regex in msg_command_parse/1

[Marqin]

I have some questions about this regex: ~r/!(.\w*:\w*:\w*|.\w*:\w*|.\w*){1}/

• Is it intentionaly to not catch unicode characters? (lack of u flag after regex)
• Why .\w*? I mean, why it can begin with any not-newline character, but then it has to be only word chars? Why eg. not just \w*?
• Why is it using this regex and not just String.starts_with?/2 ? That way it would be easier to modify it to be able to use provided char instead of ! (I'd some problems when interpolating variable inside regex).

[Marqin]

Also, isn't that {1} obsolete here? Isn't it by default matched only once?

[rmcafee]

I should have some time to finally do some refactoring after this upcoming week. I'll have to see what that will all include.

To make sure I give clear context to the library. It went from a scratch project needed within my group to something more verbose to open source. So some refactoring is expected.