Branch: Tag:

2016-10-23

2016-10-23 20:45:55 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Updated to tzdata2016h.

1:   News for the tz database    + Release 2016h - 2016-10-19 23:17:57 -0700 +  +  Changes to future time stamps +  +  Asia/Gaza and Asia/Hebron end DST on 2016-10-29 at 01:00, not +  2016-10-21 at 00:00. (Thanks to Sharef Mustafa.) Predict that +  future fall transitions will be on the last Saturday of October +  at 01:00, which is consistent with predicted spring transitions +  on the last Saturday of March. (Thanks to Tim Parenti.) +  +  Changes to past time stamps +  +  In Turkey, transitions in 1986-1990 were at 01:00 standard time +  not at 02:00, and the spring 1994 transition was on March 20, not +  March 27. (Thanks to Kıvanç Yazan.) +  +  Changes to past and future time zone abbreviations +  +  Asia/Colombo now uses numeric time zone abbreviations like "+0530" +  instead of alphabetic ones like "IST" and "LKT". Various +  English-language sources use "IST", "LKT" and "SLST", with no +  working consensus. (Usage of "SLST" mentioned by Sadika +  Sumanapala.) +  +  Changes to code +  +  zic no longer mishandles relativizing file names when creating +  symbolic links like /etc/localtime, when these symbolic links +  are outside the usual directory hierarchy. This fixes a bug +  introduced in 2016g. (Problem reported by Andreas Stieger.) +  +  Changes to build procedure +  +  New rules 'traditional_tarballs' and 'traditional_signatures' for +  building just the traditional-format distribution. (Requested by +  Deborah Goldsmith.) +  +  The file 'version' is now put into the tzdata tarball too. +  (Requested by Howard Hinnant.) +  +  Changes to documentation and commentary +  +  The 'Theory' file now has a section on interface stability. +  (Requested by Paul Koning.) It also mentions features like +  tm_zone and localtime_rz that have long been supported by the +  reference code. +  +  tz-link.htm has improved coverage of time zone boundaries suitable +  for geolocation. (Thanks to heads-ups from Evan Siroky and Matt +  Johnson.) +  +  The US commentary now mentions Allen and the "day of two noons". +  +  The Fiji commentary mentions the government's 2016-10-03 press +  release. (Thanks to Raymond Kumar.) +  +  + Release 2016g - 2016-09-13 08:56:38 -0700 +  +  Changes to future time stamps +  +  Turkey switched from EET/EEST (+02/+03) to permanent +03, +  effective 2016-09-07. (Thanks to Burak AYDIN.) Use "+03" rather +  than an invented abbreviation for the new time. +  +  New leap second 2016-12-31 23:59:60 UTC as per IERS Bulletin C 52. +  (Thanks to Tim Parenti.) +  +  Changes to past time stamps +  +  For America/Los_Angeles, spring-forward transition times have been +  corrected from 02:00 to 02:01 in 1948, and from 02:00 to 01:00 in +  1950-1966. +  +  For zones using Soviet time on 1919-07-01, transitions to UT-based +  time were at 00:00 UT, not at 02:00 local time. The affected +  zones are Europe/Kirov, Europe/Moscow, Europe/Samara, and +  Europe/Ulyanovsk. (Thanks to Alexander Belopolsky.) +  +  Changes to past and future time zone abbreviations +  +  The Factory zone now uses the time zone abbreviation -00 instead +  of a long English-language string, as -00 is now the normal way to +  represent an undefined time zone. +  +  Several zones in Antarctica and the former Soviet Union, along +  with zones intended for ships at sea that cannot use POSIX TZ +  strings, now use numeric time zone abbreviations instead of +  invented or obsolete alphanumeric abbreviations. The affected +  zones are Antarctica/Casey, Antarctica/Davis, +  Antarctica/DumontDUrville, Antarctica/Mawson, Antarctica/Rothera, +  Antarctica/Syowa, Antarctica/Troll, Antarctica/Vostok, +  Asia/Anadyr, Asia/Ashgabat, Asia/Baku, Asia/Bishkek, Asia/Chita, +  Asia/Dushanbe, Asia/Irkutsk, Asia/Kamchatka, Asia/Khandyga, +  Asia/Krasnoyarsk, Asia/Magadan, Asia/Omsk, Asia/Sakhalin, +  Asia/Samarkand, Asia/Srednekolymsk, Asia/Tashkent, Asia/Tbilisi, +  Asia/Ust-Nera, Asia/Vladivostok, Asia/Yakutsk, Asia/Yekaterinburg, +  Asia/Yerevan, Etc/GMT-14, Etc/GMT-13, Etc/GMT-12, Etc/GMT-11, +  Etc/GMT-10, Etc/GMT-9, Etc/GMT-8, Etc/GMT-7, Etc/GMT-6, Etc/GMT-5, +  Etc/GMT-4, Etc/GMT-3, Etc/GMT-2, Etc/GMT-1, Etc/GMT+1, Etc/GMT+2, +  Etc/GMT+3, Etc/GMT+4, Etc/GMT+5, Etc/GMT+6, Etc/GMT+7, Etc/GMT+8, +  Etc/GMT+9, Etc/GMT+10, Etc/GMT+11, Etc/GMT+12, Europe/Kaliningrad, +  Europe/Minsk, Europe/Samara, Europe/Volgograd, and +  Indian/Kerguelen. For Europe/Moscow the invented abbreviation MSM +  was replaced by +05, whereas MSK and MSD were kept as they are not +  our invention and are widely used. +  +  Changes to zone names +  +  Rename Asia/Rangoon to Asia/Yangon, with a backward compatibility link. +  (Thanks to David Massoud.) +  +  Changes to code +  +  zic no longer generates binary files containing POSIX TZ-like +  strings that disagree with the local time type after the last +  explicit transition in the data. This fixes a bug with +  Africa/Casablanca and Africa/El_Aaiun in some year-2037 time +  stamps on the reference platform. (Thanks to Alexander Belopolsky +  for reporting the bug and suggesting a way forward.) +  +  If the installed localtime and/or posixrules files are symbolic +  links, zic now keeps them symbolic links when updating them, for +  compatibility with platforms like OpenSUSE where other programs +  configure these files as symlinks. +  +  zic now avoids hard linking to symbolic links, avoids some +  unnecessary mkdir and stat system calls, and uses shorter file +  names internally. +  +  zdump has a new -i option to generate transitions in a +  more-compact but still human-readable format. This option is +  experimental, and the output format may change in future versions. +  (Thanks to Jon Skeet for suggesting that an option was needed, +  and thanks to Tim Parenti and Chris Rovick for further comments.) +  +  Changes to build procedure +  +  An experimental distribution format is available, in addition +  to the traditional format which will continue to be distributed. +  The new format is a tarball tzdb-VERSION.tar.lz with signature +  file tzdb-VERSION.tar.lz.asc. It unpacks to a top-level directory +  tzdb-VERSION containing the code and data of the traditional +  two-tarball format, along with extra data that may be useful. +  (Thanks to Antonio Diaz Diaz, Oscar van Vlijmen, and many others +  for comments about the experimental format.) +  +  The release version number is now more accurate in the usual case +  where releases are built from a Git repository. For example, if +  23 commits and some working-file changes have been made since +  release 2016g, the version number is now something like +  '2016g-23-g50556e3-dirty' instead of the misleading '2016g'. +  Official releases uses the same version number format as before, +  e.g., '2016g'. To support the more-accurate version number, its +  specification has moved from a line in the Makefile to a new +  source file 'version'. +  +  The experimental distribution contains a file to2050.tzs that +  contains what should be the output of 'zdump -i -c 2050' on +  primary zones. If this file is available, 'make check' now checks +  that zdump generates this output. +  +  'make check_web' now works on Fedora-like distributions. +  +  Changes to documentation and commentary +  +  tzfile.5 now documents the new restriction on POSIX TZ-like +  strings that is now implemented by zic. +  +  Comments now cite URLs for some 1917-1921 Russian DST decrees. +  (Thanks to Alexander Belopolsky.) +  +  tz-link.htm mentions JuliaTime (thanks to Curtis Vogt) and Time4J +  (thanks to Meno Hochschild) and ThreeTen-Extra, and its +  description of Java 8 has been brought up to date (thanks to +  Stephen Colebourne). Its description of local time on Mars has +  been updated to match current practice, and URLs have been updated +  and some obsolete ones removed. +  +    Release 2016f - 2016-07-05 16:26:51 +0200       Changes affecting future time stamps
393:       Changes affecting past time stamps    -  America/Whitehorse switched from UTC-9 to UTC-8 on 1967-05-28, not +  America/Whitehorse switched from UT -09 to -08 on 1967-05-28, not    1966-07-01. Also, Yukon's time zone history is documented better.    (Thanks to Brian Inglis and Dennis Ferguson.)   
427:       The 1918 transition was September 10, not September 1.    -  The UTC-4 time observed from 1932 to 1942 is now considered to be -  standard time, not year-round DST. +  The UT -04 time observed from 1932 to 1942 is now considered to +  be standard time, not year-round DST.    -  Santiago observed DST (UTC-3) from 1946-07-15 through 1946-08-31, -  then reverted to standard time, then switched its time zone to -  UTC-5 on 1947-04-01. +  Santiago observed DST (UT -03) from 1946-07-15 through +  1946-08-31, then reverted to standard time, then switched to -05 +  on 1947-04-01.       Assume transitions before 1968 were at 00:00, since we have no data    saying otherwise.
561:       Changes affecting current and future time stamps    -  Turks & Caicos' switch from US eastern time to UTC-4 year-round +  Turks & Caicos' switch from US eastern time to UT -04 year-round    did not occur on 2014-11-02 at 02:00. It's currently scheduled    for 2015-11-01 at 02:00. (Thanks to Chris Walton.)   
597:    years will use a similar pattern.       A new Zone Pacific/Bougainville, for the part of Papua New Guinea -  that plans to switch from UTC+10 to UTC+11 on 2014-12-28 at 02:00. +  that plans to switch from UT +10 to +11 on 2014-12-28 at 02:00.    (Thanks to Kiley Walbom for the heads-up.)       Changes affecting time zone abbreviations
607:    to its more-traditional value MSK on 2014-10-26 at 01:00.    (Thanks to Alexander Bokovoy for the heads-up about Belarus.)    -  The new abbreviation IDT stands for the pre-1976 use of UT+8 in -  Indochina, to distinguish it better from ICT (UT+7). +  The new abbreviation IDT stands for the pre-1976 use of UT +08 in +  Indochina, to distinguish it better from ICT (+07).       Changes affecting past time stamps   
686:    zdump -c no longer mishandles transitions near year boundaries.    (Thanks to Tim Parenti for reporting the problem.)    -  An access to uninitalized data has been fixed. +  An access to uninitialized data has been fixed.    (Thanks to Jörg Richter for reporting the problem.)       When THREAD_SAFE is defined, the code ports to the C11 memory model.
719:       Changes affecting future time stamps    -  Turks & Caicos is switching from US eastern time to UTC-4 year-round, -  modeled as a switch from EST/EDT to AST on 2014-11-02 at 02:00. +  Turks & Caicos is switching from US eastern time to UT -04 +  year-round, modeled as a switch on 2014-11-02 at 02:00.    [As noted in 2014j, this switch was later delayed.]       Changes affecting past time stamps
903:    This change does not affect UTC offsets, only time zone abbreviations.    (Thanks to Rich Tibbett and many others.)    -  Asia/Novokuznetsk shifts from NOVT to KRAT (remaining on UTC+7) +  Asia/Novokuznetsk shifts from NOVT to KRAT (remaining on UT +07)    effective 2014-10-26 at 02:00 local time.       The time zone abbreviation for Xinjiang Time (observed in Ürümqi)
932:    zones Asia/Harbin, Asia/Chongqing, and Asia/Kashgar have been    removed; backwards-compatibility links still work, albeit with    different behaviors for time stamps before May 1980. Asia/Urumqi's -  1980 transition to UTC+8 has been removed, so that it is now at -  UTC+6 and not UTC+8. (Thanks to Luther Ma and to Alois Treindl; +  1980 transition to UT +08 has been removed, so that it is now at +  +06 and not +08. (Thanks to Luther Ma and to Alois Treindl;    Treindl sent helpful translations of two papers by Guo Qingsheng.)       Some zones have been turned into links, when they differed from existing
963:    Europe/Helsinki's 1942 fall-back transition was 10-04 at 01:00, not    10-03 at 00:00. (Thanks to Konstantin Hyppönen.)    -  Pacific/Pago_Pago has been changed from UTC-11:30 to UTC-11 for the period -  from 1911 to 1950. +  Pacific/Pago_Pago has been changed from UT -11:30 to -11 for the +  period from 1911 to 1950.       Pacific/Chatham has been changed to New Zealand standard time plus    45 minutes for the period before 1957, reflecting a 1956 remark in
1291:       Changes affecting current and future time stamps:    -  Libya has switched its time zone back to UTC+2 without DST, -  instead of UTC+1 with DST. (Thanks to Even Scharning.) +  Libya has switched its UT offset back to +02 without DST, instead +  of +01 with DST. (Thanks to Even Scharning.)       Western Sahara (Africa/El_Aaiun) uses Morocco's DST rules.    (Thanks to Gwillim Law.)       Changes affecting future time stamps:    -  Acre and (we guess) western Amazonas will switch from UTC-4 to UTC-5 +  Acre and (we guess) western Amazonas will switch from UT -04 to -05    on 2013-11-10. This affects America/Rio_Branco and America/Eirunepe.    (Thanks to Steffen Thorsen.)   
1386:    Tocantins will very likely not observe DST starting this spring.    (Thanks to Steffen Thorsen.)    -  Jordan will likely stay at UTC+3 indefinitely, and will not fall +  Jordan will likely stay at UT +03 indefinitely, and will not fall    back this fall.       Palestine will fall back at 00:00, not 01:00. (Thanks to Steffen Thorsen.)
1443:    Benny Lin.) This affects Asia/Dili, Asia/Jakarta, Asia/Jayapura,    Asia/Makassar, and Asia/Pontianak.    -  Use ART (UTC-3, standard time), rather than WARST (also UTC-3, but +  Use ART (UT -03, standard time), rather than WARST (also -03, but    daylight saving time) for San Luis, Argentina since 2009.       Changes affecting Godthåb time stamps after 2037 if version mismatch
1900:      Release 2012e - 2012-08-02 20:44:55 -0700    -  * australasia (Pacific/Fakaofo): Tokelau is UTC+13, not UTC+14. +  * australasia (Pacific/Fakaofo): Tokelau is UT +13, not +14.    (Thanks to Steffen Thorsen.)       * Use a single version number for both code and data.
3519:   that talks about the changes in that release. The text has been   adapted and reformatted for the purposes of this file.    - Typically a release R consists of a pair of tarball files, + Traditionally a release R consists of a pair of tarball files,   tzcodeR.tar.gz and tzdataR.tar.gz. However, some releases (e.g.,   code2010a, data2012c) consist of just one or the other tarball, and a   few (e.g., code2012c-data2012d) have tarballs with mixed version - numbers. + numbers. Recent releases also come in an experimental format + consisting of a single tarball tzdb-R.tar.lz with extra data.      Release time stamps are taken from the release's commit (for newer, - git releases), from the newest file in the tarball (for older + Git-based releases), from the newest file in the tarball (for older   releases, where this info is available) or from the email announcing   the release (if all else fails; these are marked with a time zone of   -0000 and an "is missing!" comment).