Different output result with different word order

[templarundead]

#Example 1
input: ghost frost pos
output: ghost|frost|pos
expected output: (?:(?:fr|gh)ost|pos)
Example 2
input: pos ghost frost
output: (?:gh|fr)ost|pos
expected output: (?:(?:fr|gh)ost|pos)

[mathiasbynens]

Hmm… So let’s sort the input words by length ([...word].length) and then lexicographically?

[vegeta897]

That happens to work for ghost frost pos, but with my set of 100+ junk words, sorting from longest to shortest, then by Z-A, happens to generate the shorter RX (though not by much, 594 chars vs. the 608 with ascending length and A-Z). I also tried reversing the characters of each word when sorting A-Z and Z-A but this made no difference.

An interesting optimization problem. It would help if I actually looked at what regexgen was doing under the hood 😛