Name | Modified | Size | Downloads / Week |
---|---|---|---|
Parent folder | |||
DvObjectClass.html | 2019-05-17 | 499.1 kB | |
Readme.txt | 2019-04-12 | 1.6 kB | |
dvos230.tar.gz | 2019-04-12 | 183.1 kB | |
Totals: 3 Items | 683.7 kB | 0 |
DVOS - Data Variable Object System Imperial College, Space Plasmas, Time Series Analysis library This library provides data object handling for time series data with emphasis on Space Physics data with standard metadata. Data objects comprise sequences of multi-dimensional arrays of elements, so a single scalar value would have a sequence length of 1 with 1 dimension of size 1. Elements may be one of the following data types: double, integer, text (DvString), time (DvTime), Time Interval (DvEvent). A data object will hold data of only one type. Data objects may have lists of metadata attached, and metadata is also stored as a DVOS object. Metadata may itself have metadata attached, resulting in a self similar hierarchy. Metadata is attached through the change_xref() method which prevents infinite regression. All common mathematical operations are supported on the data, and metadata checking is performed to ensure operations are allowed (e.g. units, coordinate frames). Objects resulting from mathematical operations will have relevant metadata derived from the input objects. Joining onto a common timeline is performed on the fly for analysis, but explicit joining methods are also supported. To compile, change directory to src, edit the makefile to uncomment your platform (default is Mac OSX) and type make at the prompt. Note also, 'make clean' will remove .o files, while 'make empty' will remove all binaries (.o and libraries) to ensure a clean start. Tony Allen Space and Atmospheric Physics, Imperial College, London April 2019