Fuzzier Quote Search

[nstringham]

User Story

As a user of the compsci bot I would like to be able to find quotes without needing to type a piece of the quote exactly.

Bug Report

Steps to Reproduce

1. type !quote Tehranipour
   • observe that the bot returns a quote that contains "Tehranipour"
2. type !quote tehranipour

Expected Behavior

that the bot returns a quote that contains "Tehranipour"

Actual Behavior

the bot returns an error

Code Analysis

this line performs an exact case sensitive search for the query in the quote https://github.com/Nanosplitter/CompsciBot/blob/b8b60cd7b8082fee6cfaa38150f0fd7c285893f9/noncommands/quotes.py#L29

Suggested Improvements

I have 3 proposals that each fix this problem

Option A - Case Insensitive Match

replace keywords in line with something case insensitive such as keywords.lower() in line.lower()

Option B - Fast Fuzzy Search

replace keywords in line with a primitive fuzzy search algorithm/library to allow user to make typos and spelling mistakes

Option C - Fuzzy Search with index

use a proper fuzzy search algorithm with a index that is generated once and used for future searches

note: that this solution works well if the quotes are stored in a database

Comparison

	Case Insensitive	Misspellings and Typos	Performance Impact	Complexity
Option A	✅ yes	❌ no	🟨 slightly slower	🟩 not complex
Option B	✅ yes	✅ small mistakes	🟥 much slower	🟨 moderately complex
Option C	✅ yes	✅ medium mistakes	🟩 much faster	🟥 complex

[coxender]

I think in the long run option C would be better saved until after we migrate quotes to a database (unless that has already happened and I am unaware ). This issue should still be open then even if option A gets merged into main.

[Nanosplitter]

The quotes are now in a DB so option C is now much easier if you want to do it