1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
|
.. index:: Introduction
Introduction
============
ODB-2 is a compact data format suited for storage, transmission and archival of meteorological observation data in a tabular format, with each row corresponding to one observation. Observation is self-describing and can come from many different data sources and many types of instrument.
**pyodc** provides a thin encoder and decoder to make ODB-2 data available for the **pandas** or **numpy** ecosystem. In addition to that, it also has an interface to explore the metadata without actually decoding ODB-2 data.
.. index:: ODB-2
single: ODB-2; Format
single: ODB-2; Frame
single: ODB-2; Header
ODB-2 Format Overview
---------------------
An ODB-2 data stream is comprised of a sequence of frames. Each frame encodes a table of data, described by a header.
.. figure:: /_static/odb-2-message-stream.svg
:alt: ODB-2 Data Structure
ODB-2 Data Structure
It is important to understand that the ODB-2 should not be considered as a file format, but a data stream instead. In a stream of messages frames can also be unrelated, having entirely different data, so they must be decoded separately. However, if in a stream of messages the frames contain the the same data types, they are suitable for combined decoding and processing.
.. seealso::
For full ODB-2 format reference, please see :doc:`odc:content/reference/file-format-reference` chapter in **odc** documentation.
|
