Some RegEx isn't working?

Been trying to use the \s to look for whitespace as part of my abhorrent regex abominations but it doesn't seem to be working? It seems to just be searching for the characters \ and s separately, or maybe just s by itself?

Here's the query looking for original descriptions with ^\s+$, which I would expect to turn up any commentaries without any text in them (whitespace only). However instead it turns up commentaries with... just the letter s in it? At first glance it looks like the search for empty wiki pages in forum #355364 works properly, but you can get the same results by replacing \s with literally any other character, because it's just looking for any commentaries with as little as 0 characters of the query, because that's how the asterisk wild card works.

It's not impossible that I'm missing something about how regex works in URLs, but if I am, I'd love to hear the secret piece of the puzzle, so I can continue my insane commentary fixing crusade. Thx

The backslash character classes all seem to be broken, but the POSIX character classes still work, so you can use that instead.

https://danbooru.donmai.us/artist_commentaries.json?search[is_deleted]=no&search[order]=id_asc&search[original_description_regex]=^[[:space:]]%2B$

External links

• https://www.regular-expressions.info/posixbrackets.html
• https://www.postgresql.org/docs/current/functions-matching.html#POSIX-BRACKET-EXPRESSIONS

Edit:

As I learned on Discord, Danbooru uses ERE mode for it's regexes, which doesn't support the backslash character classes. However, as I indiciated above, the POSIX bracket character classes do work.

Danbooru's regex is forced into the POSIX ERE syntax. This means all escapes don't work (see Postgres docs). You can probably use the character classes but I haven't verified this.

My powers grow, my knowledge enhanced. Thanksmuch!