Software Setup

Sometimes I find myself forgetting how I configured and built particular software. This page is to remind myself of my base configuration and perhaps help others with similar issues.

Package Manager and Environment

I usually use the brew package manager due to its apparent and rather surprising robustness. I use MacPorts on my vintage Macs such as my iMac G3 since its PowerPC packages are still maintained. But brew seems to be far less invasive than MacPorts, going so far as to not require administrative privileges to install packages. Whatever package manager is used, I suggest being consistent, brew built software tends to easily find other brew packages - same with MacPorts. It is difficult to mix and match packages from different package managers.

Various Packages

GNU

Since this is the 21st and GNU really is the UNIX standard somewhat ironically - it is important to replace the deprecated BSD utils with their respective GNU coreutils. For example, ''sed'' becomes ''gsed'', ''readlink'' becomes ''greadlink'' and so on. Just do brew install coreutils

Also, if you want to do any sort of useful/common software development, you'll want to install the following packages.

Common

$brew install cmake
$brew install python3
$brew install pkg-config
$brew install ctags

Mathematics

If you like to do mathematics(which there is really no reason you shouldn't), you have a couple of good options when it comes to open source.

!!! note Technically this is not true. For those of you who like to reason about abstract algebras or work with the infinite, there are software like CoCoa (see Wikipedia).

I know there are other options that are big in research such as Julia(which i found to be rather brittle in my experience). But I've found the python route to be sufficient for my needs as it provides calculus operators, vector calculus operators, common linear algebra operators, and symbolic and or numerical manipulation for all the above. Sympy handles symbolics and integrates nicely with numpy for numerics. Sympy also generates LaTeX output which Jupyter readily consumes.

Python Route

!!! warning If you're not familiar with the python language, it is much easier to get started with WxMaxima. But I've found mathematics with python to be far more flexible when it comes to more complex operations such as plotting the gradient of an electromagnetic field.

$pip3 install sympy numpy scipy jupyterlab

Maxima Route

#!bash
$brew install maxima
$brew install wxmaxima
$pip3 install --user numpy scipy matplotlib ipython jupyter pandas sympy nose

FPGA Dev

#!bash
brew install yosys
brew install icarus-verilog
brew install verilator

Embedded Dev

In case you have an MBED or Arduino lying around and don't already have enough troubles as it is.

#!bash
$brew tap ArmMbed/homebrew-formulae
$brew install arm-none-eabi-gcc
$brew tap osx-cross/avr
$brew install avr-libc

GUI Dev

WxPython On MacOS - it's pretty simple

brew install wxPython

Ubuntu - it's not so simple

replace 18.04 with your ubuntu version which you can determine by invoking lsb_release -a

If you tried installing wxPython with apt - you may notice its broken.

You might have to remove the wx.py in your home directory. rm ~/wx.py

pip3 install -U -f https://extras.wxpython.org/wxPython4/extras/linux/gtk3/ubuntu-18.04 wxPython

Verify your wxPython was correctly installed by running the wxPython demo.

#!bash
$git https://github.com/wxWidgets/Phoenix.git
$cd Phoenix/demo
$python3 demo.py