dP                                                   dP   
                                    88                                                   88   
88d888b. dP    dP 88d888b. 88d888b. 88 .d8888b.    .d8888b. .d8888b. .d8888b. dP    dP d8888P 
88'  `88 88    88 88'  `88 88'  `88 88 88ooood8    Y8ooooo. 88'  `"" 88'  `88 88    88   88   
88.  .88 88.  .88 88       88.  .88 88 88.  ...          88 88.  ... 88.  .88 88.  .88   88   
88Y888P' `88888P' dP       88Y888P' dP `88888P'    `88888P' `88888P' `88888P' `88888P'   dP   
88                         88                                                                 
dP                         dP

Description

PurpleScout is an app that simplifies the scouting process for FRC teams. Data pipeline (from scouting to analysis) is as follows:

1. A Windows laptop runs a Python/Flask app that hosts a web form
2. Scouters collect data and input into the app
3. After each match, an HTML form is submitted that contains the data
4. Laptop recieves POST request
5. Data from scouts is combined and organized into a CSV file
6. CSV file is periodically synced to Google Drive
7. Google Sheets runs analytics on the data and creates suggestions for the team
8. Added capabilities to perform pit scouting at the same time as normal scouting using Blueprints. 8?? Web dashboard accessible to admins to see data at-a-glance and make decisions

Game definition

PurpleScout can be configured for any game.

1. Define fields in game.py

This app uses Flask-WTF for form gerenation. In game.py, make a subclass of FlaskForm with the name of your game.

Add instance variables in each class to represent fields, or pieces of data you want to collect. Example:

class ChargedUpScoutForm(FlaskForm):
    matchNum = IntegerField('Match Number', validators=[DataRequired()])
    teamNum = IntegerField('Team Number', validators=[DataRequired()])

    cubesScored = IntegerField('Cubes Scored', validators=[DataRequired()])
    conesScored = IntegerField('Cones Scored', validators=[DataRequired()])

    defenseQuality = RadioField('Defense Quality', choices=[
      (0,'No defense'),
      (1, 'Average defense'),
      (2,'Good defense')
    ], validators=[DataRequired()])

    comment = TextAreaField('Comment', validators=[DataRequired()])

Most fields you need will be of type IntegerField, RadioField, or TextAreaField. Refer to WTForms and Flask-WTF documentation for more information.

2. Create template in templates/forms

Create a new HTML file in templates/forms with the same filename as your class. Example:

<!-- ChargedUpScoutForm.html -->
{% extends "scout.html" %}
{% block form %}

{{ form.matchNum }}
{{ form.teamNum }}
<table class="my custom html">
  {{ form.cubesScored }}
  {{ form.conesScored }}
</table>
{{ form.defenseQuality }}
{{ form.comment }}

{% endblock %}

Each {{ form.field }} will be replaced by an <input> element of the corresponding type. You can add classes like this: form.field(class_="custom classes") and style them in scout.css.

You don't need to include a <form> wrapper or a submit button. If you want minus and plus buttons next to your integer inputs, add <button class='subtractButton'>-</button> and <button class='addButton'>+</button> before and after the {{ form.field }} line. The app will automatically add event listeners.

3. Configure main.py

Change this line to the name of your class:

app.config["GAME"] = ChargedUpScoutForm

SQLite database

PurpleScout uses an SQLite database to store data. The database is stored in data/scout.db and has the following schema:

scoutData table

• timestamp (string): the time the data was submitted, automatically generated
• matchNum (integer): the match number
• teamNum (integer): the team number
• scoutID (string): the scout's ID
• data (string): a JSON string containing all the other data from the form

Running Purple Scout as a Container

Installing a Container Runtime

To run PurpleScout as a container, you will need a container runtime. If you are running Windows, follow here: https://docs.docker.com/desktop/install/windows-install/

Building the container manually (not over GitHub Actions)

docker buildx build . -t purple-scout:latest

Running the container

In the example above, data will be read and written on the current directory

docker run -p 5000:5000 ".:/data" -t purple-scout:latest