Assignment

Updates

---

1. Object Oriented Design
2. Installation & Deployment

Object Oriented Design

---

models
    Acts
    Event
    Schedule
    Stages

In model package, there are 4 main models.
Acts is used to store details of each act.
Event is the object which stores the info of the entire Fuji Rock Festival.
Schedule is used for storing the event start and end time as a configuration
Stages contains a list of Acts which will be hold in current stage.

services
    Outputs
    Utils

Outputs and Utils are the interfaces of formatting outputs for the event and common & inputs reading methods
The reason of why I make those as interfaces are in order to extend the current program to meet different usages requirements. In current assignment, we read inputs from file and use system.out to print plans. If we get a requirement to ask us using system.in as inputs.
Then we just need to have another implementation in serviceImpl package.

Installation & Deployment

---

IMPORTANT: In case of the online editor doesn't save the correct version of my solution. I create a Backup.java file in the project. If you get any error from the online editor, please copy and paste the code from the backup file and use the test cases in testcases folder. Thank you!

1. There is no need for installation & Deployment in your local machine, in order to run the code. Here is the online editor where you can run the program.

2. You can also download the project from git and run in local.
   Run Configuration: main.java Environment: Java 17 and OpenJDK 17.0.1 The output for local will look like this

Design pattern

---

In order to match the online editor's format. The Solution class is put inside Main.java. Ideally it should be in another separated file.

Configurations

In order to make the code extendable. I make 2 configuration parameters for it.

1. ranges: an array which stores the time and length info to schedule a stage. To make a ranges like this, we can simply change the value in it and make changes to the event start/end time {9, 180, 240} this means the group of event will be started at 9am and lasts 180 to 240 mins
2. schedule: a map which stores the info of default events information

        {
          9 : "12:00PM Lunch",
          13 : "05:00PM Main models.Event"
        }

To extend or modify the schedule, we can simply add more items inside ranges and schedule to let the program find the plan for more stages and different time period

        //The schedule we would like to fill in.
        // {13, 180, 240} - the config info of a schedule time period
        // while means 13 - 1:00PM; 
        // 180, 240 - the protential time period, the event may finish in 3hr to 4hr
        int[][] ranges = {{9, 180, 180}, {13, 180, 240}, {9, 180, 180}, {13, 180, 240}}; 

        // schedule map will look like this
        // 9 : "12:00PM Lunch"
        // 13 : "05:00PM Main models.Event"
        Map<Integer, String> schedule = new HashMap<>();
        schedule.put(9, "12:00PM Lunch"); 
        schedule.put(13, "05:00PM Main models.Event");

Also there is another maintenance benefit to make those as configurations. Those data can be created by another program and passed to this code. We don't need to make any changes to this part of code if the schedule data format gets changes somewhere.

Program Process

The program will be executed in the following order

1. Take the system input and convert it into acts list -> [act lengths, act title], like this:

     [["30", "Bob Dylan & His Band"], ["45", "Post Malone"], ["60", "Mariya Takeuchi"] ...]

2. DFS fill in plans, based on schedule configuration. The program will use DFS to try different combinations of acts to match schedule requirements from configurations. And use a boolean[] to avoid duplications.

3. Print out the plan results. If we are able to find a valid plan from the input, the program will print out the information As in assignment requirements. The plan is only for 2 stages. So it will be stored like this

[List[Morning], List[Afternoon], List[Morning], List[Afternoon]]

Sample program output

Stage 1
09:00AM Bob Dylan & His Band 30min
09:30AM Burial 30min
10:00AM The Drones 30min
10:30AM Dr. Dre 30min
11:00AM Junko Ohashi 30min
11:30AM Gorillaz 30min
12:00PM Lunch
01:00PM Ross From Friends 30min
01:30PM Post Malone 45min
02:15PM Radwimps 45min
03:00PM Mariya Takeuchi 60min
05:00PM Main models.Event
-----------------------------
Stage 2
09:00AM Scandal 45min
09:45AM Talking Heads 45min
10:30AM Sambomaster 45min
11:15AM Hong Kong Express 45min
12:00PM Lunch
01:00PM One Ok Rock 60min
02:00PM Thom Yorke Tomorrow's Modern Boxes 60min
03:00PM The Stone Roses 60min
05:00PM Main models.Event
-----------------------------

Test cases

This the act data from assignment.

Mariya Takeuchi 60min
Post Malone 45min
Bob Dylan & His Band 30min
Radwimps 45min
Scandal 45min
Special announcement
One Ok Rock 60min
Talking Heads 45min
Burial 30min
The Drones 30min
Sambomaster 45min
Thom Yorke Tomorrow's Modern Boxes 60min
The Stone Roses 60min
Hong Kong Express 45min
Dr. Dre 30min
Junko Ohashi 30min
Aphex Twin 60min
Gorillaz 30min
Ross From Friends 30min

Error Handling

The program doesn't do much on error handling. It assumes the input passed to it must be valid at this point. However, I do add empty line check to let it skip empty lines from inputs. Like this

Dr. Dre 30min


Junko Ohashi 30min

Aphex Twin 60min
Gorillaz 30min

Time complexity

* The overall time complexity is o(n^2), which is calculated by 
* n - the number of different combinations of events for 2 stages.
* k - the number of events inside one combination(one solution).

Potential improvements

1. The current the program will be able to find one valid plan. However, we can make some changes on

if (start >= this.stageNum || time > ranges[start][1] || this.plans.size() >= this.stageNum) return;

and add caching to have multiple different plans.

2. For the input like Special announcement, currently I assume we can only have one during the event. Since it doesn't make sense if we have 2 stages which are only filled in Special announcement and no actual shows. However, we can make this as another configuration variable, to set a limit for it during the event. Like

Special announcement : 5

3. Error handling improvements There are always room for improving error handling. like handling invalid inputs, configurations issue