[tz] proposed changes for Win32 and a improved mktime() algorithm

Robert Elz kre at munnari.OZ.AU
Wed May 10 04:36:32 UTC 2017

    Date:        Tue, 9 May 2017 21:02:01 -0700
    From:        Paul Eggert <eggert at cs.ucla.edu>
    Message-ID:  <1c7aabd1-ae3d-e200-16d4-68d9ad50c2bc at cs.ucla.edu>

  | there is an alternative that would let us do both. 
  | We could use the glibc implementation of mktime.
  | The only downside that I can see is that it's LGPLed,

That is a very big downside, but ...

  | (Disclaimer: I wrote both the tzcode mktime and the glibc mktime.)

In that case you, or UCLA, own the copyright, and you can release it
on any licence(s) that you like...  Just because the gnusers insist
on everything being *GPL'd doesn't mean you cannot also release the
code under the MIT or BSD licence (or something similar.)

I haven't ever seen that version - I make it a point not to look at
*GPL'd code if I can possibly avoid it, but if manages to avoid all
knowledge of calendaring in its implementation, then given a suitable
licence it should be fine.

(Disclaimer; I wrote the original version of the BSD mktime() which
was the first I'm aware of to use the binary search method - though
the idea to do it that way was not mine, that came, I was told, from
Bob Kridle.)


More information about the tz mailing list