Closed mcfilib closed 7 years ago
In addition to this, according to the gettext docs, it is not unreasonable to include further data after an @
symbol e.g. cs_CZ@corporate
. I appreciate this is unlikely to be the common case but was curious whether you had considered these two cases.
I've been able to work around this by explicitly setting locales=
myself. Seems reasonable.
In the
detect_locales
method, there's a regex that is used to determine whether a directory looks like a valid locale name.This is useful for detecting the common case i.e. when a locale is a primary code (
cs
) or a primary code followed by a secondary code (cs_CZ
) but when you have the character encoding e.g.cs_CZ.UTF-8
it fails to detect the locale.Is there a rationale behind this choice?