Solve the maze with your mazerunner in bash

By . Latest revision .

You will get a finished server, written in Node.js, and a RESTful API to the server. The server implements a maze. The server is ready and you can test run it via the curl command.

Your task is to build a bash client for the server, according to a requirements specification. Your client should use the server to resolve the maze.



Read briefly on Wikipedia about what a [maze] ( can be.

This is what it may look like when you solve the maze with your script:

Note that the file extension should be mazerunner.bash.

#About the server maze

The server maze is in the course repot. There is all the source code and a specification of the server’s API.

Start by getting acquainted with the server’s API.

Here’s how to start the server.

# Go to the course repo
cd example/maze
node index.js

You can test the maze server with curl. Like this.

Please study the source code of the maze server. Had you been able to write it yourself?

#Make a copy of Maze

First, make a copy of the code in example/maze/. Save all your files in the directory me/kmom04/maze.

# Go to the course repo
cp -ri example/maze/{,index.js,maze.js,maps,router.js} me/kmom04/maze/

Now you are ready to start your own variant of the maze server.

You start the server with $ node index.js.

#To save the game ID to file

Your client needs to remember the game ID and which room you are in. You save that information easiest in a file. To avoid having to handle JSON with bash, the server has the opportunity to deliver the responses as a comma-separated string.

Try running the following commands against the server and you will see the difference.

curl localhost:1337/map
curl localhost:1337/map?type=csv

It is therefore ?type=csv that can make it easier for your bash client who will need to parse the content.


The requirements are two-parted. The first set is mandatory to pass the assignment. The second part is an optional assignment if you want to challenge yourself.

#Bash script for solving the maze (part 1)

  1. Create a script called mazerunner.bash in the folder maze/. Make the script executable.

  2. Use the API to add the following features to the script. The script should always return a response if all went well or not.

Command What should happen?
./mazerunner.bash init Initiate a game and save the game ID in a file.
./mazerunner.bash maps Show which maps are available to choose from.
./mazerunner.bash select <#map> Select a specific map by number.
./mazerunner.bash enter Enter the first room.
./mazerunner.bash info Show information about the room.
./mazerunner.bash go north Go to a new room, if the direction is supported.
./mazerunner.bash go south Go to a new room, if the direction is supported.
./mazerunner.bash go east Go to a new room, if the direction is supported.
./mazerunner.bash go west Go to a new room, if the direction is supported.

Here is how it may look when you are done.

#Bashscript in loop (part 2 - optional)

  1. Extend the functionality of mazerunner.bash so that everything happens in a loop when you start the program with ./mazerunner.bash loop. The script should start by initiating a new game and showing what maps are available. The player can then select a map and the player enters the first room. Then the loop continues and waits for the player to enter the direction north, south, east, west, or help for a help text or quit to finish.

This is what it might look like.

Validate your mazerunner.bash script by doing the following commands in the course directory in the terminal.

# Go to the course repo
$ dbwebb validate maze

Correct any errors that pop up and publish again. When it looks green you are done.

#Revision history

  • 2019-09-03: (A, lew) First edition.

Document source.

Category: unix.