tz VS lunacy

A link was added from "Europe/Kiev" to "Europe/Kyiv" in the included-by-default backward file [0], so that any user that doesn't exclude that file will simply treat the old name as an alias for the new name.

[0] https://github.com/eggert/tz/commit/e13e9c531fc48a04fb8d064a...

timeZone: Determines the current timezone to use to display the time e.g. America/Los_Angeles. Full list can be found on IANA time zone database

"Eire" is in there, for instance, to deal with software that assumes that the "is_dst" half of the year is during the (northern) summer, but Ireland technically does it the other way around -- a distinction relevant only to computers.

https://github.com/eggert/tz/blob/c3e966c59b02b1f47f0b7b0e4a...

The only other timezone that currently has a non-1h offset for DST -- Ireland's is -1 hours -- is Australia/Lord_Howe, which has a 30-min positive leap.

Prof Paul Eggert is currently the editor and coordinator of the Time Zone Database of IANA, which enables timestamps on official documents and photos.

I've looked at it, but wanted to work with the tz source repository directly (I think local-time gets their zone files from Ubuntu). Also getting zic running in a portable way seemed too much of a hassle. The text file format is not all that complicated and documented in the zic manual pretty well. This approach is also chosen by the JDK as far as I can tell.

Subscribe to tz-announce for more fun: https://www.iana.org/time-zones

Unfortunately, you just wait. The person(s) at IANA who manages The Time Zone Database is certainly aware, and from there it "just" has to percolate down to Google and your phone.

most systems use https://www.iana.org/time-zones

And have you looked at the TZ database? My God it's a hot mess, and for good reason: human governments are terrible at creating code. Worse even than beginning CS students, because at least CS students know they don't know what they're doing, while politicians are both stupid and arrogant enough to think they're not stupid.

I like having a language which can quickly do a lot more than what BASH/Sed/AWK can do fit in a 116,224 byte sized Windows 32-bit binary, but I have had to do some hacking, both making my own custom compile of Lua and making a number of Lua functions to fill in the blanks while keeping things small and trivial to install on either native Windows, as a Cygwin program, or as a program in Linux or pretty much any POSIX compatible *NIX clone (including MacOS).

Now, last time I researched this, there still isn’t a “int64_t time_64bit()” style system call in the Linux API so that newly compiled 32-bit binaries can be Y2038 compliant by using “time_64bit()” instead of “time()”.

Keep in mind that we also need 64-bit versions of “strftime()”, “localtime()”, “ctime()”, “gmtime()”, “asctime()”, and “mktime()”, and so on probably with strdtime_64bit(), localtime_64bit(), and so on. In my piece of code, I could had written my own implementations of all these — note that time zone management is a really hard problem because of things like changing daylight savings time and the Indiana time mess^2 — but I decided to simply just have my Lua fork have no timestamp support beyond a simple 64-bit timestamp. There’s also the issue with stat() and the struct timespec returned for file creation/modification/access/status change timestamps^3, so I just don’t have stat() look at timestamps.

[1] https://github.com/samboy/lunacy

[2] https://en.wikipedia.org/wiki/Time_in_Indiana

[3] To ruin a classic sadistic interview question for sys admin roles, Linux these days returns both the modification time and the mostly useless “status change” timestamp. Facebook once decided to not move forward because I said that file timestamp was “modification time” and not “status change”; if Facebook is still asking that question, their knowledge is out of date.