Julun <HOST.HAIKU@xxxxxx> wrote: > i had a look at the implementation and i will try how far i come with > my > recent changes to the time prefs to update the timezone stuff. > > while going thru the code i have seen a todo for which i can propose > a > different algorithm if required. it is based on Julian day > calculation > and should be well proven. see attached diff. Thanks a lot! I've changed it a bit to get rid of d/2/3, and also renamed N/L to n/l according to our style guide. I've applied it in r22557. Bye, Axel.