hps-submarine-architecture

Hello guys! This is the architecture for the submarine problem.

There are multiple parts to this project:

• The Submarine Alert UI to go with the game
• The actual server/client code that plays the game

Setup Client / Server Code

1. This project uses python3.6. If you haven't already, please install it.
2. Download dependencies: pip install -r requirements.txt
3. cd into this project on your machine and run python3.6 run_game.py

You should see some output that looks similar to this:

sub {'m': 10, 'L': 4, 'pos': 78}
trench {'d': 36, 'y': 6, 'r': 6, 'm': 10, 'L': 4, 'p': 2}
The trench manager's final cost is: 120. The safety condition was satisfied.
Your final cost is: 120. The safety condition was satisfied.

If you do not, try to debug what's wrong on your side. If problems persist, please reach out.

Including Your Client Code

Under the clients folder you will find three files. You should only care about two of them:

• submarine_captain_client.py: You will insert your submarine client code in the your_algorithm() function. You don't need to touch anything else. Follow the instructions detailed in the docstring.
• trench_manager_client.py: You will split up your trench client code into two functions: send_probes and choose_alert. I split it up into two separate functions for modularity purposes and the fact that they do have different logic. You don't need to touch anything else. Follow the instructions detailed in each docstring.

Playing a Manual Game

In order to play a manual game (i.e., you enter in your moves every time using stdin), use the --manual flag. It has the following options:

• --manual both you play the submarine captain and trench manager manually
• --manual sub you play the submarine captain manually
• --manual trench you play the trench manager manually

So, when you'd run the python code, you'd just run: python3.6 run_game.py --manual both for instance.

Modifying the Initial Game State

If you want to test out various initial game state values, you only need to modify one line of code.

In run_game.py on line 24, I initialize controller = GameServer(). To change the game state, inject any of the following parameters into GameServer: d, y, r, m, L, p, and/or gui.

controller = GameServer(d=45, y=10, r=50, m=3, L=10, p=3, gui=False)

Running the Web Application

Okay, so you want to see the awesome UI. It's stored under the submarine-vis folder. However, you do need to install some packages (it's written in React.js):

1. Install Node.js: for Macs with homebrew installed, it's as easy as: brew install node. Or, you can visit the node website: https://nodejs.org/en/download/
2. Install yarn: for Macs with homebrew installed, it's as easy as brew install yarn. Or, you can visit the yarn website: https://yarnpkg.com/lang/en/docs/install
3. cd into submarine-vis
4. Run yarn install. This will install all of the needed node dependencies
5. Afterwards, run yarn start and the app should open up in a new tab. It's hosted on localhost:3000
6. To have the GameServer provide an endpoint pass True to the gui optional parameter. ie. GameServer(gui=True). This will provide an endpoint on localhost:8000

TIPS: The players will not be allowed to connect until the gui has connected to the GameServer. This means you should change your run_server.py to sleep for a longer amount of time prior to connecting to the game server.

You can independently open the gui in a browser by going to the address localhost:8000... assuming you're running the react app on your computer. You could also run both of your clients independently and connect to the game server after you connect the gui.