neila/Minerva-ASM-elections

a coded voting in the Schultze Method, "tailored" for Minerva Associated Students. Note that other methods are supported but edits are required.

ASM_Elections_Code

IMPORTANT

Before using this to actually run an election, read the segment on preparing the data for processing. Results will be incorrect otherwise.

p.s. this code and instructions were written on/for Mac OS (Sierra, 10.12.6).

How the CSV file needs to be structured?

First Column: Class ID

in theory, if ASM representatives are per-cohort this ID should be unique for class+cohort. eg. M23A, M23B, M24, M25, M26A, M26B, etc.

First Row: Headers

Question headers and Candidate Names

Candidate Columns

• the candidate of each class (or cohort) should be sequential.

ie all candidates for M22 should span columns 1-10, not columns 1, 4, 10-15, 21-23.

• all candidate columns should be sequential.

ie. no columns contatining other information should exist between two candidate names, EVEN if those are candidates of different classes/cohorts.

• The code can be run multiple times on multiple files, but as long as the above is true, it will identify the candidates and voters of each class and spew-out the winners.

Voters Don't Have to be Sequential

rows represent a ballot/vote. Voters can be in any order, as long as column 0 is class IDs.

Voters Not Spoiling Their Ballots Must Rank ALL Candidates

it's how our voting works. If it ever changes, changes have to be made to the code.

these changes include, but may not be limited to:

• the way run_schulze.py counts votes
• the way asm_vote_counter.py recognizes candidate columns

What's In Here?

Exists:

• vote-core

This is the core of the code – an implementation of the election method.

• originally from: https://github.com/bradbeattie/python-vote-core
• there's a Py3 version: https://pypi.org/project/python3-vote-core/
• the authors say it isn't "clean" though, but works.

• pygraph

This is required for vote-core. Why? IDK. Uses the algorithms in it.

• originally from https://github.com/pmatiello/python-graph
• updated version exist in: https://github.com/Shoobx/python-graph honestly, I haven't tested the new version. May break everything.

• asm_vote_counter.py

• asks for the first column in the CSV that has candidate names
• asks for number of desired winners
• calls run_shulze.py on each class ballot

ie. it recognizes in a larger CSV which lines belond to each class as well as the corrosponding cadidates.

• run_schulze.py

counts the ballots and runs the schulz method on the data

• UI_elements.py

Containes code for the buttons and pop-ups uses Tk and Tkinter, which will require updating in Python 3

Missing:

• Processing from Excel to csv

Cautionary Note (Roiman, M19, 2019)

Honestly, I don't know all of the ins and outs of this code. It works. It worked when we needed it to work. But it's one of those "sold as-is" things. Schultze is definitely an awesome method. But maybe get some CS majors to rebuild it.

Dependencies:

The code used by the ASM has several dependencies. These include python-vote-core and python-graph-master. These cant be found with pip, as far as I could tell, and we definitely should re-work them ASAP.