td-logger 0.4.3

A Python logging handler for Treasure Data Cloud

td-loger: logging library for Treasure Data Cloud
=================================================

Setup
-----

Before using this logging handler, td-agent must be properly configured at the
localhost. Please confirm these settings are in your
/etc/td-agent/td-agent.conf.

## built-in TCP input
<source>
type tcp
</source>

# Treasure Data output
# match events whose tag is td.DATABASE.TABLE
<match td.*.*>
type tdlog
apikey YOUR_API_KEY
</match>

For more information, please look at the documentation.

* http://docs.treasure-data.com/

Install
-------

$ easy_install td-logger

or

$ pip install td-logger

Usage
-----

Check out the tests folder for more samples.

import logging
from tdlog import logger

logging.basicConfig(level=logging.INFO)
l = logging.getLogger('td_logger.test')
l.addHandler(logger.TreasureDataHandler())

l.info('Some message')
js = { "semicolon" : ";", "at" : "@" }
l.info(js)

This will throw the log entries to the local td-agent. And there's an important
point. This logger accepts only 'dict' and 'json string'.

Also, these parameters are logged by default.

* sys_host
* sys_name
* sys_module
* sys_lineno
* sys_levelno
* sys_levelname
* sys_filename
* sys_funcname
* sys_exc_info
* msg

These parameters can be specified at TreasureDataHander constructor.

* host: td-agent host (default: 127.0.0.1)
* port: td-agent port (default: 24224)
* db: td database name (default: log)
* table: td table name (default: default)
* bufmax: buffer size max when td-agent is unavailable (default: 1*1024*1024)
* timeout: network timeout (default: 3.0)

Have fun!

How to Develop
--------------

Folloings are the way to setup development envionment on MacOSX.

$ sudo easy_install virtualenv
$ virtualenv --no-site-packages .
New python executable in ./bin/python
Installing setuptools............done.
Installing pip...............done.
$ source bin/activate
$ bin/pip install msgpack-python
$ python run_tests.py

Special Thanks
--------------

- [Joshua Kuntz](https://github.com/j3kuntz) for the original work