    return Scaffold(
      appBar: AppBar(title: Text('Home')),
      body: ElevatedButton(
        onPressed: _handleButtonPress,
        child: Text('Submit'),
      ),
    );
  }
}

Performance monitoring

Log performance metrics to identify bottlenecks:

dart

Future<void> performExpensiveOperation(AxiomLogger logger) async {
  final stopwatch = Stopwatch()..start();

  try {
    await someExpensiveTask();

    stopwatch.stop();
    await logger.info('Operation completed', metadata: {
      'operation': 'expensive_task',
      'duration_ms': stopwatch.elapsedMilliseconds,
      'status': 'success',
    });
  } catch (e) {
    stopwatch.stop();
    await logger.error('Operation failed', metadata: {
      'operation': 'expensive_task',
      'duration_ms': stopwatch.elapsedMilliseconds,
      'status': 'failed',
      'error': e.toString(),
    });
  }
}

Best practices

Use appropriate log levels: Reserve critical for system failures, error for recoverable errors, warning for potential issues, info for important events, and debug for development details.
Add contextual metadata: Include relevant information like user IDs, session IDs, device info, and operation context to make logs more useful.
Dispose properly: Always call logger.dispose() when your app closes to ensure buffered logs are sent.
Handle errors gracefully: Wrap logging calls in try-catch blocks to prevent logging failures from crashing your app.
Use batching wisely: Adjust batchSize and flushInterval based on your app's logging volume and network conditions.

Reference

List of log fields

Field Category	Field Name	Description
Core Fields
	_time	ISO 8601 formatted timestamp when the log event occurred.
	level	Log severity level (DEBUG, INFO, WARNING, ERROR, CRITICAL).
	message	The main log message describing the event.
Custom Metadata
	app_name	Name of the app generating the log.
	version	Application version number.
	environment	Deployment environment (development, staging, production).
	user_id	Unique identifier for the user associated with the event.
	session_id	Unique identifier for the user session.
	error_code	Application-specific error code.
	duration_ms	Duration of an operation in milliseconds.
	status	Status of an operation (success, failed, processing).
Device & Platform
	platform	Operating system or platform (iOS, Android, Web, Desktop).
	device_model	Specific device model generating the log.
	os_version	Operating system version.
Custom Fields
	*	Any custom fields added via the metadata parameter.

Logger configuration options

AxiomLoggerConfig

The AxiomLoggerConfig class configures how the logger connects to Axiom:

domain: The base URL of your Axiom deployment (for example, https://us-east-1.aws.edge.axiom.co).
dataset: The name of the Axiom dataset where logs are sent.
apiToken: Your Axiom API token for authentication.
timeout: Maximum time to wait for HTTP requests (default: 10 seconds).
enableDebugLogs: Enable verbose HTTP logging for debugging (default: false).

AxiomLogger

The AxiomLogger class manages log creation and transmission:

batchSize: Number of logs to buffer before automatically flushing to Axiom (default: 10).
flushInterval: Time interval for automatic flushing (default: 5 seconds). Note: Automatic interval-based flushing isn't yet implemented but can be added.

Log levels

The logger supports five log levels in increasing order of severity:

DEBUG: Detailed information for diagnosing problems, typically used during development.
INFO: Confirmation that things are working as expected, such as successful operations.
WARNING: Indication that something unexpected happened, but the app continues to work normally.
ERROR: A more serious problem that prevented a specific operation from completing.
CRITICAL: A severe error that may cause the app to fail or require immediate attention. Critical logs are sent immediately, bypassing the buffer.

Key methods

Logging methods

log(level, message, {metadata, sendImmediately}): Core logging method that accepts any log level.
debug(message, {metadata}): Convenience method for DEBUG level logs.
info(message, {metadata}): Convenience method for INFO level logs.
warning(message, {metadata}): Convenience method for WARNING level logs.
error(message, {metadata}): Convenience method for ERROR level logs.
critical(message, {metadata}): Convenience method for CRITICAL level logs (sends immediately).

Management methods

flush(): Manually send all buffered logs to Axiom. Returns a boolean indicating success.
dispose(): Clean up resources, flush remaining logs, and close HTTP connections. Call this when shutting down the logger.

Dependencies

dio

The Dio package is a powerful HTTP client for Dart that provides:

Request and response interceptors for debugging and modifying HTTP traffic.
Support for timeouts, custom headers, and authentication.
Error handling and retry mechanisms.
FormData, file uploading, and downloading capabilities.

In this logger, Dio handles all HTTP communication with Axiom's ingest API, including authentication via Bearer tokens and proper JSON serialization of log batches.

intl

The intl package provides internationalization and localization support, including:

Date and time formatting using standard patterns.
Number formatting for different locales.
Message translation support.

In this logger, intl is used to format timestamps in ISO 8601 format (yyyy-MM-dd'T'HH:mm:ss.SSS'Z'), ensuring consistent and parseable timestamp strings across all log entries.

Error handling

The logger includes built-in error handling:

Network failures are caught and logged to the console without crashing the app.
Failed log sends return false from the flush() method, allowing you to implement retry logic.
Uninitialized logger calls are safely ignored with console warnings.

Thread safety

The logger is designed for use in Flutter's single-threaded Dart environment. All async operations use Dart's Future API, ensuring proper sequencing of log operations without race conditions.