loghandler 0.5.0

Python logging library with support for multiple destinations

loghandler

Easy logging package for all your logging needs.

Features

• Log to multiple endpoints at once
• Support for STDOUT, Elasticsearch, Database (MySQL, PostgreSQL, SQLite), and more coming soon.
• Easy syntax
• Fail-over reporting (If one endpoint fails it will be reported to the other endpoints)

Installing

Install loghandler via pip

pip install loghandler

Using

In your code import LogHandler and initalize it.

from loghandler import LogHandler

logger = LogHandler({
    "log_level": "DEBUG",
    "outputs": [
        {
            "type": "STDOUT"
        }
    ]
})

You can now log messages to all your outputs via:

logger.log('fatal', Exception("Something went HORRIBLY wrong"))

Endpoints

The following endpoints are currently in the works and will be supported soon.

• logstash
• sentry

General Configuration

All endpoints accept a few global settings. They are shown below.

log_level: For the output it's applied to, this will overrule the global configuration level

report_error: If an output fails to send a message, this error will be reported to the other outputs, by default this is True. (It's not recommended to turn this off.)

retry_after: Defines how long an output should wait before retrying, by default this is 15s (Defined in seconds)

STDOUT

To use STDOUT as a log endpoint, add the following to your outputs array.

{
    "type": "STDOUT"
}

Elasticsearch

To use elasticsearch as a log endpoint, add the following to your outputs array.

{
    "type": "elasticsearch", 
    "config": {
        "hosts": ["https://your-es-host.com:9243"],
        "ssl": True,
        "verify_certs": True,
        "refresh": "wait_for",  # Must be either "true", "false" or "wait_for"
        "index": "your-index",  # Index will be created if it doesn't exist
        "api_key": ("your-api-key-id", "your-api-key-secret")
    }
}

Next time something is logged you should see something like the following under your index:

{
  "_index" : "logs",
  "_type" : "_doc",
  "_id" : "some-id",
  "_score" : 1.0,
  "_source" : {
    "timestamp" : "2021-11-05T04:16:25.250206",
    "level" : "DEBUG",
    "hostname" : "YOUR-HOSTNAME",
    "message" : "division by zero",
    "occurred_at" : {
      "path" : "/somepath/test.py",
      "line" : 22
    }
  }
}

Database

Table Structure

Table(
    db_config["table_name"],
    metadata,
    Column("id", Integer, primary_key=True),
    Column("message", Text),
    Column("level", String),
    Column("origin", String),
    Column("timestamp", DateTime),
)

sqlite

To use sqlite as a log endpoint, add the following to your outputs array.

{
    "type": "sqlite", 
    "config": {
        "table_name": "logs",  # Will be created if it doesn't exist
        "db_path": "/path/to/db.sqlite",  # Will be created if it doesn't exist
    }
}

Next time something is logged you should see something like the following under your table:

('division by zero', 'DEBUG', '/somepath/test.py:31', '2021-11-07 01:27:24.755989')

mysql

To use mysql as a log endpoint, add the following to your outputs array.

{
    "type": "mysql",
    "config": {
        "table_name": "logs",
        "connection_string": "root:example@localhost:3306/example_db"
    }
}

Next time something is logged you should see something like the following under your table:

division by zero | DEBUG | /somepath/test.py:22 | 2021-11-07 01:46:58

pgsql (PostgreSQL)

To use pgsql as a log endpoint, add the following to your outputs array.

{
    "type": "pgsql",
    "config": {
        "table_name": "logs",
        "connection_string": "postgres:postgres@localhost:5432/example"
    }
}

Next time something is logged you should see something like the following under your table:

division by zero | DEBUG | /somepath/test.py:22 | 2021-11-07 01:46:58