__       __    __
.--.--.--|__.-----|  |--|  |--.-----.-----.-----.
|  |  |  |  |__ --|     |  _  |  _  |     |  -__|
|________|__|_____|__|__|_____|_____|__|__|_____|
                                   version 2.1.2

Build composable event pipeline servers with minimal effort.


=======================
wishbone.encode.flatten
=======================

Version: 0.1.0

Flattens a dict structure of arbitrary depth into individual metric events.
---------------------------------------------------------------------------


    This module takes a <dict> structure and recursively travels it
    flattening the namespace into a dotted format untill a numeric value
    is encountered. For each metric a <wishbone.event.Metric>
    datastructure is created and stored in the @data part of
    <wishbone.event.Event>.

    Non-numeric values are ignored.

    For example:

        {"server": {"host01": {"memory": {"free": 10, "consumed": 90}}}}

        Would generate following metrics:

        server.host01.memory.free
        server.host01.memory.consumed

    These metrics are converted to the Wishbone metric data format:

        http://wishbone.readthedocs.org/en/latest/logs%20and%20metrics.html#format

    The module is expecting a Python <dict> type.  That means you should have
    already decoded the incoming data using a module like wishbone.decode.json.


    Parameters:

        - type(str)("wishbone")
           |  An arbitrary string to assign to the "type" field of the Metric
           |  datastructure.

        - source(str)("wishbone")
           |  An arbitrary string to assign to the "source" field of the Metric
           |  datastructure.

        - tags(set)()
           |  An arbitrary set of tags assign to the "tags" field of the Metric
           |  datastructure.


    Queues:

        - inbox:    Incoming events.

        - outbox:   Outgoing events with @data containing the wishbone.event.Metric data.