The prefer-switch rule can behave strangely in certain situations

[kv-bh]

The prefer-switch rule is applied in some situations where it doesn't really make sense.

Example code:

  // eslint-disable-next-line unicorn/prefer-switch
  if (typeof document.exitFullscreen === 'function') {
    await document.exitFullscreen();
  } else if (typeof document.webkitExitFullscreen === 'function') {
    await document.webkitExitFullscreen();
  } else if (typeof document.mozExitFullscreen === 'function') {
    await document.mozExitFullscreen();
  }

Example after "fixing":

  switch ('function') {
    case typeof document.exitFullscreen: {
      await document.exitFullscreen();

      break;
    }
    case typeof document.webkitExitFullscreen: {
      await document.webkitExitFullscreen();

      break;
    }
    case typeof document.mozExitFullscreen: {
      await document.mozExitFullscreen();

      break;
    }
  // No default
  }

While this may technically work, it doesn't seem like a situation where this rule should be applied. The result is unintuitive and it also causes TypeScript's type inferences to break in this situation.

[fisker]

Ha, this is the first issue about switch (constant), and someone don't like my little secret, much longer than I expected to see this issue.

When I implemented this rule, I was thinking "It's fixable, so why not?"

eslint-plugin-unicorn/test/prefer-switch.js

Lines 108 to 113 in e8798da

	// Compare to constant
	outdent`
	if (true === foo) {}
	else if (bar.bar === true) {}
	else if (true === baz()) {}
	`,