JeremyTubongbanua / WikiPathFinder

HackTheJob 2022 Submission - WikiPathFinder is a program that finds the fastest possible path from one wiki page to another by only clicking links.

WikiPathFinder

WikiPathFinder is a tool that tells you the quickest path from START_WKIPEDIA_URL to END_WIKIPEDIA_URL.

Table of Contents

• WikiPathFinder
• Table of Contents
• For Users
  • Background
  • How to Use
  • Example Usage (with DCP)
  • Example Usage (without DCP)
• For Developers
  • Installing Dependencies
  • Node Dependencies
  • How we use DCP
    • Using DCP Pros:
    • Using DCP Cons:
    • Without DCP Pros:
    • Without DCP Cons:

For Users

Background

WikiRacing is a commonly known game where 2 people start at the same START_WIKIPEDIA_URL and try to get to the same END_WIKIPEDIA_URL. Whoever gets to END_WIKIPEDIA_URL the fastest is the winner.

How to Use

1. Install all necessary node modules

• cheerio (for web scraping): npm i cheerio
• request-promise npm i request-promise
• dcp: npm i dcp-client
• prompt-sync npm install prompt-sync
• require() function: npm i --save-dev @types/node

2. Start terminal program via node out/index.js
3. Enter start URL (eg: https://en.wikipedia.org/wiki/Gamer)
4. Enter end URL (eg: https://en.wikipedia.org/wiki/Google)
5. Use DCP (y/n) ? (eg: y)
6. Obtain results. ([] is path not found)

Example Usage (with DCP)

[Using DCP]

1. Start terminal program via node out/index.js
2. Enter start URL (eg: https://en.wikipedia.org/wiki/Gamer)
3. Enter end URL (eg: https://en.wikipedia.org/wiki/Google)
4. Use DCP (y/n) ? (eg: y)
5. Obtain results.

Result:
START https://en.wikipedia.org/wiki/Gamer
TO https://en.wikipedia.org/wiki/Gamer
TO END https://en.wikipedia.org/wiki/Google

Example Usage (without DCP)

[Without DCP]

1. Start terminal program via node out/index.js
2. Enter start URL (eg: https://en.wikipedia.org/wiki/Among_Us)
3. Enter end URL (eg: https://en.wikipedia.org/wiki/University_of_Guelph)
4. Use DCP (y/n) ? (eg: n)
5. Obtain results.

Result
START https://en.wikipedia.org/wiki/Among_Us
TO https://en.wikipedia.org/wiki/Cover_art
TO https://en.wikipedia.org/wiki/Illustration
TO https://en.wikipedia.org/wiki/Ben_Shneiderman
TO END https://en.wikipedia.org/wiki/University_of_Guelph

For Developers

Installing Dependencies

• Update all dependencies listed in package.json: npm install

Node Dependencies

If npm install doesn't work:

• type script npm i -g type-script
• cheerio (for web scraping): npm i cheerio
• request-promise npm i request-promise
• request-promise types npm i --save-dev @types/request-promise
• prompt-sync npm i prompt-sync
• prompt-sync types npm i --save-dev @types/prompt-sync

How we use DCP

The original plan was to do all of the web scraping and run the findPath algorithm on DCP. However, we were told by a mentor that DCP could not use the internet. So we were severely limited with integrating DCP into our idea. So for now, we give the user 2 options (to use DCP and to not use DCP) where each have their own pros and cons:

Using DCP Pros:

• findPath algorithm is super fast

Using DCP Cons:

• Have to do all web-scraping on system beforehand because no internet access.
• The way our algorithm works: it was either using 1 worker or 300+ workers (300+ is how many URLs are in the first wikipedia start page).

Without DCP Pros:

• Web Scraping is dynamic.
• The user can enter any possible wikipedia links.

Without DCP Cons:

• Takes very long to path find
• Web scraping speed is limited to user's system