1. CODAS Installation

This guide is for installing the CODAS processing system, which works for shipboard ADCP data acquired using either the UHDAS or the VmDAS data acquisition system. It does not cover installation of either of those systems. There is also limited legacy support for old data sets acquired with DAS 2.48, which ran on MSDOS machines. Supported file types include several single-ping and pre-averaged styles.

CODAS software is written in Python 3.6+ and C. Like all modern software, it takes advantage of many libraries that must be installed first. Our software can be built on MacOS and Linux but not on Windows. If you must run it on a Windows computer you can do it via a virtual machine as described below.

There are three supported ways to get a computer set up with CODAS processing software, including the libraries on which it depends.

(Option #1) Virtual Computer:

We provide a fully configured virtual machine running Linux that you can import and run via virtualization software such Virtual Box, which is free.

  • Advantages:

    • fewer steps than other options

    • works on machines running Windows, MacOS, and Linux

    • CODAS software and data files are already installed and tested

    • you can delete it when you don’t want it anymore

    • the demos (for processing) are all copy-paste tested on the virtual machine

    • actual user’s data are kept on the user’s computer (not inside the virtual computer) so they can be accessed and backed up with familiar tools

  • Disadvantages:

    • Administrative privileges are required for the initial Virtual Box installation.

    • It requires more machine resources than a native installation.

    • Display response is slow compared to a native installation.

    • There is a learning curve in operating and managing the virtual machine.

This is the only supported method if you have a Windows computer. It is also the method we require for our CODAS workshops, where we need to be sure all participants are ready to go with identical setups at the start.

(Option #2) Using the conda package manager:

The conda package manager has recently become our favorite tool for installing most CODAS dependencies natively on Linux and MacOS, for personal use.

  • advantages:

    • No administrative privileges are needed. Everything is installed in, and run from, your user home directory tree.

    • Everything is done with the operating system you are familiar with.

    • It’s easy to update.

    • You can make multiple environments, so you can test a new version of code or a library without disturbing your working environment.

  • disadvantages:

    • You must execute additional steps that are already done for you in the Virtual Computer.

(Option #3) Ubuntu Linux native:

Install a recommended version of Ubuntu (presently Xubuntu 20.04) and use the native package manager to install the dependencies.

  • advantages:

    • Concise and tested installation instructions, at least for one version.

    • The packages are stable.

  • disadvantages:

    • Your machine is now dedicated to this version of Linux.

    • You can only do this if you have full control over the machine.

From the sections below, choose one of the 3 options and proceed as directed. The CODAS installation and testing steps are needed only for options 2 and 3.

Let us know (uhdas@hawaii.edu) if you are installing CODAS programs and want to be informed if there are updates or bug fixes.