>>>We have a feature freeze scheduled for either June 1 or
>whenever a major
>>>feature is completed, whichever is last.
>>>
>>>For Win32, we have open:
>>>
>>> fsync (Tom)
>>> relative installs (Bruce)
>>> timezone (Magnus)
>>>
>
>Is timezone still an open item? I thought it was done.
>Certainly we pass the regression tests.
Yes. I am working on two things here:
1) Replacing the TZ library for Unix as well, so we have the same code.
For one thing, it will be much better tested that way. This part is
pretty much done. This will also allow some changes to the APIs to get
rid of a lot of workarounds currently in the code (for example, tzset()
by default does not let the caller know if a valid timezone was
specified, it just sets it to GMT if an invalid one was. The code I have
now returns an error directly).
2) Pick upt he default OS timezone. This still need a little bit more
work. But the code as it stands on Win32 will give you GMT on startup
unless you specify a TZ in postgresql.conf. On Unix, it goes with the OS
timezone.
I expect to be done with this not too long from now, though.
//Magnus