Lab 6: Representation Invariants and the JobManager

Overview

In Lab 5, you designed and programmed your own implementation of the JobManager abstract data type (ADT). In this lab, you're given a buggy implementation of the JobManager, and you're tasked with formulating its representation invariant (RI) and programming this RI in the form of a checkRep method, which ensures all conditions of the RI are met. Using your checkRep method, you should find & fix all the bugs in the given implementation. You are not allowed to change the representation (fields) of the implementation in any way.

The JobManager ADT Operations (Same as Lab 5)

A JobManager is seeded with n (the number of jobs it is managing). Each of these n jobs are initially “unassigned” (i.e., not assigned to a robot). One can then add Robots, assign jobs to the Robots, and move jobs among the Robots.

The principal operations that the JobManager ADT supports are as follows:

• Create Instance JobManager(int n) → Create a new JobManager instance with n jobs with all IDs in [1, n].

• Add robot addRobot(Robot robot) → Add a new robot to the pool of robots that can complete jobs.

• Check has robot hasRobot(Robot robot) → Check if robot exists in the pool of Robots maintained.

• Remove robot removeRobot(Robot) → Remove a robot from the pool of robots; all jobs assigned to the robot that was removed will be “unassigned”.

• Assign jobs assignJobs(Robot robot, int jobId) → Assign all unassigned jobs with id $\le$ jobId to robot.

• Check if job is assigned isAssigned(int jobId) → Checks if the job with id = jobId is assigned to any Robot in the pool.

• Move some jobs moveJobs(Robot srcRobot, dstRobot, int jobId) → Move all jobs with id $\le$ jobID from srcRobot to dstRobot.

• Move all jobs moveJobs(Robot srcRobot, dstRobot) → Move all jobs from srcRobot to dstRobot.

• Get assigned robot getRobot(int jobId) → Identify the Robot that is responsible for the job with jobId as its id.

• Get highest priority job assigned to a robot getHighestPriorityJob(Robot robot, int jobId) → Assuming that the job id indicates priority (higher id ⇒ higher priority), obtain the highest priority job assigned to robot with job id $\le$ jobId.

• Check equality equals(Object o) → checks for equality with another object based on a specified notion of representation equality.

• Get hashcode hashCode() → returns a hashcode that obeys the contract with equals() and that tries to minimize hash collisions through a Robot id-aware hash function.

Tasks

You are tasked with completing the following. You are not allowed to change the representation (fields) of the provided implementation in any way.

1. Implement the checkRep() method to throw an AssertionError if and only if the representation invariant (RI) for the provided implementation of JobManager is violated.

   • You cannot use the assert keyword in checkRep(). Instead, you must throw new AssertionError() if the representation invariant is violated. Otherwise, you will not pass the tests on PrairieLearn.

2. Run the provided test suite for checkRep() in CheckRepTests.java

3. Once you have checkRep() working, call this.checkRep() before all exit points in other methods like assignJobs.

4. Now, you can use checkRep() to help you find bugs in the implementation.

   • Hint: Look at which test cases were previously passing but started failing right after you added calls to checkRep() in the relevant methods.

5. Fix all the bugs in the implementation, and ensure you’re passing all provided test cases in JobManagerTests.java .

6. Once you’re confident in your solution, submit your code on PrairieLearn.

7. Answer the questions in the second part of the lab on PrairieLearn. (Or you can start with this if you like too.)

Grading

You will answer some questions on PrairieLearn and you will submit your implementation to PrairieLearn as well. This programming task is worth 6 points and the other questions on PrairieLearn are worth 3 points.