dzlog

dzlog

dzlog is a lightweight and extensible logger for Go applications, designed for backend services and microservices that need structured logs, context propagation and correlation (trace) IDs, without pulling heavy logging frameworks.

It provides an Apache-style log format, configurable log levels, message splitting for large payloads, and seamless context.Context integration.

✨ Features

• Log levels: DEBUG, INFO, WARN, ERROR
• Environment-based log level via LOG_LEVEL
• Context-aware logging (context.Context)
• Trace / correlation ID support (tracerId)
• Automatic caller info (file + line)
• Large message splitting
• Apache-style log format
• Zero configuration by default
• No reflection, no unsafe

📦 Installation

go get github.com/dmytrozilnyk/dzlog

🚀 Quick start

import "github.com/dmytrozilnyk/dzlog"

log := dzlog.New(
	dzlog.LogModuleName("my-service"),
	dzlog.LogLevel("INFO"),
	dzlog.LogEnvironment("dev"),
)

log.Info("Service started")
log.Warning("Cache miss")
log.Error("Database error")

⚙️ Configuration

dzlog uses functional options for configuration.

Log level

You can configure the log level in two ways:

1️⃣ Programmatically

dzlog.LogLevel("DEBUG")

Supported values:

• DEBUG
• INFO
• WARN
• ERROR

If an invalid value is provided, it falls back to INFO.

2️⃣ Environment variable (overrides code)

export LOG_LEVEL=DEBUG

If LOG_LEVEL is set, it always overrides the configured value.

Module name

Identifies the service or component emitting logs.

dzlog.LogModuleName("users-api")

Environment name

Useful to distinguish logs between environments (dev, staging, prod, etc.).

dzlog.LogEnvironment("prod")

Max message length

Messages longer than the configured length are automatically split into multiple log entries.

dzlog.LogMaxLength(3000)

If the value is <= 0, the default (3000) is used.

Each split entry includes an entries: x/y indicator.

🧠 Context & Trace ID

dzlog supports context.Context to propagate metadata such as trace / correlation IDs.

The context key used internally is:

"tracerId"

Example

ctx := context.WithValue(context.Background(), "tracerId", "abc-123")

log.InfoCtx(ctx, "Processing request")
log.ErrorCtx(ctx, "Request failed")

If the context contains a tracerId, it will automatically appear in the logs.

⏱️ Time measurement (DEBUG only)

Useful for performance tracing. Time measurement is only active when the log level is DEBUG.

start := log.GetTimeNow()

// do work...

elapsed := log.CalculateDifference(start)
log.Debug("Elapsed time:", elapsed)

If the log level is not DEBUG, GetTimeNow() returns a zero time.Time and CalculateDifference returns 0.

🔁 Returning context

Some methods return the same context to allow chaining or propagation.

ctx = log.InfoReturnCtx(ctx, "Step completed")

Available methods:

• DebugReturnCtx
• InfoReturnCtx
• WarningReturnCtx

📄 Log format

Example output:

[2026-02-04 15:04:05.123][users-api:INFO][env:dev][pid:1234 tid:0][tracerId: abc-123][entries: 1/1][handler.go:42] User created

Each log entry includes:

• Timestamp
• Module name
• Log level
• Environment
• Process ID (PID)
• Thread ID (TID)
• Trace ID (tracerId) if present
• Entry index when message splitting is used
• Source file and line number
• Message content

📄 License

MIT License