Requiremnts and Installation

Requirements

The table below describes Kenozooid core (mandatory) and optional dependencies.

The availability column provides information about a package being provided with an operating system (i.e. Linux distro, FreeBSD) or a binary available to install on given platform (Windows, Mac OS X).

Name

Version

Type

Availability

Description

Core

Python

3.6.4

execution environment

Arch, Debian Wheezy, Fedora 15, Mac OS X, PLD Linux, Ubuntu Natty, Windows

Kenozooid is written in Python language

lxml

2.3

Python module

Fedora 15, Mac OS X, PLD Linux, Windows

XML parsing and searching

dirty

1.0.2

Python module

Mac OS X, PLD Linux, Windows

XML data generation

dateutil

2.0

Python module

Mac OS X, PLD Linux, Windows

date and time parsing

cytoolz

0.9.0

Python module

data processing

Optional

decotengu

0.14.0

Python module

required by decompression dive planner

pyserial

2.6

Python module

PLD Linux, Windows

required by OSTC driver

btzen

0.2.3

Python module

required by hwOS driver

libdivecomputer

library

required by Sensus Ultra driver

R

3.1.3

R scripts execution environment

Arch, Debian Wheezy, Fedora 15, Mac OS X, PLD Linux, Ubuntu Natty, Windows

plotting and dive data analysis

rpy

3.0.5

Python module

PLD Linux

used to communicate with R

Hmisc

R package

used for plotting

colorspace

R package

used for overlay plotting

Installation

Kenozooid Installation

Kenozooid can be downloaded from PyPI

Unpack the downloaded archive and use the following command to install Kenozooid:

$ python3 setup.py install

The Checking Dependencies section describes how to check additional Kenozooid dependencies like R and R packages needed for plotting and data analysis.

At the Bleeging Edge of Kenozooid Development

Kenozooid can be used directly after fetching source code from its Git repository. This allows to follow Kenozooid development and use its unstable but latest features.

Kenozooid source code is hosted on Savannah at

To check out from the repository simply type:

git clone http://git.savannah.gnu.org/r/kenozooid.git

After fetching the source code enter Kenozooid directory, set up paths and simply invoke Kenozooid command line interface, for example:

cd kenozooid
export PYTHONPATH=.:$PYTHONPATH
export PATH=./bin:$PATH
kz --help

Checking Dependencies

The Kenozooid dependencies can be checked with ‘setup.py’ script, which is part of Kenozooid source code. The dependency checking verifies version of Python used and installation of Python modules and R packages.

To check the dependencies execute the following command from Kenozooid’s source code directory:

python3 setup.py deps

Example, fully successful output of dependency check, can be as follows:

$ python3 setup.py deps
running deps
Checking Kenozooid dependencies
Checking Python version >= 3.2... ok
Checking core Python module lxml... ok
Checking core Python module dirty >= 1.0.2... ok
Checking core Python module dateutil... ok
Checking optional Python module rpy2... ok
Checking optional Python module serial... ok
Checking R package Hmisc... ok
Checking R package colorspace... ok

Example, successful output of dependency check, but with missing optional dependencies, might look in the following way:

$ python3 setup.py deps
running deps
Checking Kenozooid dependencies
Checking Python version >= 3.2... ok
Checking core Python module lxml... ok
Checking core Python module dirty >= 1.0.2... ok
Checking core Python module dateutil... ok
Checking optional Python module rpy2... ok
Checking optional Python module serial... not found
Checking R package Hmisc... not found
Checking R package colorspace... ok

Missing optional dependencies:

  Install serial Python module with command

      easy_install-3.2 --user pyserial

  Install Hmisc R package by starting R and invoking command

      install.packages('Hmisc')

R Packages Tips

R is very sophisticated and powerful statistical software with many addons distributed via The Comprehensive R Archive Network.

When installing R packages required by Kenozooid, some additional software might be needed

  • Fortran compiler is required to compile some R packages, i.e. Hmisc; on Linux gcc-fortran package should be installed