It is _much_ faster than the text tzdb that we currently ship. A few things stand in the way of this: - [ ] Being able to set the binary tzdb path at runtime https://github.com/HowardHinnant/date/pull/639 - [ ] Allow for specification of a binary tzdb path on Windows, and ensure the binary parser works on Windows https://github.com/HowardHinnant/date/issues/564 - [ ] Extend the transition table indefinitely when using the binary tzdb https://github.com/HowardHinnant/date/issues/563 - [ ] Ensure that we use the (no longer default) "fat" format when generating the binary tzdb, unless this gets fixed https://github.com/HowardHinnant/date/issues/614
It is much faster than the text tzdb that we currently ship.
A few things stand in the way of this:
USE_OS_TZDBHowardHinnant/date#563