scandroid

Scans english verse in iambic and anapestic meters. The version 1.1 (2015) source is copied from Charles O. Hartman's page.

Version 1.1 was written in python 2. I have updated the code for python 3. My ultimate goal is to use this for song lyric writing.

User's Manual

The manual for version 1.1 (pdf) from 2015 is copied from Hartman's page to the doc/reference folder. I have ported the manual to markdown format, so that it will be covered under GPL3 and can be maintained moving forward for future versions. The ported manual (markdown) is located in the doc folder.

What is Metrical Foot?

It is an unit consists of a pattern of stressed and unstressed syllables.

Iambic: Pronounced duh-DUH. Consists of two syllables- an unstressed syllable followed by a stressed syllable. An iamb can be made up of one word with two syllables or two different words.

Trochee: Pronounced DUH-duh. A line of verse with one or more trochees is said to have a trochaic meter. This stress pattern is the opposite of iambic meter.

Spondee: Spondees have two stressed syllables, pronounced DUH-DUH.

Dactyl: Dactyls are pronounced DUH-duh-duh. This foot has a stressed syllable followed by two unstressed syllables.

Anapest: Pronounced duh-duh-DUH. Anapestic meter typically divides its syllables across multiple words.

VS code setting

For linting correctly in VS code, we need to whitelist wxPython binary package for pylint.

File .vscode/settings.json

{
    "python.linting.pylintArgs": [
        "--extension-pkg-whitelist=wx"
    ],
}

I use VS code for development and I have multiple versions of python installed and therefore, point pythonPath to correct python 3 version that has wxPython installed in it so that pylint can find wxPython.

File .vscode/settings.json

{
    "python.pythonPath": "~/AppData/Local/Programs/Python/Python39/python.exe",
}

You can also interactively select python version in VS code status bar.

Pre-built binaries (Executable file) for releases

Executable is the most straight forward option since they have batteries included. You can run them directly without any additional steps or installing any additional components.

Github provides windows, MacOs and ubuntu virtual machines for github actions. As such, we can only pre-build windows, MacOs and ubuntu binaries using PyInstaller right at github. The virtual machines are created fresh right before the run and after the run is finished the virtual machines are disposed. As such this comes with a high probability of being free of malicious influences and should be trust worthy. However bugs are always possible.

If you don't need to build the binaries yourself then you can download the pre-built executable binaries for windows, MacOs and ubuntu for any binary release from our github release page.

Running Scandroid from source code using Python

Of course you have to have Python 3 installed. I have tested with Python 3.7.3. Following are the instructions for running scandroid directly from source code with python. You will also need wxPython installed for the user interface.

Windows O/S

pip install -U wxPython

MacOs, linux and other O/S

To install wxPython consult .github/workflows folder. This folder contains commands for setting up wxPython using github actions for windows, MacOs and ubuntu.

Now you should be able to run the source directly using python interpreter from cloned scandroid directory.

py ./source/Scandroid.py

Building executable

If you can run scandroid from source using python successfully on a machine then you can use pyinstaller to build executable for the same o/s in the same machine. Executable can be a preferred option since running executable won't need a machine with python and wxPython installed. You can just distribute the executable bundle. The bundle will have everything it needs to run.

Windows O/S

First install PyInstaller.

pip install pyinstaller

Now you should be able to build the executable from cloned scandroid directory.

build with one file option

This option is useful for easy distribution. Run PyInstaller with -F option,

pyinstaller -F --add-data "./source/scandictionary.txt;." "./source/Scandroid.py"

This will generate a single executable file in a subdirectory called dist. You simply distribute this executable.

build with one folder option

This option is useful for debugging a bundle. Run PyInstaller with -D option,

pyinstaller -D --add-data "./source/scandictionary.txt;." "./source/Scandroid.py"

This will generate the executable bundle folder in a subdirectory called dist. You can zip the bundle folder inside dist and distribute.

There are other advanced options available for building. For more details, see the pyinstaller manual.

PyInstaller is not a cross platform builder. That means if you want to build a Mac executable, you have to build the executable in a Mac.

MacOs, linux and other O/S

Building binary for a platform consult .github/workflows folder. This folder contains commands for building binaries using github actions. We use github actions to build binaries for windows, MacOs and ubuntu each time we do a commit and release.