- tz - lists.iana.org

zdump man page misleading?
by Martin Burnicki Feb. 26, 2021

Feb. 26, 2021

Hi, the zdump.8.txt file generated from the current github code says: -c [loyear,]hiyear ... Cutoffs are at the start of each year, where the lower-bound timestamp is *exclusive* and the upper is *inclusive*; for example, -c 1970,2070 selects transitions after 1970-01-01 00:00:00 UTC and on or before 2070-01-01 00:00:00 UTC. The default cutoff is -500,2500. and $ ./zdump -V -c 2021,2022 Europe/Berlin Europe/Berlin Sun Mar 28 00:59:59 2021 UT = Sun Mar 28 01:59:59 .. Europe/Berlin Sun Mar 28 01:00:00 2021 UT = Sun Mar 28 03:00:00 .. Europe/Berlin Sun Oct 31 00:59:59 2021 UT = Sun Oct 31 02:59:59 .. Europe/Berlin Sun Oct 31 01:00:00 2021 UT = Sun Oct 31 02:00:00 .. (lines shorted to avoid line-wrapping) So the lower year is *inclusive* and the upper year is *exclusive*, and from my understanding these words should be swapped in the text of the man page. Or am I missing something? If I'm correct, should I provide a patch or pull request for the git repo? Martin -- Martin Burnicki Senior Software Engineer MEINBERG Funkuhren GmbH & Co. KG Email: martin.burnicki(a)meinberg.de Phone: +49 5281 9309-414 Linkedin: https://www.linkedin.com/in/martinburnicki/ Lange Wand 9, 31812 Bad Pyrmont, Germany Amtsgericht Hannover 17HRA 100322 Geschäftsführer/Managing Directors: Günter Meinberg, Werner Meinberg, Andre Hartmann, Heiko Gerstung Websites: https://www.meinberg.de https://www.meinbergglobal.com Training: https://www.meinberg.academy

3 3

time zones only ?
by William Karlsson Feb. 25, 2021

Feb. 25, 2021

Is there a simple way to compile only the timezones ? i.e. US/Central etc... No cities ? -- // Karlsson

3 2

New IANA-timezone support date/time library
by Kerry Shetline Feb. 24, 2021

Feb. 24, 2021

I’ve just finished an extensive update of a JavaScript/TypeScript date/time library that I’d previously used only for my own astronomy website, skyviewcafe.com <http://skyviewcafe.com/>. I’ve done enough updates to this library now that it may be of interest to many other software developers, especially if they want to handle the odd edge cases that exist in timezone handling, like 47-hour days and missing days and the like. One unique features is that it's possible for users to get live updates of the tz database, which usually don’t become available to applications without, at the very least, restarting an application, if not rebooting one’s computer. Libraries in this category are hardly unique, but one of the most popular libraries, Moment.js, is being phased out, so this seemed to be to be a good time to introduce a new one. Among the (possible) unique features, you can do things like find the specific moments of DST transitions and other transitions, like UTC offset changes, and get descriptions of these changes, like this: new DateTime('2021-11-07', 'America/New_York').getDiscontinuityDuringDay() → { start: '02:00:00', end: '01:00:00', delta: -3600000 } // fall back! // As soon as it’s midnight on the 30th, it’s instantly midnight on the 31st, erasing 24 hours:<br> new DateTime('2011-12-30', 'Pacific/Apia').getDiscontinuityDuringDay()) → { start: '00:00:00', end: '24:00:00', delta: 86400000 } // As soon as it’s midnight on the 31th, turn back to 1AM on the 30th, adding 23 hours to the day:<br> new DateTime('1969-09-30', 'Pacific/Kwajalein').getDiscontinuityDuringDay() → { start: '24:00:00', end: '01:00:00', delta: -82800000 } My updated library can be found here: https://www.npmjs.com/package/@tubular/time <https://www.npmjs.com/package/@tubular/time> Besides the IANA timezone support, this library also features the ability to handle variable switch-over from the Julian to the Gregorian calendar, which can be of helpful to those handling historical dates.

8 9

Re: [tz] Gratuitous optimization
by Paul Eggert Feb. 23, 2021

Feb. 23, 2021

Thanks, I installed your performance fix via the attached proposed patch. The performance bug seems to have been introduced by my commit 30364485a6fdc0333f48905db387e07abb70bac1 (2013-09-08) "Improve the support for perpetual DST" which first appeared in tzdb 2013e.

1 0

Re: [tz] New IANA-timezone support date/time library
by Kerry Shetline Feb. 23, 2021

Feb. 23, 2021

On 2021-02-22, at 14:25, Paul Gilmartin <PaulGBoulder(a)AIM.com> wrote: > But the "Help" link (<https://skyviewcafe.com/help/>) fails with 404. The “Help” link is working now, so I’m not sure what the issue was at the time you clicked on it. > I don't know whether you've followed the "Kiev"/"Kyiv" debate on tz(a)iana.org. > But I was curious what skyviewcafe.com does about it. I read a little of that, but haven’t followed it. At least the pronunciation of Kyiv isn’t an issue! ;) My geographic data base finds the Ukrainian capital by either spelling, so that’s not a problem. I’m not sure if showing “Europe/Kyiv” in my interface would be confusing or not, since that’s not the existing IANA name for the associated timezone. Users are generally expected to click “Find”, and find cities by name, thus getting longitude, latitude, and timezone automatically without fussing with a manual selection, so the IANA names aren’t quite so important. I also intend to update my manual interface to make timezone selections by standard UTC offset, and then a list of zones in that offset, as the default way to make a manual timezone selection, with region/zone being a secondary option. > I tried to enter > "50°25'N 30°31'E". I learned: > o It's tedious to enter such coordinates digit-by-digit. > o I found no way to refresh the page with the coordinates entered. It’s definitely tedious compared to just looking up a city and getting lat/long automatically. But if one wants to enter info manually, I’m curious what you’d prefer as a means of input. The manual entry field also supports rolling digits up and down, which is a fun way to explore how the sky changes as you change latitude and longitude. > Tz, trying to pivot away from the debate, says that internal directory > and filenames should not be obvious to users. Thus, applications should > not show me "America", "Denver", or "Kiev”. I hadn’t really thought of those names like “America/New_York” as file names, but as names for timezones as abstract timezone entities. In fact, there is no “America/New_York” file in the tzdb. There’s a file named “northamerica”, and inside that file, a zone entry named "America/New_York”. Perhaps that’s still something a bit too rawly technical to be considered user-friendly, however. -Kerrt

3 2

[PATCH 1/5] Don't cite Howse where no longer used
by Paul Eggert Feb. 23, 2021

Feb. 23, 2021

--- africa | 3 --- asia | 3 --- 2 files changed, 6 deletions(-) diff --git a/africa b/africa index 28168cf..fca9af7 100644 --- a/africa +++ b/africa @@ -30,9 +30,6 @@ # Milne J. Civil time. Geogr J. 1899 Feb;13(2):173-94. # https://www.jstor.org/stable/1774359 # -# A reliable and entertaining source about time zones is -# Derek Howse, Greenwich time and longitude, Philip Wilson Publishers (1997). -# # European-style abbreviations are commonly used along the Mediterranean. # For sub-Saharan Africa abbreviations were less standardized. # Previous editions of this database used WAT, CAT, SAT, and EAT diff --git a/asia b/asia index ed94413..ac7def8 100644 --- a/asia +++ b/asia @@ -34,9 +34,6 @@ # Byalokoz EL. New Counting of Time in Russia since July 1, 1919. # (See the 'europe' file for a fuller citation.) # -# A reliable and entertaining source about time zones is -# Derek Howse, Greenwich time and longitude, Philip Wilson Publishers (1997). -# # The following alphabetic abbreviations appear in these tables # (corrections are welcome): # std dst -- 2.27.0

2 6

Some time zone history (2)
by P Chan Feb. 21, 2021

Feb. 21, 2021

Bahamas There were two periods of DST observed in 1941-1945: 1941-05-01 midnight to 1944-12-31 and 1945-02-01 to 1945-10-17 midnight. War Time Order 1942 [1942-05-01] and War Time (No. 2) Order 1942 [1942-09-29] Appendix to the Statutes of 7 George VI. and the Year 1942. p 34, 43 https://books.google.com/books?id=5rlNAQAAIAAJ&pg=RA3-PA34 https://books.google.com/books?id=5rlNAQAAIAAJ&pg=RA3-PA43 War Time Order 1943 [1943-03-31] and War Time Order 1944 [1943-12-29] Appendix to the Statutes of 8 George VI. and the Year 1943. p 9-10, 28-29 https://books.google.com/books?id=5rlNAQAAIAAJ&pg=RA4-PA9 https://books.google.com/books?id=5rlNAQAAIAAJ&pg=RA4-PA28 War Time Order 1945 [1945-01-31] and the Order which revoke War Time Order 1945 [1945-10-16] Appendix to the Statutes of 9 George VI. and the Year 1945. p 160, 247-248 https://books.google.com/books?id=5rlNAQAAIAAJ&pg=RA6-PA160 https://books.google.com/books?id=5rlNAQAAIAAJ&pg=RA6-PA247 ---------- Guyana British Guiana/Guyana adopted GMT-4 on 1911-08-01 , GMT-3:75 on 1915-03-01, GMT-3 on 1975-08-01 and GMT-4 on 1992-03-29. The Official Gazette of British Guiana. (New Series.) Vol. XL. July to December, 1915, p 1547 https://books.google.com/books?id=5-5CAQAAMAAJ&pg=SA1-PA547 Interpretation and general clauses (Amendment) Act 1975 (Act No. 27 of 1975) [1975-07-31] https://parliament.gov.gy/documents/acts/10923-act_no._27_of_1975_-_interpr… Interpretation and general clauses (Amendment) Act 1992 (Act No. 6 of 1992) [1992-04-18] https://parliament.gov.gy/documents/acts/5885-6_of_1992_interpretation_and_… Circular No. 10/1992 dated 1992-03-30 https://dps.gov.gy/wp-content/uploads/2018/12/1992-03-20-Circular-010.pdf ---------- Nigeria On 1914-01-01, standard time of GMT+0:30 was adopted for the unified Nigeria. Colonial Reports - Annual. No. 878. Nigeria. Report for 1914. (April 1916), p 27 https://libsysdigi.library.illinois.edu/ilharvest/Africana/Books2011-05/306… Interpretation Ordinance, 1914 (No. 4 of 1914) [1914-10-01] https://books.google.com.hk/books?id=mm5CAQAAMAAJ&pg=PA172 In 1919, standard time was changed to GMT+1. Interpretation Ordinance (Cap 2) The Laws of Nigeria, Containing the Ordinances of Nigeria, in Force on the 1st Day of January, 1923, Vol.I p 5-18 https://books.google.com.hk/books?id=BOMrAQAAMAAJ&pg=PA16 For the standard time used in Lagos before 1914, the Interpretation Ordinance of Southern Nigeria, 1908 revised edition defined “Time” to mean Greenwich time. That section was originated from Proclamation No. 11 of 1905 of the Protectorate of Southern Nigeria. But the Colony of Lagos was not merged into Southern Nigeria until May 1906. Moreover, the 1906 Report of Southern Nigeria says “all laws in force in any part of the territory prior to the commencement of the new system of government shall continue in force until further provision is made by the local legislature.” Lloyd's Register of Shipping (1915) says “Hitherto the time observed in Lagos was the local mean time. On 1st January, 1914, standard time for the whole of Nigeria was introduced … Lagos time has been advanced about 16 minutes accordingly.” Therefore it is possible that no standard time was adopted in Lagos until 1914. Interpretation Ordinance (Cap 2) Laws of the Colony of Southern Nigeria : being the schedule to the Statute laws revision ordinance, 1908, Vol.I, p 3-11 https://babel.hathitrust.org/cgi/pt?id=mdp.35112105426037&view=1up&seq=61 Colonial Reports - Annual. No. 554. Southern Nigeria. Report for 1906. (February 1908), p 43 https://libsysdigi.library.illinois.edu/ilharvest/Africana/Books2011-05/306… Lloyd's Register of Shipping. 1915-16. Vol. II.-Appendix. ,p 655 https://books.google.com/books?id=b_ROAQAAIAAJ&pg=PA655 ---------- Seychelles Standard Time was adopted on 1907-01-01. Standard Time Ordinance (Chapter 237) The Laws of Seychelles in Force on the 31st December, 1971, Vol. 6, p 571 https://books.google.com/books?id=efE-AQAAIAAJ&pg=PA571 ---------- Turks and Caicos Islands Standard Time Declaration Order 2015 (L.N. 15/2015) http://online.fliphtml5.com/fizd/czin/#p=2 Standard Time Declaration Order 2017 (L.N. 31/2017) http://online.fliphtml5.com/fizd/dmcu/#p=2 ---------- Vanuatu DST was observed in 1973/1974. The spring forward in 1984 should be on Sep 23. Joint Daylight Saving Regulation No 59 of 1973 New Hebrides Condominium Gazette No 336. December 1973 http://www.paclii.org/vu/other/VUNHGovGaz//1973/11.pdf#page=15 Joint Daylight Saving (Repeal) Regulation No 10 of 1974 New Hebrides Condominium Gazette No 336. March 1974 http://www.paclii.org/vu/other/VUNHGovGaz//1974/3.pdf#page=11 Summer Time Act No. 35 of 1982 [1983-09-01] http://www.paclii.org/vu/other/VUGovGaz/1982/32.pdf#page=48 Summer Time Act (Cap 157) Laws of the Republic of Vanuatu Revised Edition 1988 http://www.paclii.org/cgi-bin/sinodisp/vu/legis/consol_act1988/sta147/sta14… Summer Time (Amendment) Act No. 6 of 1991 [1991-11-11] http://www.paclii.org/vu/legis/num_act/sta1991227/ Summer Time (Repeal) Act No. 4 of 1993 [1993-05-03] http://www.paclii.org/vu/other/VUGovGaz/1993/15.pdf#page=59

4 11

[PATCH] Fixups for today’s patches
by Paul Eggert Feb. 18, 2021

Feb. 18, 2021

* NEWS: Mention the recent all-year DST change. * zic.c (stringzone): Fix sign bug. --- NEWS | 7 +++++++ zic.c | 2 +- 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/NEWS b/NEWS index af5dd9f..f3c94ed 100644 --- a/NEWS +++ b/NEWS @@ -20,6 +20,13 @@ Unreleased, experimental changes last time transition disagreed with the TZ string, contrary to Internet RFC 8563 section 3.3. + zic now generates a POSIX-conforming TZ string for TZif files + where all-year DST is predicted for the indefinite future. + For example, for all-year Eastern Daylight Time, zic now generates + "XXX3EDT4,0/0,J365/23" where it previously generated + "EST5EDT,0/0,J365/25" or "". (Thanks to Michael Deckers for + noting the possibility of POSIX conformance.) + Release 2021a - 2021-01-24 10:54:57 -0800 diff --git a/zic.c b/zic.c index 4893a32..b91cb6c 100644 --- a/zic.c +++ b/zic.c @@ -2519,7 +2519,7 @@ stringzone(char *result, struct zone const *zpfirst, ptrdiff_t zonecount) Fake a timezone with negative DST. */ stdzp = &zstr[0]; dstzp = &zstr[1]; - zstr[0].z_stdoff = zp->z_stdoff - 2 * save; + zstr[0].z_stdoff = zp->z_stdoff + 2 * save; zstr[0].z_format = "XXX"; /* Any 3 letters will do. */ zstr[0].z_format_specifier = 0; zstr[1].z_stdoff = zstr[0].z_stdoff; -- 2.27.0

1 0

[PATCH 1/4] Avoid use of local statics
by Paul Eggert Feb. 18, 2021

Feb. 18, 2021

* zic.c (inrule, inzsub, writezone): Avoid use of static when auto will do. This can help GCC catch uninitialized-variable errors. --- zic.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/zic.c b/zic.c index 51977b2..8653fb0 100644 --- a/zic.c +++ b/zic.c @@ -1341,7 +1341,7 @@ getsave(char *field, bool *isdst) static void inrule(char **fields, int nfields) { - static struct rule r; + struct rule r; if (nfields != RULE_FIELDS) { error(_("wrong number of fields on Rule line")); @@ -1418,7 +1418,7 @@ inzsub(char **fields, int nfields, bool iscont) { register char * cp; char * cp1; - static struct zone z; + struct zone z; register int i_stdoff, i_rule, i_format; register int i_untilyear, i_untilmonth; register int i_untilday, i_untiltime; @@ -1868,8 +1868,6 @@ writezone(const char *const name, const char *const string, char version, register FILE * fp; register ptrdiff_t i, j; register int pass; - static const struct tzhead tzh0; - static struct tzhead tzh; bool dir_checked = false; zic_t one = 1; zic_t y2038_boundary = one << 31; @@ -2000,6 +1998,7 @@ writezone(const char *const name, const char *const string, char version, for (pass = 1; pass <= 2; ++pass) { register ptrdiff_t thistimei, thistimecnt, thistimelim; register int thisleapi, thisleapcnt, thisleaplim; + struct tzhead tzh; int currenttype, thisdefaulttype; bool locut, hicut; zic_t lo; @@ -2168,7 +2167,7 @@ writezone(const char *const name, const char *const string, char version, thistimelim = thistimei; } #define DO(field) fwrite(tzh.field, sizeof tzh.field, 1, fp) - tzh = tzh0; + memset(&tzh, 0, sizeof tzh); memcpy(tzh.tzh_magic, TZ_MAGIC, sizeof tzh.tzh_magic); tzh.tzh_version[0] = version; convert(utcnt, tzh.tzh_ttisutcnt); -- 2.27.0

1 3

State of Palestine Time zone - need to update
by heba hamad Feb. 14, 2021

Feb. 14, 2021

Dear Sir/Madam, In reference to your recent response at 25 Oct,2020 regarding State of Palestine's request to correct the 'Time Zone Database', we would like to confirm that 'TZ' related to the State of Palestine is as follows: PS State of Palestine Asia/East Jerusalem UTC +2:00 Also, we would like to inform you that the State of Palestine comprises of the West Bank, including East Jerusalem and Gaza, all of which are integral parts of the State of Palestine, thus, as one territorial unit all areas belong to the same time zone. We would like to also emphasize that East Jerusalem is an integral part of the West Bank as stipulated under international law and by numerous international resolutions. Therefore, we urge you to adopt the official name and correct your data base. Best Regards, Heba Hamad 11 Feb,2021

22 35

[PATCH 1/2] Remove unused C macros
by Paul Eggert Feb. 13, 2021

Feb. 13, 2021

Also, pacify ‘gcc -Wunused-macros’, which was employed to find these unused macros. * localtime.c (NETBSD_INSPIRED_EXTERN): Define only if STD_INSPIRED is defined. * private.h (sys_time) [!TZ_TIME_T]: Inspect LOCALTIME_IMPLEMENTATION. * zdump.c (MAX_STRING_LENGTH): * zic.c (PRIdZIC): Remove these macros, which are no longer used. --- localtime.c | 16 ++++++++-------- private.h | 7 ++++--- zdump.c | 4 ---- zic.c | 1 - 4 files changed, 12 insertions(+), 16 deletions(-) diff --git a/localtime.c b/localtime.c index d3e406b..82a3b70 100644 --- a/localtime.c +++ b/localtime.c @@ -28,14 +28,6 @@ static int lock(void) { return 0; } static void unlock(void) { } #endif -/* NETBSD_INSPIRED_EXTERN functions are exported to callers if - NETBSD_INSPIRED is defined, and are private otherwise. */ -#if NETBSD_INSPIRED -# define NETBSD_INSPIRED_EXTERN -#else -# define NETBSD_INSPIRED_EXTERN static -#endif - #ifndef TZ_ABBR_MAX_LEN #define TZ_ABBR_MAX_LEN 16 #endif /* !defined TZ_ABBR_MAX_LEN */ @@ -2247,6 +2239,14 @@ leapcorr(struct state const *sp, time_t t) #ifdef STD_INSPIRED +/* NETBSD_INSPIRED_EXTERN functions are exported to callers if + NETBSD_INSPIRED is defined, and are private otherwise. */ +# if NETBSD_INSPIRED +# define NETBSD_INSPIRED_EXTERN +# else +# define NETBSD_INSPIRED_EXTERN static +# endif + /* ** IEEE Std 1003.1 (POSIX) says that 536457599 ** shall correspond to "Wed Dec 31 23:59:59 UTC 1986", which diff --git a/private.h b/private.h index 9efcd78..a919e4f 100644 --- a/private.h +++ b/private.h @@ -418,10 +418,11 @@ typedef unsigned long uintmax_t; # define TZ_TIME_T 0 #endif -#if TZ_TIME_T -# ifdef LOCALTIME_IMPLEMENTATION +#if defined LOCALTIME_IMPLEMENTATION && TZ_TIME_T static time_t sys_time(time_t *x) { return time(x); } -# endif +#endif + +#if TZ_TIME_T typedef time_tz tz_time_t; diff --git a/zdump.c b/zdump.c index b532fe3..57e363e 100644 --- a/zdump.c +++ b/zdump.c @@ -42,10 +42,6 @@ #define ZDUMP_HI_YEAR 2500 #endif /* !defined ZDUMP_HI_YEAR */ -#ifndef MAX_STRING_LENGTH -#define MAX_STRING_LENGTH 1024 -#endif /* !defined MAX_STRING_LENGTH */ - #define SECSPERNYEAR (SECSPERDAY * DAYSPERNYEAR) #define SECSPERLYEAR (SECSPERNYEAR + SECSPERDAY) #define SECSPER400YEARS (SECSPERNYEAR * (intmax_t) (300 + 3) \ diff --git a/zic.c b/zic.c index b37cdc9..51977b2 100644 --- a/zic.c +++ b/zic.c @@ -21,7 +21,6 @@ typedef int_fast64_t zic_t; #define ZIC_MIN INT_FAST64_MIN #define ZIC_MAX INT_FAST64_MAX -#define PRIdZIC PRIdFAST64 #define SCNdZIC SCNdFAST64 #ifndef ZIC_MAX_ABBR_LEN_WO_WARN -- 2.27.0

1 1

[PATCH] Clarify GCC pacification in previous fix
by Paul Eggert Feb. 13, 2021

Feb. 13, 2021

* zic.c (outzone): Make it clear that prevktime is initialized only to pacify GCC. --- zic.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/zic.c b/zic.c index a864bc4..b37cdc9 100644 --- a/zic.c +++ b/zic.c @@ -2715,7 +2715,8 @@ outzone(const struct zone *zpfirst, ptrdiff_t zonecount) for (i = 0; i < zonecount; ++i) { struct rule *prevrp = NULL; - zic_t prevktime = min_time; + zic_t prevktime; + INITIALIZE(prevktime); /* ** A guess that may well be corrected later. */ -- 2.27.0

1 0

[PATCH] Fix bug in last time type of Asia/Gaza etc.
by Paul Eggert Feb. 13, 2021

Feb. 13, 2021

* zic.c (outzone): When avoiding bloat due to transitions deducible from the TZ string, don’t eliminate transitions that are before any lower-bound cutoff established by -r. All but the last of these transitions will be discarded later by limitrange, and the last transition is needed so that the dummy time type after the TZif file’s last transition agrees with the TZ string as required by Internet RFC 8536 section 3.3. This problem can be observed in tzdb 2021a if you run the command ‘zic -r@2145916800 asia’: the output file Asia/Gaza’s only time transition, which is at the specified time 2038-01-01 00:00:00 UTC, has an associated time type that specifies +03 with DST, rather than the correct +02 without DST. --- NEWS | 6 ++++++ zic.c | 4 +++- 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/NEWS b/NEWS index b9d5777..a22e8ff 100644 --- a/NEWS +++ b/NEWS @@ -6,6 +6,12 @@ Unreleased, experimental changes Portugal observed DST in 1950. (Thanks to Alois Treindl.) + Changes to code + + Fix bug in zic -r; in some cases, the dummy time type after the + last time transition disagreed with the TZ string, contrary to + Internet RFC 8563 section 3.3. + Release 2021a - 2021-01-24 10:54:57 -0800 diff --git a/zic.c b/zic.c index a902b34..a864bc4 100644 --- a/zic.c +++ b/zic.c @@ -2715,6 +2715,7 @@ outzone(const struct zone *zpfirst, ptrdiff_t zonecount) for (i = 0; i < zonecount; ++i) { struct rule *prevrp = NULL; + zic_t prevktime = min_time; /* ** A guess that may well be corrected later. */ @@ -2851,7 +2852,7 @@ outzone(const struct zone *zpfirst, ptrdiff_t zonecount) rp->r_isdst, rp->r_save, false); offset = oadd(zp->z_stdoff, rp->r_save); if (!want_bloat() && !useuntil && !do_extend - && prevrp + && prevrp && lo_time <= prevktime && rp->r_hiyear == ZIC_MAX && prevrp->r_hiyear == ZIC_MAX) break; @@ -2865,6 +2866,7 @@ outzone(const struct zone *zpfirst, ptrdiff_t zonecount) lastatmax = timecnt; addtt(ktime, type); prevrp = rp; + prevktime = ktime; } } if (usestart) { -- 2.27.0

1 0

Replying to government officials
by John Hawkinson Feb. 12, 2021

Feb. 12, 2021

I wonder if we should have some kind of understanding of when it is appropriate for the general members of the list to respond to emails from government officials, versus deferring that task to the project coordinators. (I have thought this was a concern for some time, but multiple recent messages have amplified it in my mind.) -- jhawk(a)alum.mit.edu John Hawkinson

2 1

DST existed in Portugal 1950
by Alois Treindl Feb. 10, 2021

Feb. 10, 2021

TZ claims that in 1950 there was no DST in Portugal, zone Europe/Lisbon. While it was continuously applied between 1916 and 1965. The sources quoted in file europe are poor, it is only Whitman versus Shanks/Pottenger. I have research online law collection for Portugal at https://dre.pt/ and searched for 'hora de verão'. The results are sparse, but what I find on legal texts does not indicate any interruption of the 1916 law for the year 1950. This Wikipedia page shows a list https://pt.wikipedia.org/wiki/Lista_de_per%C3%ADodos_em_que_vigorou_o_hor%C… and mentions the Decreto-Lei nº 37048 for 1949-1965. That law says 'first Sunday in April to first Sunday in October. No exception for 1950. Here is a list from the Lisbon astronomical observatory, which says the same thing. http://oal.ul.pt/documentos/2018/01/hl1911a2018.pdf/ I suggest that the rules for Europe/Lisbon should be updated accordingly.

3 6

Request for change to the tz database
by Nadia Kobyliak Feb. 9, 2021

Feb. 9, 2021

Dear Arthur and Paul, I would like to draw your attention to one of the names of the UTC+2 time zone: Europe/Kiev . And suggest changing it to Europe/Kyiv. This name is derived from the name of Ukrainian capital “Київ”. Its official transliteration(1) is “Kyiv” and not “Kiev”. The transliteration “Kyiv” was approved by the Tenth United Nations Conference on Standardization of Geographical Names(2). This is a warranty of the campaign KyivNotKiev launched by Ministry of Foreign Affairs of Ukraine(3). This is a very important issue for Ukrainians in establishing our national identity in times of an armed conflict and hybrid war by propaganda and misinformation. “To appreciate the significance of the “Kyiv vs. Kiev” debate, one must first step back and view it in terms of the deep-rooted national identity crisis caused by centuries of Tsarist and Soviet russification.” (4) It would make a huge difference if you considered this suggestion! And please let me know, if there is something else I should do to substantiate this proposal. References: 1. According to the rules approved by Governmental decree from 27 January 2010 №55: https://www.kmu.gov.ua/npas/243262567 This particular example could be found there. 2. Resolution X/9. Romanization of Ukrainian geographical names could be found in the Report of the Conference: https://unstats.un.org/unsd/geoinfo/UNGEGN/docs/10th-uncsgn-docs/econf/E_CO… Document approved by the resolution: “Romanization System in Ukraine” https://unstats.un.org/unsd/geoinfo/UNGEGN/docs/10th-uncsgn-docs/econf/E_CO… 1. It may also be interesting for you to find out more about the movement on Wikipedia page: https://en.wikipedia.org/wiki/KyivNotKiev 2. An article on this matter in English could be found on Atlantic Council web-site: https://www.atlanticcouncil.org/blogs/ukrainealert/kyiv-not-kiev-why-spelli… Kind regards, Nadia Kobyliak [Изображение выглядит как знак Автоматически созданное описание] Nadia Kobyliak Senior project manager SOEs and Corporate Governance reform ________________________________ cell.: +38 097 642 35 29 12/2 Hrushevskoho str. me.gov.ua<http://www.me.gov.ua/> kobyliak.nadia(a)gmail.com<mailto:kobyliak.nadia@gmail.com> 01008 Kyiv Twitter<http://twitter.com/mineconomdev> NKobyliak(a)me.gov.ua<mailto:NKobyliak@me.gov.ua> Ukraine Facebook<http://facebook.com/mineconomdev>

17 33

Re: [tz] Request for change to the tz database
by Hal Murray Feb. 8, 2021

Feb. 8, 2021

tim(a)timtimeonline.com said: > Yes, and as the proposal moves forward, we could provide the IANA moderators > with updated boilerplate as necessary. I'd expect the boilerplate to be a pointer to a FAQ entry. I think a note on the page about the list would be helpful and friendly. https://mm.icann.org/mailman/listinfo/tz -- These are my opinions. I hate spam.

1 0

Re: [tz] South Sudan to change time zone?
by Paul Eggert Feb. 8, 2021

Feb. 8, 2021

On 2/7/21 10:16 AM, Hayo Schmidt wrote: > you created a new > edge case by defining 00:00 of a day and then shifting back the date by > one day (and 23 hours forward). This sort of edge case is not new to Africa/Juba. For example, it occurred on 2020-03-22 at 00:00 in America/Asuncion, and is scheduled to occur again on 2021-03-28 at 00:00 at the same location. > A datum could have been created when reaching the first time exactly > 00:00:00.000. This is modeled in tzdb by the clock ticking 23:59:59.999999999 (as many 9s as you like) followed by 23:00:00.000000000 (the same number of 0s). That is, all the timestamps from 23:00:00.000000000 through 23:59:59.999999999 occur twice, but 00:00:00.000000000 the next day occurs only once. In this sense, there's no difference between a transition at 24:00, and one at 02:00. You can seem more details by running the command "zdump -v Africa/Juba".

1 0

Fwd: Timezone issue
by Yaroslav Bilko Feb. 5, 2021

Feb. 5, 2021

Hi, hope you're doing great! I've noticed some libraries use quite an irrelevant name for the capital of Ukraine: "Timezone: Europe/Kiev". And it comes from TZDB. C'mon lads, everybody in the whole civilized world knows that Kiev is wrong but you use it for some reason (only bloody russians talk like this and without any permission from our country (ukrainians), they push that name just because they want conquer our country and erase ukrainnian culture and you help them to achieve this goal!), the only one official name of this city is Kyiv! Yes, many people say "Kiev" even in Ukraine, but their number is getting down very fast. People start to figure it out. And the most important thing - they use it in non-public conversations. In all public places all information must be written Ukrainian (sometimes dubbing with English and that's it). It is protected by the law ( https://zakon.rada.gov.ua/laws/show/2704-19#Text) All Latin transliteration of our cities written in this law: https://zakon.rada.gov.ua/laws/show/z0957-14#Text [image: image.png] No Kiev as you can see. Only Kyiv. Soviet name "Kiev" disappeared 30 years ago. There's no need to exhume this corpse from the grave. Ukraine started a official campaign ( https://mfa.gov.ua/news/67521-correctua-mzs-ukrajini-zvertajetysya-do-svitu… ) some time ago to show other countries and mass media that Kyiv is only one true name. Even Google accepted it and maps show the correct name. That hit me and my friends and colleagues right in a heart! We hope you find some time to fix this little typo, but this is little only for first look! Hope for your care.

4 5

Names of Areas
by Andrea Boyle Feb. 3, 2021

Feb. 3, 2021

Hello, I am writing to you to request a small but significant change to the area options that are currently listed in the database. Recently I used software that uses the database for the area and location information and I was initially very puzzled when trying to find Canada. America was there but no other country in the North or South American continents. After a little research I realized that the database has America represent both. The trouble I have is that America is also a well used name for the United States and represents the name of a country rather than the two continents. I will be honest that it definitely rubbed the wrong way to have to choose America to find a city in Canada and I would also bet that outside of the US it would not be intuitive to choose America when looking for other countries within North and South America. To that end, I think it would be a great improvement to update the area to "Americas" or "the Americas" so that it was much clearer that it represents more than just one country. Thank you -- Andrea Boyle

3 2

(Near) UNIX epoch in the wild
by Arthur David Olson Feb. 1, 2021

Feb. 1, 2021

Today at the Paul S. Sarbanes Silver Spring (Maryland) Transit Center. https://twitter.com/dashdashado/status/1356321947005628418 @dashdashado

2 1

Re: [tz] CCTF survey on Time and Frequency Metrology - Telecom
by Hal Murray Feb. 1, 2021

Feb. 1, 2021

Brian.Inglis(a)SystematicSw.ab.ca said: > My reply suggested they divorce the time scales to allow business systems to > use approximate POSIX like common law legal solar time as they ignore leap > seconds one way or another; allow scientific systems to use atomic time and > ignore legal solar time; and just ignore the ITU as broadcast radio time > signals (DCF/JJY/MSF/WWV/B/H) are unavailable most places, may need > equipment no longer commonly available, and too low accuracy for modern > needs; other commonly available sources are delayed to uselessness by > digital processing. Could you please say more about the "delayed to uselessness"? -- These are my opinions. I hate spam.

9 12

How can I add a new time zone: 'America/Joao Pessoa'?
by Fabrício Cabral Jan. 31, 2021

Jan. 31, 2021

Hi all! How can I add a new time zone? I'd like add 'America/Joao Pessoa'. Joao Pessoa[1] is the capital of Paraiba state, in Brazil. [1] https://en.wikipedia.org/wiki/Jo%C3%A3o_Pessoa,_Para%C3%ADba Best Regards, -- --fx

4 5

Re: [tz] South Sudan to change time zone?
by Chris Hill Jan. 30, 2021

Jan. 30, 2021

On Tue, 19 Jan 2021 at 19:04, Paul Eggert <eggert at cs.ucla.edu> wrote: > > This article says the change will take place at "midnight, February 1": > > > > http://southsudan24news.com/south-sudan-change-its-time-zone-in-february-1-… > > The version I now see says, "The change will occur at 00:00 (midnight) > on February 1, when clocks will be set back 1 hour to 23:00 (11:00 pm)." > This is a bit confusing, because if the change is indeed at 00:00 on > February 1, the quoted text should have ended in "23:00 (11:00 pm) the > previous day" to avoid confusion. > > Perhaps the official announcement (if one is made and someone can > .provide it) will be less confusing. The official announcement (accessible without a Facebook account at https://www.facebook.com/www.motps.goss.org/videos/3770463003014509/ concurs with this: 08:23 'So, from the 1st of February, we are going back to our actual Greenwich Mean Time. We are going back to our local zone, of 2 hours' difference between Greenwich Mean Time, which is the zero hour, and our local time which will be 2 hours, which will be Greenwich Mean Time plus 2 hours' 09:43 'On the 31st, and the morning starts, that morning will not start at 12, but when it is 12 midnight of the 31st of January, it will be 11 o'clock. And you continue like that... Our 12 midnight of the 31st January will repeat itself; it will be 11, and then you go again for 12 midnight. That is when it becomes according to our time now.' We're based in South Sudan so was keen to get this right. Hope it helps. Kind regards, -- Chris Hill IT Officer Mission Aviation Fellowship International T +44 (0)1233 895517 F +44 (0)1233 895570 Operations Centre, Henwood, Ashford, Kent TN24 8DH Mission Aviation Fellowship International. A company Limited by guarantee, registered in England & Wales Registered Charity Number: 1058226. Registered Company Number: 3144199. Registered Office: Operations Centre, Henwood, Ashford, Kent TN24 8DH https://www.mafint.org/ "NOTICE: Any opinions expressed in this message are those of the individual and not necessarily the company. This email (and any attachment) is confidential and may be legally privileged. If you are not the intended recipient, you must not use, disclose, forward, copy, print or retain this email. If you have received this email in error, please notify the sender before deleting it from your computer system. Email transmissions may not be completely secure, may be intercepted or corrupted and may arrive late or not at all. The sender accepts no liability for viruses and we recommend that you scan all emails for viruses before opening them. For more information please contact dataprotection(a)mafint.org<mailto:dataprotection@mafint.org>. To view our Privacy Notice click Here<https://www.mafint.org/privacy-notice.html>. To view our Terms and Conditions click Here<https://www.mafint.org/terms-and-conditions.html>."

6 8

Old London Meridian
by Tim Jacobs Jan. 29, 2021

Jan. 29, 2021

Lines 106 to 125 of tz/europe mention the old meridian before Greenwich. The estimated position is 51° 28' 30" N, 0° 18' 45" W, which would put it north of the river in the grounds of Syon House. The meridian markers are in fact all south of the river, and can be seen by walking the Thames Path on the south bank. The meridian passes through the King's Observatory, with obelisks to the north and south. The map on the information boards shows a total of four obelisks marking the meridian. The website for the King's Observatory quotes the following: "This task was later transferred to Greenwich Observatory but not before three obelisks had been erected in the Old Deer Park in June 1778 to assist in the exact alignment of instruments in the Observatory. The northern one and the western one of the southern pair are on the true north-south meridian line passing through the west room of the Observatory which housed a tracking telescope. The southern obelisk is due south from the east room that at the time housed its great mural quadrant." The position of the King's Observatory is given on Wikipedia as 51°28′08″N 0°18′53″W, which would give STDOFF = -0:01:15.53 I don't suppose anyone is particularly concerned about this, but I thought I would point it out.

1 0