https://github.com/inimaz/millenium-falcon-developer-test
https://github.com/inimaz/millenium-falcon-developer-test
Last synced: 2 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/inimaz/millenium-falcon-developer-test
- Owner: inimaz
- Created: 2023-05-23T18:17:50.000Z (about 2 years ago)
- Default Branch: master
- Last Pushed: 2024-05-04T11:45:59.000Z (about 1 year ago)
- Last Synced: 2025-02-15T19:28:09.581Z (4 months ago)
- Language: TypeScript
- Size: 21 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Calculate the odds
To see full info about this, check [Exercise Readme]('./docs/README.md').
> **Note**: See the [TODO](TODO.md) page to see the things that are yet to be developed.
# How to use it
## CLI
Install Nodejs and NPM
go to the cli folder and install the cli.
```
cd cli
npm install -g .
```
Now you can run it doing
```
give-me-the-odds
```
## BE
```
cd backend
npm install
npm start
```
There are 2 endpoints:
- GET **/routes** gives you all the routes
- POST **/odds** with body
```
{
"countdown": 8,
"bounty_hunters": [
{"planet": "Hoth", "day": 6 },
{"planet": "Hoth", "day": 7 },
{"planet": "Hoth", "day": 8 }
]
}
```
returns the odds in the format
```
{
"odds": 0
}
```
### ENV VARIABLE
`MILLENIUM_FALCON_CONFIG_FILE`:string. (Optional) If mentioned, it directs to the local json file where the millenium-falcon.json file is stored.
# About the algorithm
We started using dijkstra algorithm to find the shortest path in between 2 planets. This works well for minimizing distance. But in this case we want to minimize distance AND odds of being caught. So we have to maximize for a ficticious distance that is a combination of the 2 factors.
For now the combination used is
```
newDistance = sqrt((oldDistance)^2 + (X)^2)
```
where this X is Infinity if there is a bounty hunter and 0 if there is none. In the future we can try to find the ideal X for each case so the ship takes sometimes the path even though there might be hunters.
With this new distance, we use the following algorithm:
1. we calculate the shortest path ignoring the hunters
2. If it does not exist --> there is no possible path
3. If there are no hunters in this path --> this is the wanted path let's call it `shortPathNoHunters`
4. If there are hunters in this path, calculate the shortest path avoiding the hunters
5. If this path exists --> this is the wanted path let's call it `shortPathWithHunters`
6. If the path does not exist --> we know the number of capture tries of `shortPathNoHunters`, we know that this is the worst case.