chore: Updated lint rule suppression comment

[jsumners-nr]

This PR resolves #2859.

I worked on a couple of different algorithms in an attempt to get around the rule violation, but the regular expression is our best option. The simplest algorithm I was able to devise still used regular expressions, but ended up being slower (despite being easier to follow). I have attached my benchmark below.

Benchmark

'use strict'

const assert = require('node:assert')

const tests = [
  ['use test_db;', 'test_db'],
  ['USE INIT', 'INIT'],
  ['use test_db', 'test_db'],
  ['use test_db;;;;;;', 'test_db'],
  ['            use test_db;', 'test_db'],
  ['use            test_db;', 'test_db'],
  ['use test_db            ;', 'test_db'],
  ['use test_db;            ', 'test_db'],
  ['use `test_db`;', '`test_db`'],
  ['use `☃☃☃☃☃☃`;', '`☃☃☃☃☃☃`'],
  ['use cxvozicjvzocixjv`oasidfjaosdfij`;', null],
  ['use `oasidfjaosdfij`123;', null],
  ['use `oasidfjaosdfij` 123;', null],
  ['use \u0001;', null],
  ['use oasidfjaosdfij 123;', null]
]
const iterations = 10_000

const regexA = process.hrtime.bigint()
for (let i = 0; i < iterations; i += 1) {
  for (const [input, expected] of tests) {
    const found = extractWithRegex(input)
    assert.equal(found, expected)
  }
}
const regexB = process.hrtime.bigint()

const withoutA = process.hrtime.bigint()
for (let i = 0; i < iterations; i += 1) {
  for (const [input, expected] of tests) {
    const found = extractWithoutRegex(input)
    assert.equal(found, expected)
  }
}
const withoutB = process.hrtime.bigint()

console.log('with regex time:', regexB - regexA)
console.log('without regex time:', withoutB - withoutA)
console.log('with < without:', (regexB - regexA) < (withoutB - withoutA))

function extractWithRegex(input) {
  const match = /^\s*use[^\w`]+([\w$_\u0080-\uFFFF]+|`[^`]+`)[\s;]*$/i.exec(input)
  return (match && match[1]) || null
}

function extractWithoutRegex(input) {
  const dbName = input
    .trim()
    .replace(/use\s+/i, '')
    .replace(/;+$/, '')
    .trim()
  if (/^`?[\w$_\u0080-\uFFFF]+`?$/.test(dbName) === true) {
    return dbName
  }
  return null
}

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 97.35%. Comparing base (cc65763) to head (3f6202d).
Report is 6 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2895      +/-   ##
==========================================
- Coverage   97.39%   97.35%   -0.05%     
==========================================
  Files         310      310              
  Lines       47512    47512              
==========================================
- Hits        46276    46254      -22     
- Misses       1236     1258      +22     

Flag	Coverage Δ
integration-tests-cjs-18.x	73.03% <0.00%> (ø)
integration-tests-cjs-20.x	73.03% <0.00%> (ø)
integration-tests-cjs-22.x	73.06% <0.00%> (ø)
integration-tests-esm-18.x	49.78% <0.00%> (ø)
integration-tests-esm-20.x	49.79% <0.00%> (ø)
integration-tests-esm-22.x	49.83% <0.00%> (ø)
unit-tests-18.x	88.99% <100.00%> (ø)
unit-tests-20.x	88.99% <100.00%> (ø)
unit-tests-22.x	89.00% <100.00%> (ø)
versioned-tests-18.x	79.27% <100.00%> (-0.19%)	⬇️
versioned-tests-20.x	?
versioned-tests-22.x	?

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mrickard]

Am I reading this correctly? It seems as though the regex has changed from the previous commit.

[jsumners-nr]

Notice that the second suppressed rule has been removed from suppression.

[mrickard]

I'm thinking about the underscore (\w$_\u0080 to \w$\u0080), which has been in that regex for a long time prior to this change.

[jsumners-nr]

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_expressions/Character_classes