logx

Python implementation of structured logging compatible with NETSTARS log specification v1.3

Introduction

logx is a universal logging library that conforms to the NETSTARS log specification requirements. By default, it supports NETSTARS Log Specification v1.3. Under specification v1.3, logx outputs logs in JSON format, as shown in the following example:

{"compliance":"1.3","log_app_id":"test_app_id","log_app_version":"0.1","log_category":"srv-exception-log","log_event_id":"","log_event_time":"2022-01-21T10:53:21.774+09:00","log_level":"fatal","log_time":"2022-01-21T10:53:21.774+09:00","message":"this is a test message"}

JSON format logs can nest JSON, and newlines and double quotes in key-value content are automatically escaped.

During development and debugging, you can set the compliance level to "0" to output simple semicolon-separated log information:

2022-01-21T10:56:43.901+09:00 [INF] item_1: 23; log_event_time: 2022-01-21T10:56:43.901+09:00; message: this is a test message

Note: Specification v1.3 meets SLA requirements and can be parsed by Elasticsearch; specification 0 is only recommended for development and testing phases.

Log Levels

Log levels in logx are integers conforming to the v1.3 log specification. Valid range is [0,6], where 0 means no log output, and 1-6 correspond to fatal, error, warn, info, debug, trace respectively. Larger values provide richer output information. The maximum output log level can be specified in the logx.init() initialization method to control the level of detail. You can also dynamically adjust the maximum log output level using the logx.set_max_level() method.

Usage

Complete Usage Example

from logx import init, errorf, Log, InfoLvl, New, CustomItem, DebugLvl

# Initialize logging
init(DebugLvl, "my_app", "1.0.0", "1.3")  # Set to debug level to show all logs

# Simple log output
errorf("用户登录失败: %s", "密码错误")

# Custom log items
Log(InfoLvl, "用户操作",
    CustomItem("user_id", "12345"),
    CustomItem("action", "login"))

# Use Loggerz object
logger = New(
    CustomItem("service", "user-service"),
    CustomItem("version", "1.0.0"),
)
logger.printf("处理用户请求: %s", "GET /api/users")

Output

Running the above code will output the following JSON formatted logs:

{"compliance":"1.3","log_app_id":"my_app","log_app_version":"1.0.0","log_category":"srv-exception-log","log_event_id":"","log_event_time":"2025-10-22T19:58:30.785+08:00","log_level":"error","log_time":"2025-10-22T19:58:30.785+08:00","message":"用户登录失败: 密码错误"}
{"compliance":"1.3","log_app_id":"my_app","log_app_version":"1.0.0","log_category":"srv-biz-process-log","log_event_id":"","log_event_time":"2025-10-22T19:58:30.785+08:00","log_level":"info","log_time":"2025-10-22T19:58:30.785+08:00","message":"用户操作","user_id":"12345","action":"login"}
{"compliance":"1.3","log_app_id":"my_app","log_app_version":"1.0.0","log_category":"srv-biz-process-log","log_event_id":"","log_event_time":"2025-10-22T19:58:30.785+08:00","log_level":"info","log_time":"2025-10-22T19:58:30.785+08:00","message":"处理用户请求: GET /api/users","service":"user-service","version":"1.0.0"}

Detailed Description

• Initialization: Call the logx.init() method at program startup to set the maximum output log level. Usually this method is called in the main() function, and the parameter values can be provided through the application's configuration file.

  • DebugLvl: Sets the log level to debug, which outputs all level logs
  • "my_app": Application ID
  • "1.0.0": Application version number
  • "1.3": Log specification version number. If "0" is specified, simple information is output, recommended only during debugging

• Simple log output: Use methods like logx.printf(), logx.errorf() to output simple log information:

logx.errorf("用户登录失败: %s", "密码错误")

• Custom log items: Use logx.log() and logx.CustomItem() methods to achieve maximum flexibility in log information customization:

logx.Log(InfoLvl, "用户操作",
    CustomItem("user_id", "12345"),
    CustomItem("action", "login"))

• Using Loggerz objects: Use the logx.new() method to create a new Loggerz object and save some common log information, so you don't have to manually pass this information each time:

logger = logx.new(
    CustomItem("service", "user-service"),
    CustomItem("version", "1.0.0"),
)
logger.printf("处理用户请求: %s", "GET /api/users")

Installation

From GitHub

pip install git+https://github.com/0xNetx/logx.git

From Local Directory

If you've cloned the repository:

cd logx
pip install -e .

Or install directly:

pip install -e /path/to/logx

Version

• v0.0.1: Initial release version with complete log formatting functionality
• Supports NETSTARS v1.3 specification JSON format output
• Supports simple text format output (compliance 0)
• Includes hook mechanism and Loggerz object functionality

API Reference

Constants

• DisabledLvl = 0: No log output
• FatalLvl = 1: Fatal error
• ErrorLvl = 2: General error
• WarnLvl = 3: Warning
• InfoLvl = 4: General info
• DebugLvl = 5: Debug information
• TraceLvl = 6: Trace information

Functions

Initialization

• init(level, app_id, app_version, compliance): Initialize log configuration
• set_max_level(level): Set maximum log output level
• get_max_level(): Get maximum log output level

Logging Functions

• log(level, message, *log_items): Log a message with specified level
• fatal(*args): Log fatal message
• error(*args): Log error message
• warn(*args): Log warning message
• print(*args): Log info message
• debug(*args): Log debug message
• trace(*args): Log trace message

Each level also has *f (formatted) and *ln (with newline) variants.

Options

• CustomItem(key, value): Add custom log item
• Compliance(version): Set compliance version
• LogTime(time): Set log time
• EventTime(time): Set event time
• Level(level): Set log level
• Category(category): Set log category
• AppId(app_id): Set application ID
• AppVersion(version): Set application version
• EventId(event_id): Set event ID

Objects

• new(*items) -> Loggerz: Create a new logger object

Loggerz Class

The Loggerz class provides all the functionality of the global functions plus:

• enable_log_elapsed_time(): Enable elapsed time tracking
• get_items(): Get all log items
• set_items(*items): Update log items
• add_hook(hook): Add a hook function