4 <title>Theory and pragmatics of the tz code and data</title>
7 pre {margin-left: 2em; white-space: pre-wrap;}
12 <h1>Theory and pragmatics of the <code><abbr>tz</abbr></code> code and data</h1>
16 <li><a href="#scope">Scope of the <code><abbr>tz</abbr></code>
18 <li><a href="#naming">Timezone identifiers</a></li>
19 <li><a href="#abbreviations">Time zone abbreviations</a></li>
20 <li><a href="#accuracy">Accuracy of the <code><abbr>tz</abbr></code>
22 <li><a href="#functions">Time and date functions</a></li>
23 <li><a href="#stability">Interface stability</a></li>
24 <li><a href="#leapsec">Leap seconds</a></li>
25 <li><a href="#calendar">Calendrical issues</a></li>
26 <li><a href="#planets">Time and time zones on other planets</a></li>
31 <h2 id="scope">Scope of the <code><abbr>tz</abbr></code> database</h2>
34 href="https://www.iana.org/time-zones"><code><abbr>tz</abbr></code>
35 database</a> attempts to record the history and predicted future of
37 It organizes <a href="tz-link.html">time zone and daylight saving time
38 data</a> by partitioning the world into <a
39 href="https://en.wikipedia.org/wiki/List_of_tz_database_time_zones"><dfn>timezones</dfn></a>
40 whose clocks all agree about timestamps that occur after the <a
41 href="https://en.wikipedia.org/wiki/Unix_time">POSIX Epoch</a>
42 (1970-01-01 00:00:00 <a
43 href="https://en.wikipedia.org/wiki/Coordinated_Universal_Time"><abbr
44 title="Coordinated Universal Time">UTC</abbr></a>).
45 Although 1970 is a somewhat-arbitrary cutoff, there are significant
46 challenges to moving the cutoff earlier even by a decade or two, due
47 to the wide variety of local practices before computer timekeeping
49 Most timezones correspond to a notable location and the database
50 records all known clock transitions for that location;
51 some timezones correspond instead to a fixed <abbr>UTC</abbr> offset.
55 Each timezone typically corresponds to a geographical region that is
56 smaller than a traditional time zone, because clocks in a timezone
57 all agree after 1970 whereas a traditional time zone merely
58 specifies current standard time. For example, applications that deal
59 with current and future timestamps in the traditional North
60 American mountain time zone can choose from the timezones
61 <code>America/Denver</code> which observes US-style daylight saving
62 time (<abbr>DST</abbr>),
63 and <code>America/Phoenix</code> which does not observe <abbr>DST</abbr>.
64 Applications that also deal with past timestamps in the mountain time
65 zone can choose from over a dozen timezones, such as
66 <code>America/Boise</code>, <code>America/Edmonton</code>, and
67 <code>America/Hermosillo</code>, each of which currently uses mountain
68 time but differs from other timezones for some timestamps after 1970.
72 Clock transitions before 1970 are recorded for location-based timezones,
73 because most systems support timestamps before 1970 and could
74 misbehave if data entries were omitted for pre-1970 transitions.
75 However, the database is not designed for and does not suffice for
76 applications requiring accurate handling of all past times everywhere,
77 as it would take far too much effort and guesswork to record all
78 details of pre-1970 civil timekeeping.
79 Although some information outside the scope of the database is
80 collected in a file <code>backzone</code> that is distributed along
81 with the database proper, this file is less reliable and does not
82 necessarily follow database guidelines.
86 As described below, reference source code for using the
87 <code><abbr>tz</abbr></code> database is also available.
88 The <code><abbr>tz</abbr></code> code is upwards compatible with <a
89 href="https://en.wikipedia.org/wiki/POSIX">POSIX</a>, an international
91 href="https://en.wikipedia.org/wiki/Unix">UNIX</a>-like systems.
92 As of this writing, the current edition of POSIX is: <a
93 href="https://pubs.opengroup.org/onlinepubs/9699919799/"> The Open
94 Group Base Specifications Issue 7</a>, IEEE Std 1003.1-2017, 2018
96 Because the database's scope encompasses real-world changes to civil
97 timekeeping, its model for describing time is more complex than the
98 standard and daylight saving times supported by POSIX.
99 A <code><abbr>tz</abbr></code> timezone corresponds to a ruleset that can
100 have more than two changes per year, these changes need not merely
101 flip back and forth between two alternatives, and the rules themselves
103 Whether and when a timezone changes its clock,
104 and even the timezone's notional base offset from <abbr>UTC</abbr>,
106 It does not always make sense to talk about a timezone's
107 "base offset", which is not necessarily a single number.
113 <h2 id="naming">Timezone identifiers</h2>
115 Each timezone has a name that uniquely identifies the timezone.
116 Inexperienced users are not expected to select these names unaided.
117 Distributors should provide documentation and/or a simple selection
118 interface that explains each name via a map or via descriptive text like
119 "Czech Republic" instead of the timezone name "<code>Europe/Prague</code>".
120 If geolocation information is available, a selection interface can
121 locate the user on a timezone map or prioritize names that are
122 geographically close. For an example selection interface, see the
123 <code>tzselect</code> program in the <code><abbr>tz</abbr></code> code.
124 The <a href="https://cldr.unicode.org">Unicode Common Locale Data
125 Repository</a> contains data that may be useful for other selection
126 interfaces; it maps timezone names like <code>Europe/Prague</code> to
127 locale-dependent strings like "Prague", "Praha", "Прага", and "布拉格".
131 The naming conventions attempt to strike a balance
132 among the following goals:
137 Uniquely identify every timezone where clocks have agreed since 1970.
138 This is essential for the intended use: static clocks keeping local
142 Indicate to experts where the timezone's clocks typically are.
145 Be robust in the presence of political changes.
146 For example, names are typically not tied to countries, to avoid
147 incompatibilities when countries change their name (e.g.,
148 Swaziland→Eswatini) or when locations change countries (e.g., Hong
149 Kong from UK colony to China).
150 There is no requirement that every country or national
151 capital must have a timezone name.
154 Be portable to a wide variety of implementations.
157 Use a consistent naming conventions over the entire world.
162 Names normally have the form
163 <var>AREA</var><code>/</code><var>LOCATION</var>, where
164 <var>AREA</var> is a continent or ocean, and
165 <var>LOCATION</var> is a specific location within the area.
166 North and South America share the same area, '<code>America</code>'.
167 Typical names are '<code>Africa/Cairo</code>',
168 '<code>America/New_York</code>', and '<code>Pacific/Honolulu</code>'.
169 Some names are further qualified to help avoid confusion; for example,
170 '<code>America/Indiana/Petersburg</code>' distinguishes Petersburg,
171 Indiana from other Petersburgs in America.
175 Here are the general guidelines used for
176 choosing timezone names,
177 in decreasing order of importance:
182 Use only valid POSIX file name components (i.e., the parts of
183 names other than '<code>/</code>').
184 Do not use the file name components '<code>.</code>' and
186 Within a file name component, use only <a
187 href="https://en.wikipedia.org/wiki/ASCII">ASCII</a> letters,
188 '<code>.</code>', '<code>-</code>' and '<code>_</code>'.
189 Do not use digits, as that might create an ambiguity with <a
190 href="https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html#tag_08_03">POSIX
191 <code>TZ</code> strings</a>.
192 A file name component must not exceed 14 characters or start with
194 E.g., prefer <code>America/Noronha</code> to
195 <code>America/Fernando_de_Noronha</code>.
196 Exceptions: see the discussion of legacy names below.
199 A name must not be empty, or contain '<code>//</code>', or
200 start or end with '<code>/</code>'.
203 Do not use names that differ only in case.
204 Although the reference implementation is case-sensitive, some
205 other implementations are not, and they would mishandle names
206 differing only in case.
209 If one name <var>A</var> is an initial prefix of another
210 name <var>AB</var> (ignoring case), then <var>B</var> must not
211 start with '<code>/</code>', as a regular file cannot have the
212 same name as a directory in POSIX.
213 For example, <code>America/New_York</code> precludes
214 <code>America/New_York/Bronx</code>.
217 Uninhabited regions like the North Pole and Bouvet Island
218 do not need locations, since local time is not defined there.
221 If all the clocks in a timezone have agreed since 1970,
222 do not bother to include more than one timezone
223 even if some of the clocks disagreed before 1970.
224 Otherwise these tables would become annoyingly large.
227 If boundaries between regions are fluid, such as during a war or
228 insurrection, do not bother to create a new timezone merely
229 because of yet another boundary change. This helps prevent table
230 bloat and simplifies maintenance.
233 If a name is ambiguous, use a less ambiguous alternative;
234 e.g., many cities are named San José and Georgetown, so
235 prefer <code>America/Costa_Rica</code> to
236 <code>America/San_Jose</code> and <code>America/Guyana</code>
237 to <code>America/Georgetown</code>.
240 Keep locations compact.
241 Use cities or small islands, not countries or regions, so that any
242 future changes do not split individual locations into different
244 E.g., prefer <code>Europe/Paris</code> to <code>Europe/France</code>,
246 <a href="https://en.wikipedia.org/wiki/Time_in_France#History">France
247 has had multiple time zones</a>.
250 Use mainstream English spelling, e.g., prefer
251 <code>Europe/Rome</code> to <code>Europa/Roma</code>, and
252 prefer <code>Europe/Athens</code> to the Greek
253 <code>Ευρώπη/Αθήνα</code> or the Romanized
254 <code>Evrópi/Athína</code>.
255 The POSIX file name restrictions encourage this guideline.
258 Use the most populous among locations in a region,
259 e.g., prefer <code>Asia/Shanghai</code> to
260 <code>Asia/Beijing</code>.
261 Among locations with similar populations, pick the best-known
262 location, e.g., prefer <code>Europe/Rome</code> to
263 <code>Europe/Milan</code>.
266 Use the singular form, e.g., prefer <code>Atlantic/Canary</code> to
267 <code>Atlantic/Canaries</code>.
270 Omit common suffixes like '<code>_Islands</code>' and
271 '<code>_City</code>', unless that would lead to ambiguity.
272 E.g., prefer <code>America/Cayman</code> to
273 <code>America/Cayman_Islands</code> and
274 <code>America/Guatemala</code> to
275 <code>America/Guatemala_City</code>, but prefer
276 <code>America/Mexico_City</code> to
277 <code>America/Mexico</code>
278 because <a href="https://en.wikipedia.org/wiki/Time_in_Mexico">the
279 country of Mexico has several time zones</a>.
282 Use '<code>_</code>' to represent a space.
285 Omit '<code>.</code>' from abbreviations in names.
286 E.g., prefer <code>Atlantic/St_Helena</code> to
287 <code>Atlantic/St._Helena</code>.
290 Do not change established names if they only marginally violate
291 the above guidelines.
292 For example, do not change the existing name <code>Europe/Rome</code> to
293 <code>Europe/Milan</code> merely because Milan's population has grown
294 to be somewhat greater than Rome's.
297 If a name is changed, put its old spelling in the
298 '<code>backward</code>' file as a link to the new spelling.
299 This means old spellings will continue to work.
300 Ordinarily a name change should occur only in the rare case when
301 a location's consensus English-language spelling changes; for example,
302 in 2008 <code>Asia/Calcutta</code> was renamed to <code>Asia/Kolkata</code>
303 due to long-time widespread use of the new city name instead of the old.
308 Guidelines have evolved with time, and names following old versions of
309 these guidelines might not follow the current version. When guidelines
310 have changed, old names continue to be supported. Guideline changes
311 have included the following:
316 Older versions of this package used a different naming scheme.
317 See the file '<code>backward</code>' for most of these older names
318 (e.g., '<code>US/Eastern</code>' instead of '<code>America/New_York</code>').
319 The other old-fashioned names still supported are
320 '<code>WET</code>', '<code>CET</code>', '<code>MET</code>', and
321 '<code>EET</code>' (see the file '<code>europe</code>').
325 Older versions of this package defined legacy names that are
326 incompatible with the first guideline of location names, but which are
328 These legacy names are mostly defined in the file
329 '<code>etcetera</code>'.
330 Also, the file '<code>backward</code>' defines the legacy names
331 '<code>Etc/GMT0</code>', '<code>Etc/GMT-0</code>', '<code>Etc/GMT+0</code>',
332 '<code>GMT0</code>', '<code>GMT-0</code>' and '<code>GMT+0</code>',
333 and the file '<code>northamerica</code>' defines the legacy names
334 '<code>EST5EDT</code>', '<code>CST6CDT</code>',
335 '<code>MST7MDT</code>', and '<code>PST8PDT</code>'.
339 Older versions of these guidelines said that
340 there should typically be at least one name for each <a
341 href="https://en.wikipedia.org/wiki/ISO_3166-1"><abbr
342 title="International Organization for Standardization">ISO</abbr>
343 3166-1</a> officially assigned two-letter code for an inhabited
344 country or territory.
345 This old guideline has been dropped, as it was not needed to handle
346 timestamps correctly and it increased maintenance burden.
351 The file <code>zone1970.tab</code> lists geographical locations used
353 It is intended to be an exhaustive list of names for geographic
354 regions as described above; this is a subset of the timezones in the data.
355 Although a <code>zone1970.tab</code> location's
356 <a href="https://en.wikipedia.org/wiki/Longitude">longitude</a>
358 its <a href="https://en.wikipedia.org/wiki/Local_mean_time">local mean
359 time (<abbr>LMT</abbr>)</a> offset with one hour for every 15°
360 east longitude, this relationship is not exact.
361 The backward-compatibility file <code>zone.tab</code> is similar
362 but conforms to the older-version guidelines related to <abbr>ISO</abbr> 3166-1;
363 it lists only one country code per entry and unlike <code>zone1970.tab</code>
364 it can list names defined in <code>backward</code>.
368 The database defines each timezone name to be a zone, or a link to a zone.
369 The source file <code>backward</code> defines links for backward
370 compatibility; it does not define zones.
371 Although <code>backward</code> was originally designed to be optional,
372 nowadays distributions typically use it
373 and no great weight should be attached to whether a link
374 is defined in <code>backward</code> or in some other file.
375 The source file <code>etcetera</code> defines names that may be useful
376 on platforms that do not support POSIX-style <code>TZ</code> strings;
377 no other source file other than <code>backward</code>
378 contains links to its zones.
379 One of <code>etcetera</code>'s names is <code>Etc/UTC</code>,
380 used by functions like <code>gmtime</code> to obtain leap
381 second information on platforms that support leap seconds.
382 Another <code>etcetera</code> name, <code>GMT</code>,
383 is used by older code releases.
388 <h2 id="abbreviations">Time zone abbreviations</h2>
390 When this package is installed, it generates time zone abbreviations
391 like '<code>EST</code>' to be compatible with human tradition and POSIX.
392 Here are the general guidelines used for choosing time zone abbreviations,
393 in decreasing order of importance:
398 Use three to six characters that are ASCII alphanumerics or
399 '<code>+</code>' or '<code>-</code>'.
400 Previous editions of this database also used characters like
401 space and '<code>?</code>', but these characters have a
402 special meaning to the
403 <a href="https://en.wikipedia.org/wiki/Unix_shell">UNIX shell</a>
404 and cause commands like
405 '<code><a href="https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#set">set</a>
406 `<a href="https://pubs.opengroup.org/onlinepubs/9699919799/utilities/date.html">date</a>`</code>'
407 to have unexpected effects.
408 Previous editions of this guideline required upper-case letters, but the
409 Congressman who introduced
410 <a href="https://en.wikipedia.org/wiki/Chamorro_Time_Zone">Chamorro
411 Standard Time</a> preferred "ChST", so lower-case letters are now
413 Also, POSIX from 2001 on relaxed the rule to allow '<code>-</code>',
414 '<code>+</code>', and alphanumeric characters from the portable
415 character set in the current locale.
416 In practice ASCII alphanumerics and '<code>+</code>' and
417 '<code>-</code>' are safe in all locales.
420 In other words, in the C locale the POSIX extended regular
421 expression <code>[-+[:alnum:]]{3,6}</code> should match the
423 This guarantees that all abbreviations could have been specified by a
424 POSIX <code>TZ</code> string.
428 Use abbreviations that are in common use among English-speakers,
429 e.g., 'EST' for Eastern Standard Time in North America.
430 We assume that applications translate them to other languages
431 as part of the normal localization process; for example,
432 a French application might translate 'EST' to 'HNE'.
435 <small>These abbreviations (for standard/daylight/etc. time) are:
436 ACST/ACDT Australian Central,
437 AST/ADT/APT/AWT/ADDT Atlantic,
438 AEST/AEDT Australian Eastern,
439 AHST/AHDT Alaska-Hawaii,
441 AWST/AWDT Australian Western,
443 CAT/CAST Central Africa,
444 CET/CEST/CEMT Central European,
446 CST/CDT/CWT/CPT/CDDT Central [North America],
448 GMT/BST/IST/BDST Greenwich,
450 EST/EDT/EWT/EPT/EDDT Eastern [North America],
451 EET/EEST Eastern European,
453 HST/HDT/HWT/HPT Hawaii,
454 HKT/HKST/HKWT Hong Kong,
460 MET/MEST Middle European (a backward-compatibility alias for
463 MST/MDT/MWT/MPT/MDDT Mountain,
464 NST/NDT/NWT/NPT/NDDT Newfoundland,
465 NST/NDT/NWT/NPT Nome,
466 NZMT/NZST New Zealand through 1945,
467 NZST/NZDT New Zealand 1946–present,
469 PST/PDT/PWT/PPT/PDDT Pacific,
474 WAT/WAST West Africa,
475 WET/WEST/WEMT Western European,
476 WIB Waktu Indonesia Barat,
477 WIT Waktu Indonesia Timur,
478 WITA Waktu Indonesia Tengah,
479 YST/YDT/YWT/YPT/YDDT Yukon</small>.
484 For times taken from a city's longitude, use the
485 traditional <var>x</var>MT notation.
486 The only abbreviation like this in current use is '<abbr>GMT</abbr>'.
487 The others are for timestamps before 1960,
488 except that Monrovia Mean Time persisted until 1972.
489 Typically, numeric abbreviations (e.g., '<code>-</code>004430' for
490 MMT) would cause trouble here, as the numeric strings would exceed
491 the POSIX length limit.
495 <small>These abbreviations are:
496 AMT Asunción, Athens;
497 BMT Baghdad, Bangkok, Batavia, Bermuda, Bern, Bogotá, Bridgetown,
499 CMT Calamarca, Caracas, Chisinau, Colón, Córdoba;
505 HMT Havana, Helsinki, Horta, Howrah;
506 IMT Irkutsk, Istanbul;
508 KMT Kaunas, Kyiv, Kingston;
509 LMT Lima, Lisbon, local, Luanda;
510 MMT Macassar, Madras, Malé, Managua, Minsk, Monrovia, Montevideo,
513 PMT Paramaribo, Paris, Perm, Pontianak, Prague;
516 RMT Rangoon, Riga, Rome;
519 SMT Santiago, Simferopol, Singapore, Stanley;
527 <small>A few abbreviations also follow the pattern that
528 <abbr>GMT</abbr>/<abbr>BST</abbr> established for time in the UK.
530 BMT/BST for Bermuda 1890–1930,
531 CMT/BST for Calamarca Mean Time and Bolivian Summer Time
533 DMT/IST for Dublin/Dunsink Mean Time and Irish Summer Time
535 MMT/MST/MDST for Moscow 1880–1919, and
536 RMT/LST for Riga Mean Time and Latvian Summer time 1880–1926.
541 Use '<abbr>LMT</abbr>' for local mean time of locations before the
542 introduction of standard time; see "<a href="#scope">Scope of the
543 <code><abbr>tz</abbr></code> database</a>".
546 If there is no common English abbreviation, use numeric offsets like
547 <code>-</code>05 and <code>+</code>0530 that are generated
548 by <code>zic</code>'s <code>%z</code> notation.
551 Use current abbreviations for older timestamps to avoid confusion.
552 For example, in 1910 a common English abbreviation for time
553 in central Europe was 'MEZ' (short for both "Middle European
554 Zone" and for "Mitteleuropäische Zeit" in German).
555 Nowadays 'CET' ("Central European Time") is more common in
556 English, and the database uses 'CET' even for circa-1910
557 timestamps as this is less confusing for modern users and avoids
558 the need for determining when 'CET' supplanted 'MEZ' in common
562 Use a consistent style in a timezone's history.
563 For example, if a history tends to use numeric
564 abbreviations and a particular entry could go either way, use a
565 numeric abbreviation.
569 <a href="https://en.wikipedia.org/wiki/Universal_Time">Universal Time</a>
570 (<abbr>UT</abbr>) (with time zone abbreviation '<code>-</code>00') for
571 locations while uninhabited.
572 The leading '<code>-</code>' is a flag that the <abbr>UT</abbr> offset is in
573 some sense undefined; this notation is derived
574 from <a href="https://datatracker.ietf.org/doc/html/rfc3339">Internet
575 <abbr title="Request For Comments">RFC</abbr> 3339</a>.
580 Application writers should note that these abbreviations are ambiguous
581 in practice: e.g., 'CST' means one thing in China and something else
582 in North America, and 'IST' can refer to time in India, Ireland or
584 To avoid ambiguity, use numeric <abbr>UT</abbr> offsets like
585 '<code>-</code>0600' instead of time zone abbreviations like 'CST'.
590 <h2 id="accuracy">Accuracy of the <code><abbr>tz</abbr></code> database</h2>
592 The <code><abbr>tz</abbr></code> database is not authoritative, and it
594 Corrections are welcome and encouraged; see the file <code>CONTRIBUTING</code>.
595 Users requiring authoritative data should consult national standards
596 bodies and the references cited in the database's comments.
600 Errors in the <code><abbr>tz</abbr></code> database arise from many sources:
605 The <code><abbr>tz</abbr></code> database predicts future
606 timestamps, and current predictions
607 will be incorrect after future governments change the rules.
608 For example, if today someone schedules a meeting for 13:00 next
609 October 1, Casablanca time, and tomorrow Morocco changes its
610 daylight saving rules, software can mess up after the rule change
611 if it blithely relies on conversions made before the change.
614 The pre-1970 entries in this database cover only a tiny sliver of how
615 clocks actually behaved; the vast majority of the necessary
616 information was lost or never recorded.
617 Thousands more timezones would be needed if
618 the <code><abbr>tz</abbr></code> database's scope were extended to
619 cover even just the known or guessed history of standard time; for
620 example, the current single entry for France would need to split
621 into dozens of entries, perhaps hundreds.
622 And in most of the world even this approach would be misleading
623 due to widespread disagreement or indifference about what times
627 href="https://www.hup.harvard.edu/catalog.php?isbn=9780674286146">The
628 Global Transformation of Time, 1870–1950</a></cite>,
630 "Outside of Europe and North America there was no system of time
631 zones at all, often not even a stable landscape of mean times,
632 prior to the middle decades of the twentieth century".
633 See: Timothy Shenk, <a
634 href="https://www.dissentmagazine.org/blog/booked-a-global-history-of-time-vanessa-ogle">Booked:
635 A Global History of Time</a>. <cite>Dissent</cite> 2015-12-17.
638 Most of the pre-1970 data entries come from unreliable sources, often
639 astrology books that lack citations and whose compilers evidently
640 invented entries when the true facts were unknown, without
641 reporting which entries were known and which were invented.
642 These books often contradict each other or give implausible entries,
643 and on the rare occasions when they are checked they are
644 typically found to be incorrect.
647 For the UK the <code><abbr>tz</abbr></code> database relies on
648 years of first-class work done by
649 Joseph Myers and others; see
650 "<a href="https://www.polyomino.org.uk/british-time/">History of
651 legal time in Britain</a>".
652 Other countries are not done nearly as well.
655 Sometimes, different people in the same city maintain clocks
656 that differ significantly.
657 Historically, railway time was used by railroad companies (which
659 agree with each other), church-clock time was used for birth
661 More recently, competing political groups might disagree about
662 clock settings. Often this is merely common practice, but
663 sometimes it is set by law.
664 For example, from 1891 to 1911 the <abbr>UT</abbr> offset in France
665 was legally <abbr>UT</abbr> +00:09:21 outside train stations and
666 <abbr>UT</abbr> +00:04:21 inside. Other examples include
667 Chillicothe in 1920, Palm Springs in 1946/7, and Jerusalem and
671 Although a named location in the <code><abbr>tz</abbr></code>
672 database stands for the containing region, its pre-1970 data
673 entries are often accurate for only a small subset of that region.
674 For example, <code>Europe/London</code> stands for the United
675 Kingdom, but its pre-1847 times are valid only for locations that
676 have London's exact meridian, and its 1847 transition
677 to <abbr>GMT</abbr> is known to be valid only for the L&NW and
678 the Caledonian railways.
681 The <code><abbr>tz</abbr></code> database does not record the
682 earliest time for which a timezone's
683 data entries are thereafter valid for every location in the region.
684 For example, <code>Europe/London</code> is valid for all locations
685 in its region after <abbr>GMT</abbr> was made the standard time,
686 but the date of standardization (1880-08-02) is not in the
687 <code><abbr>tz</abbr></code> database, other than in commentary.
688 For many timezones the earliest time of
692 The <code><abbr>tz</abbr></code> database does not record a
693 region's boundaries, and in many cases the boundaries are not known.
694 For example, the timezone
695 <code>America/Kentucky/Louisville</code> represents a region
696 around the city of Louisville, the boundaries of which are
700 Changes that are modeled as instantaneous transitions in the
701 <code><abbr>tz</abbr></code>
702 database were often spread out over hours, days, or even decades.
705 Even if the time is specified by law, locations sometimes
706 deliberately flout the law.
709 Early timekeeping practices, even assuming perfect clocks, were
710 often not specified to the accuracy that the
711 <code><abbr>tz</abbr></code> database requires.
714 The <code><abbr>tz</abbr></code> database cannot represent stopped clocks.
715 However, on 1911-03-11 at 00:00, some public-facing French clocks
716 were changed by stopping them for a few minutes to effect a transition.
717 The <code><abbr>tz</abbr></code> database models this via a
718 backward transition; the relevant French legislation does not
719 specify exactly how the transition was to occur.
722 Sometimes historical timekeeping was specified more precisely
723 than what the <code><abbr>tz</abbr></code> code can handle.
724 For example, from 1880 to 1916 clocks in Ireland observed Dublin Mean
725 Time (estimated to be <abbr>UT</abbr>
726 −00:25:21.1); although the <code><abbr>tz</abbr></code>
727 source data can represent the .1 second, TZif files and the code cannot.
728 In practice these old specifications were rarely if ever
729 implemented to subsecond precision.
732 Even when all the timestamp transitions recorded by the
733 <code><abbr>tz</abbr></code> database are correct, the
734 <code><abbr>tz</abbr></code> rules that generate them may not
735 faithfully reflect the historical rules.
736 For example, from 1922 until World War II the UK moved clocks
737 forward the day following the third Saturday in April unless that
738 was Easter, in which case it moved clocks forward the previous
740 Because the <code><abbr>tz</abbr></code> database has no
741 way to specify Easter, these exceptional years are entered as
742 separate <code><abbr>tz</abbr> Rule</code> lines, even though the
743 legal rules did not change.
744 When transitions are known but the historical rules behind them are not,
745 the database contains <code>Zone</code> and <code>Rule</code>
746 entries that are intended to represent only the generated
747 transitions, not any underlying historical rules; however, this
748 intent is recorded at best only in commentary.
751 The <code><abbr>tz</abbr></code> database models time
753 href="https://en.wikipedia.org/wiki/Proleptic_Gregorian_calendar">proleptic
754 Gregorian calendar</a> with days containing 24 equal-length hours
755 numbered 00 through 23, except when clock transitions occur.
756 Pre-standard time is modeled as local mean time.
757 However, historically many people used other calendars and other timescales.
758 For example, the Roman Empire used
759 the <a href="https://en.wikipedia.org/wiki/Julian_calendar">Julian
761 and <a href="https://en.wikipedia.org/wiki/Roman_timekeeping">Roman
762 timekeeping</a> had twelve varying-length daytime hours with a
763 non-hour-based system at night.
764 And even today, some local practices diverge from the Gregorian
765 calendar with 24-hour days. These divergences range from
766 relatively minor, such as Japanese bars giving times like "24:30" for the
767 wee hours of the morning, to more-significant differences such as <a
768 href="https://www.pri.org/stories/2015-01-30/if-you-have-meeting-ethiopia-you-better-double-check-time">the
769 east African practice of starting the day at dawn</a>, renumbering
770 the Western 06:00 to be 12:00. These practices are largely outside
771 the scope of the <code><abbr>tz</abbr></code> code and data, which
772 provide only limited support for date and time localization
773 such as that required by POSIX.
774 If <abbr>DST</abbr> is not used a different time zone
775 can often do the trick; for example, in Kenya a <code>TZ</code> setting
776 like <code><-03>3</code> or <code>America/Cayenne</code> starts
777 the day six hours later than <code>Africa/Nairobi</code> does.
780 Early clocks were less reliable, and data entries do not represent
784 The <code><abbr>tz</abbr></code> database assumes Universal Time
785 (<abbr>UT</abbr>) as an origin, even though <abbr>UT</abbr> is not
786 standardized for older timestamps.
787 In the <code><abbr>tz</abbr></code> database commentary,
788 <abbr>UT</abbr> denotes a family of time standards that includes
789 Coordinated Universal Time (<abbr>UTC</abbr>) along with other
790 variants such as <abbr>UT1</abbr> and <abbr>GMT</abbr>,
791 with days starting at midnight.
792 Although <abbr>UT</abbr> equals <abbr>UTC</abbr> for modern
793 timestamps, <abbr>UTC</abbr> was not defined until 1960, so
794 commentary uses the more-general abbreviation <abbr>UT</abbr> for
795 timestamps that might predate 1960.
796 Since <abbr>UT</abbr>, <abbr>UT1</abbr>, etc. disagree slightly,
797 and since pre-1972 <abbr>UTC</abbr> seconds varied in length,
798 interpretation of older timestamps can be problematic when
799 subsecond accuracy is needed.
802 Civil time was not based on atomic time before 1972, and we do not
804 <a href="https://en.wikipedia.org/wiki/Earth's_rotation">earth's
805 rotation</a> accurately enough to map <a
806 href="https://en.wikipedia.org/wiki/International_System_of_Units"><abbr
807 title="International System of Units">SI</abbr></a> seconds to
808 historical <a href="https://en.wikipedia.org/wiki/Solar_time">solar time</a>
809 to more than about one-hour accuracy.
810 See: Stephenson FR, Morrison LV, Hohenkerk CY.
811 <a href="https://dx.doi.org/10.1098/rspa.2016.0404">Measurement of
812 the Earth's rotation: 720 BC to AD 2015</a>.
813 <cite>Proc Royal Soc A</cite>. 2016;472:20160404.
814 Also see: Espenak F. <a
815 href="https://eclipse.gsfc.nasa.gov/SEhelp/uncertainty2004.html">Uncertainty
819 The relationship between POSIX time (that is, <abbr>UTC</abbr> but
820 ignoring <a href="https://en.wikipedia.org/wiki/Leap_second">leap
821 seconds</a>) and <abbr>UTC</abbr> is not agreed upon after 1972.
823 clock officially stops during an inserted leap second, at least one
824 proposed standard has it jumping back a second instead; and in
825 practice POSIX clocks more typically either progress glacially during
826 a leap second, or are slightly slowed while near a leap second.
829 The <code><abbr>tz</abbr></code> database does not represent how
830 uncertain its information is.
831 Ideally it would contain information about when data entries are
833 Partial temporal knowledge is a field of active research, though,
834 and it is not clear how to apply it here.
839 In short, many, perhaps most, of the <code><abbr>tz</abbr></code>
840 database's pre-1970 and future timestamps are either wrong or
842 Any attempt to pass the
843 <code><abbr>tz</abbr></code> database off as the definition of time
844 should be unacceptable to anybody who cares about the facts.
845 In particular, the <code><abbr>tz</abbr></code> database's
846 <abbr>LMT</abbr> offsets should not be considered meaningful, and
847 should not prompt creation of timezones
848 merely because two locations
849 differ in <abbr>LMT</abbr> or transitioned to standard time at
855 <h2 id="functions">Time and date functions</h2>
857 The <code><abbr>tz</abbr></code> code contains time and date functions
858 that are upwards compatible with those of POSIX.
859 Code compatible with this package is already
860 <a href="tz-link.html#tzdb">part of many platforms</a>, where the
861 primary use of this package is to update obsolete time-related files.
862 To do this, you may need to compile the time zone compiler
863 '<code>zic</code>' supplied with this package instead of using the
864 system '<code>zic</code>', since the format of <code>zic</code>'s
865 input is occasionally extended, and a platform may still be shipping
866 an older <code>zic</code>.
869 <h3 id="POSIX">POSIX properties and limitations</h3>
873 In POSIX, time display in a process is controlled by the
874 environment variable <code>TZ</code>.
875 Unfortunately, the POSIX
876 <code>TZ</code> string takes a form that is hard to describe and
877 is error-prone in practice.
878 Also, POSIX <code>TZ</code> strings cannot deal with daylight
879 saving time rules not based on the Gregorian calendar (as in
880 Iran), or with situations where more than two time zone
881 abbreviations or <abbr>UT</abbr> offsets are used in an area.
885 The POSIX <code>TZ</code> string takes the following form:
889 <var>stdoffset</var>[<var>dst</var>[<var>offset</var>][<code>,</code><var>date</var>[<code>/</code><var>time</var>]<code>,</code><var>date</var>[<code>/</code><var>time</var>]]]
897 <dt><var>std</var> and <var>dst</var></dt><dd>
898 are 3 or more characters specifying the standard
899 and daylight saving time (<abbr>DST</abbr>) zone abbreviations.
900 Starting with POSIX.1-2001, <var>std</var> and <var>dst</var>
901 may also be in a quoted form like '<code><+09></code>';
902 this allows "<code>+</code>" and "<code>-</code>" in the names.
904 <dt><var>offset</var></dt><dd>
906 '<code>[±]<var>hh</var>:[<var>mm</var>[:<var>ss</var>]]</code>'
907 and specifies the offset west of <abbr>UT</abbr>.
908 '<var>hh</var>' may be a single digit;
909 0≤<var>hh</var>≤24.
910 The default <abbr>DST</abbr> offset is one hour ahead of
913 <dt><var>date</var>[<code>/</code><var>time</var>]<code>,</code><var>date</var>[<code>/</code><var>time</var>]</dt><dd>
914 specifies the beginning and end of <abbr>DST</abbr>.
915 If this is absent, the system supplies its own ruleset
916 for <abbr>DST</abbr>, and its rules can differ from year to year;
917 typically <abbr>US</abbr> <abbr>DST</abbr> rules are used.
919 <dt><var>time</var></dt><dd>
921 '<var>hh</var><code>:</code>[<var>mm</var>[<code>:</code><var>ss</var>]]'
922 and defaults to 02:00.
923 This is the same format as the offset, except that a
924 leading '<code>+</code>' or '<code>-</code>' is not allowed.
926 <dt><var>date</var></dt><dd>
927 takes one of the following forms:
929 <dt>J<var>n</var> (1≤<var>n</var>≤365)</dt><dd>
930 origin-1 day number not counting February 29
932 <dt><var>n</var> (0≤<var>n</var>≤365)</dt><dd>
933 origin-0 day number counting February 29 if present
935 <dt><code>M</code><var>m</var><code>.</code><var>n</var><code>.</code><var>d</var>
936 (0[Sunday]≤<var>d</var>≤6[Saturday], 1≤<var>n</var>≤5,
937 1≤<var>m</var>≤12)</dt><dd>
938 for the <var>d</var>th day of week <var>n</var> of
939 month <var>m</var> of the year, where week 1 is the first
940 week in which day <var>d</var> appears, and
941 '<code>5</code>' stands for the last week in which
942 day <var>d</var> appears (which may be either the 4th or
944 Typically, this is the only useful form; the <var>n</var>
945 and <code>J</code><var>n</var> forms are rarely used.
952 Here is an example POSIX <code>TZ</code> string for New
954 It says that standard time (<abbr>NZST</abbr>) is 12 hours ahead
955 of <abbr>UT</abbr>, and that daylight saving time
956 (<abbr>NZDT</abbr>) is observed from September's last Sunday at
957 02:00 until April's first Sunday at 03:00:
960 <pre><code>TZ='NZST-12NZDT,M9.5.0,M4.1.0/3'</code></pre>
963 This POSIX <code>TZ</code> string is hard to remember, and
964 mishandles some timestamps before 2008.
965 With this package you can use this instead:
968 <pre><code>TZ='Pacific/Auckland'</code></pre>
971 POSIX does not define the <abbr>DST</abbr> transitions
972 for <code>TZ</code> values like
973 "<code>EST5EDT</code>".
974 Traditionally the current <abbr>US</abbr> <abbr>DST</abbr> rules
975 were used to interpret such values, but this meant that the
976 <abbr>US</abbr> <abbr>DST</abbr> rules were compiled into each
977 program that did time conversion. This meant that when
978 <abbr>US</abbr> time conversion rules changed (as in the United
979 States in 1987), all programs that did time conversion had to be
980 recompiled to ensure proper results.
983 The <code>TZ</code> environment variable is process-global, which
984 makes it hard to write efficient, thread-safe applications that
985 need access to multiple timezones.
988 In POSIX, there is no tamper-proof way for a process to learn the
989 system's best idea of local (wall clock) time.
990 This is important for applications that an administrator wants
991 used only at certain times – without regard to whether the
993 <code>TZ</code> environment variable.
994 While an administrator can "do everything in <abbr>UT</abbr>" to
995 get around the problem, doing so is inconvenient and precludes
996 handling daylight saving time shifts – as might be required to
997 limit phone calls to off-peak hours.
1000 POSIX provides no convenient and efficient way to determine
1001 the <abbr>UT</abbr> offset and time zone abbreviation of arbitrary
1002 timestamps, particularly for timezones
1003 that do not fit into the POSIX model.
1006 POSIX requires that <code>time_t</code> clock counts exclude leap
1010 The <code><abbr>tz</abbr></code> code attempts to support all the
1011 <code>time_t</code> implementations allowed by POSIX.
1012 The <code>time_t</code> type represents a nonnegative count of seconds
1013 since 1970-01-01 00:00:00 <abbr>UTC</abbr>, ignoring leap seconds.
1014 In practice, <code>time_t</code> is usually a signed 64- or 32-bit
1015 integer; 32-bit signed <code>time_t</code> values stop working after
1016 2038-01-19 03:14:07 <abbr>UTC</abbr>, so new implementations these
1017 days typically use a signed 64-bit integer.
1018 Unsigned 32-bit integers are used on one or two platforms, and 36-bit
1019 and 40-bit integers are also used occasionally.
1020 Although earlier POSIX versions allowed <code>time_t</code> to be a
1021 floating-point type, this was not supported by any practical system,
1022 and POSIX.1-2013 and the <code><abbr>tz</abbr></code> code both
1023 require <code>time_t</code> to be an integer type.
1027 <h3 id="POSIX-extensions">Extensions to POSIX in the
1028 <code><abbr>tz</abbr></code> code</h3>
1032 The <code>TZ</code> environment variable is used in generating
1033 the name of a file from which time-related information is read
1034 (or is interpreted à la POSIX); <code>TZ</code> is no longer
1035 constrained to be a string containing abbreviations
1036 and numeric data as described <a href="#POSIX">above</a>.
1037 The file's format is <dfn><abbr>TZif</abbr></dfn>,
1038 a timezone information format that contains binary data; see
1039 <a href="https://datatracker.ietf.org/doc/html/8536">Internet
1040 <abbr>RFC</abbr> 8536</a>.
1041 The daylight saving time rules to be used for a
1042 particular timezone are encoded in the
1043 <abbr>TZif</abbr> file; the format of the file allows <abbr>US</abbr>,
1044 Australian, and other rules to be encoded, and
1045 allows for situations where more than two time zone
1046 abbreviations are used.
1049 It was recognized that allowing the <code>TZ</code> environment
1050 variable to take on values such as '<code>America/New_York</code>'
1051 might cause "old" programs (that expect <code>TZ</code> to have a
1052 certain form) to operate incorrectly; consideration was given to using
1053 some other environment variable (for example, <code>TIMEZONE</code>)
1054 to hold the string used to generate the <abbr>TZif</abbr> file's name.
1055 In the end, however, it was decided to continue using
1056 <code>TZ</code>: it is widely used for time zone purposes;
1057 separately maintaining both <code>TZ</code>
1058 and <code>TIMEZONE</code> seemed a nuisance; and systems where
1059 "new" forms of <code>TZ</code> might cause problems can simply
1060 use legacy <code>TZ</code> values such as "<code>EST5EDT</code>" which
1061 can be used by "new" programs as well as by "old" programs that
1062 assume pre-POSIX <code>TZ</code> values.
1066 The code supports platforms with a <abbr>UT</abbr> offset member
1067 in <code>struct tm</code>, e.g., <code>tm_gmtoff</code>,
1068 or with a time zone abbreviation member in
1069 <code>struct tm</code>, e.g., <code>tm_zone</code>. As noted
1070 in <a href="https://austingroupbugs.net/view.php?id=1533">Austin
1071 Group defect 1533</a>, a future version of POSIX is planned to
1072 require <code>tm_gmtoff</code> and <code>tm_zone</code>.
1075 Functions <code>tzalloc</code>, <code>tzfree</code>,
1076 <code>localtime_rz</code>, and <code>mktime_z</code> for
1077 more-efficient thread-safe applications that need to use multiple
1079 The <code>tzalloc</code> and <code>tzfree</code> functions
1080 allocate and free objects of type <code>timezone_t</code>,
1081 and <code>localtime_rz</code> and <code>mktime_z</code> are
1082 like <code>localtime_r</code> and <code>mktime</code> with an
1083 extra <code>timezone_t</code> argument.
1084 The functions were inspired by <a href="https://netbsd.org/">NetBSD</a>.
1087 Negative <code>time_t</code> values are supported, on systems
1088 where <code>time_t</code> is signed.
1091 These functions can account for leap seconds;
1092 see <a href="#leapsec">Leap seconds</a> below.
1096 <h3 id="vestigial">POSIX features no longer needed</h3>
1098 POSIX and <a href="https://en.wikipedia.org/wiki/ISO_C"><abbr>ISO</abbr> C</a>
1099 define some <a href="https://en.wikipedia.org/wiki/API"><abbr
1100 title="application programming interface">API</abbr>s</a> that are vestigial:
1101 they are not needed, and are relics of a too-simple model that does
1102 not suffice to handle many real-world timestamps.
1103 Although the <code><abbr>tz</abbr></code> code supports these
1104 vestigial <abbr>API</abbr>s for backwards compatibility, they should
1105 be avoided in portable applications.
1106 The vestigial <abbr>API</abbr>s are:
1110 The POSIX <code>tzname</code> variable does not suffice and is no
1112 To get a timestamp's time zone abbreviation, consult
1113 the <code>tm_zone</code> member if available; otherwise,
1114 use <code>strftime</code>'s <code>"%Z"</code> conversion
1118 The POSIX <code>daylight</code> and <code>timezone</code>
1119 variables do not suffice and are no longer needed.
1120 To get a timestamp's <abbr>UT</abbr> offset, consult
1121 the <code>tm_gmtoff</code> member if available; otherwise,
1122 subtract values returned by <code>localtime</code>
1123 and <code>gmtime</code> using the rules of the Gregorian calendar,
1124 or use <code>strftime</code>'s <code>"%z"</code> conversion
1125 specification if a string like <code>"+0900"</code> suffices.
1128 The <code>tm_isdst</code> member is almost never needed and most of
1129 its uses should be discouraged in favor of the abovementioned
1131 Although it can still be used in arguments to
1132 <code>mktime</code> to disambiguate timestamps near
1133 a <abbr>DST</abbr> transition when the clock jumps back on
1134 platforms lacking <code>tm_gmtoff</code>, this
1135 disambiguation does not work when standard time itself jumps back,
1136 which can occur when a location changes to a time zone with a
1137 lesser <abbr>UT</abbr> offset.
1141 <h3 id="other-portability">Other portability notes</h3>
1144 The <a href="https://en.wikipedia.org/wiki/Version_7_Unix">7th Edition
1145 UNIX</a> <code>timezone</code> function is not present in this
1146 package; it is impossible to reliably map <code>timezone</code>'s
1147 arguments (a "minutes west of <abbr>GMT</abbr>" value and a
1148 "daylight saving time in effect" flag) to a time zone
1149 abbreviation, and we refuse to guess.
1150 Programs that in the past used the <code>timezone</code> function
1151 may now examine <code>localtime(&clock)->tm_zone</code>
1152 (if <code>TM_ZONE</code> is defined) or
1153 <code>tzname[localtime(&clock)->tm_isdst]</code>
1154 (if <code>HAVE_TZNAME</code> is nonzero) to learn the correct time
1155 zone abbreviation to use.
1159 href="https://en.wikipedia.org/wiki/History_of_the_Berkeley_Software_Distribution#4.2BSD"><abbr>4.2BSD</abbr></a>
1160 <code>gettimeofday</code> function is not
1161 used in this package.
1162 This formerly let users obtain the current <abbr>UTC</abbr> offset
1163 and <abbr>DST</abbr> flag, but this functionality was removed in
1164 later versions of <abbr>BSD</abbr>.
1167 In <abbr>SVR2</abbr>, time conversion fails for near-minimum or
1168 near-maximum <code>time_t</code> values when doing conversions
1169 for places that do not use <abbr>UT</abbr>.
1170 This package takes care to do these conversions correctly.
1171 A comment in the source code tells how to get compatibly wrong
1175 The functions that are conditionally compiled
1176 if <code>STD_INSPIRED</code> is defined should, at this point, be
1177 looked on primarily as food for thought.
1178 They are not in any sense "standard compatible" – some are
1179 not, in fact, specified in <em>any</em> standard.
1180 They do, however, represent responses of various authors to
1181 standardization proposals.
1184 Other time conversion proposals, in particular those supported by the
1185 <a href="https://howardhinnant.github.io/date/tz.html">Time Zone
1186 Database Parser</a>, offer a wider selection of functions
1187 that provide capabilities beyond those provided here.
1188 The absence of such functions from this package is not meant to
1189 discourage the development, standardization, or use of such
1191 Rather, their absence reflects the decision to make this package
1192 contain valid extensions to POSIX, to ensure its broad
1194 If more powerful time conversion functions can be standardized, so
1201 <h2 id="stability">Interface stability</h2>
1203 The <code><abbr>tz</abbr></code> code and data supply the following interfaces:
1208 A set of timezone names as per
1209 "<a href="#naming">Timezone identifiers</a>" above.
1212 Library functions described in "<a href="#functions">Time and date
1213 functions</a>" above.
1216 The programs <code>tzselect</code>, <code>zdump</code>,
1217 and <code>zic</code>, documented in their man pages.
1220 The format of <code>zic</code> input files, documented in
1221 the <code>zic</code> man page.
1224 The format of <code>zic</code> output files, documented in
1225 the <code>tzfile</code> man page.
1228 The format of zone table files, documented in <code>zone1970.tab</code>.
1231 The format of the country code file, documented in <code>iso3166.tab</code>.
1234 The version number of the code and data, as the first line of
1235 the text file '<code>version</code>' in each release.
1240 Interface changes in a release attempt to preserve compatibility with
1242 For example, <code><abbr>tz</abbr></code> data files typically do not
1243 rely on recently-added <code>zic</code> features, so that users can
1244 run older <code>zic</code> versions to process newer data files.
1245 <a href="tz-link.html#download">Downloading
1246 the <code><abbr>tz</abbr></code> database</a> describes how releases
1247 are tagged and distributed.
1251 Interfaces not listed above are less stable.
1252 For example, users should not rely on particular <abbr>UT</abbr>
1253 offsets or abbreviations for timestamps, as data entries are often
1254 based on guesswork and these guesses may be corrected or improved.
1258 Timezone boundaries are not part of the stable interface.
1259 For example, even though the <samp>Asia/Bangkok</samp> timezone
1260 currently includes Chang Mai, Hanoi, and Phnom Penh, this is not part
1261 of the stable interface and the timezone can split at any time.
1262 If a calendar application records a future event in some location other
1263 than Bangkok by putting "<samp>Asia/Bangkok</samp>" in the event's record,
1264 the application should be robust in the presence of timezone splits
1265 between now and the future time.
1270 <h2 id="leapsec">Leap seconds</h2>
1272 The <code><abbr>tz</abbr></code> code and data can account for leap seconds,
1273 thanks to code contributed by Bradley White.
1274 However, the leap second support of this package is rarely used directly
1275 because POSIX requires leap seconds to be excluded and many
1276 software packages would mishandle leap seconds if they were present.
1277 Instead, leap seconds are more commonly handled by occasionally adjusting
1278 the operating system kernel clock as described in
1279 <a href="tz-link.html#precision">Precision timekeeping</a>,
1280 and this package by default installs a <samp>leapseconds</samp> file
1282 <a href="https://www.ntp.org"><abbr title="Network Time Protocol">NTP</abbr></a>
1283 software that adjusts the kernel clock.
1284 However, kernel-clock twiddling approximates UTC only roughly,
1285 and systems needing more-precise UTC can use this package's leap
1286 second support directly.
1290 The directly-supported mechanism assumes that <code>time_t</code>
1291 counts of seconds since the POSIX epoch normally include leap seconds,
1292 as opposed to POSIX <code>time_t</code> counts which exclude leap seconds.
1293 This modified timescale is converted to <abbr>UTC</abbr>
1294 at the same point that time zone and <abbr>DST</abbr>
1295 adjustments are applied –
1296 namely, at calls to <code>localtime</code> and analogous functions –
1297 and the process is driven by leap second information
1298 stored in alternate versions of the <abbr>TZif</abbr> files.
1299 Because a leap second adjustment may be needed even
1300 if no time zone correction is desired,
1301 calls to <code>gmtime</code>-like functions
1302 also need to consult a <abbr>TZif</abbr> file,
1303 conventionally named <samp><abbr>Etc/UTC</abbr></samp>
1304 (<samp><abbr>GMT</abbr></samp> in previous versions),
1305 to see whether leap second corrections are needed.
1306 To convert an application's <code>time_t</code> timestamps to or from
1307 POSIX <code>time_t</code> timestamps (for use when, say,
1308 embedding or interpreting timestamps in portable
1309 <a href="https://en.wikipedia.org/wiki/Tar_(computing)"><code>tar</code></a>
1311 the application can call the utility functions
1312 <code>time2posix</code> and <code>posix2time</code>
1313 included with this package.
1317 If the POSIX-compatible <abbr>TZif</abbr> file set is installed
1318 in a directory whose basename is <samp>zoneinfo</samp>, the
1319 leap-second-aware file set is by default installed in a separate
1320 directory <samp>zoneinfo-leaps</samp>.
1321 Although each process can have its own time zone by setting
1322 its <code>TZ</code> environment variable, there is no support for some
1323 processes being leap-second aware while other processes are
1324 POSIX-compatible; the leap-second choice is system-wide.
1325 So if you configure your kernel to count leap seconds, you should also
1326 discard <samp>zoneinfo</samp> and rename <samp>zoneinfo-leaps</samp>
1327 to <samp>zoneinfo</samp>.
1328 Alternatively, you can install just one set of <abbr>TZif</abbr> files
1329 in the first place; see the <code>REDO</code> variable in this package's
1330 <a href="https://en.wikipedia.org/wiki/Makefile">makefile</a>.
1335 <h2 id="calendar">Calendrical issues</h2>
1337 Calendrical issues are a bit out of scope for a time zone database,
1338 but they indicate the sort of problems that we would run into if we
1339 extended the time zone database further into the past.
1340 An excellent resource in this area is Edward M. Reingold
1341 and Nachum Dershowitz, <cite><a
1342 href="https://www.cambridge.org/fr/academic/subjects/computer-science/computing-general-interest/calendrical-calculations-ultimate-edition-4th-edition">Calendrical
1343 Calculations: The Ultimate Edition</a></cite>, Cambridge University Press (2018).
1344 Other information and sources are given in the file '<code>calendars</code>'
1345 in the <code><abbr>tz</abbr></code> distribution.
1346 They sometimes disagree.
1351 <h2 id="planets">Time and time zones on other planets</h2>
1353 Some people's work schedules have used
1354 <a href="https://en.wikipedia.org/wiki/Timekeeping_on_Mars">Mars time</a>.
1355 Jet Propulsion Laboratory (JPL) coordinators kept Mars time on
1357 <a href="https://en.wikipedia.org/wiki/Mars_Pathfinder">Mars
1358 Pathfinder</a> mission (1997).
1359 Some of their family members also adapted to Mars time.
1360 Dozens of special Mars watches were built for JPL workers who kept
1361 Mars time during the
1362 <a href="https://en.wikipedia.org/wiki/Mars_Exploration_Rover">Mars
1363 Exploration Rovers (MER)</a> mission (2004–2018).
1364 These timepieces looked like normal Seikos and Citizens but were adjusted
1365 to use Mars seconds rather than terrestrial seconds, although
1366 unfortunately the adjusted watches were unreliable and appear to have
1367 had only limited use.
1371 A Mars solar day is called a "sol" and has a mean period equal to
1372 about 24 hours 39 minutes 35.244 seconds in terrestrial time.
1373 It is divided into a conventional 24-hour clock, so each Mars second
1374 equals about 1.02749125 terrestrial seconds.
1375 (One MER worker noted, "If I am working Mars hours, and Mars hours are
1376 2.5% more than Earth hours, shouldn't I get an extra 2.5% pay raise?")
1380 The <a href="https://en.wikipedia.org/wiki/Prime_meridian">prime
1381 meridian</a> of Mars goes through the center of the crater
1382 <a href="https://en.wikipedia.org/wiki/Airy-0">Airy-0</a>, named in
1383 honor of the British astronomer who built the Greenwich telescope that
1384 defines Earth's prime meridian.
1385 Mean solar time on the Mars prime meridian is
1386 called Mars Coordinated Time (<abbr>MTC</abbr>).
1390 Each landed mission on Mars has adopted a different reference for
1391 solar timekeeping, so there is no real standard for Mars time zones.
1392 For example, the MER mission defined two time zones "Local
1393 Solar Time A" and "Local Solar Time B" for its two missions, each zone
1394 designed so that its time equals local true solar time at
1395 approximately the middle of the nominal mission.
1396 The A and B zones differ enough so that an MER worker assigned to
1397 the A zone might suffer "Mars lag" when switching to work in the B zone.
1398 Such a "time zone" is not particularly suited for any application
1399 other than the mission itself.
1403 Many calendars have been proposed for Mars, but none have achieved
1405 Astronomers often use Mars Sol Date (<abbr>MSD</abbr>) which is a
1406 sequential count of Mars solar days elapsed since about 1873-12-29
1407 12:00 <abbr>GMT</abbr>.
1411 In our solar system, Mars is the planet with time and calendar most
1413 On other planets, Sun-based time and calendars would work quite
1415 For example, although Mercury's
1416 <a href="https://en.wikipedia.org/wiki/Rotation_period">sidereal
1417 rotation period</a> is 58.646 Earth days, Mercury revolves around the
1418 Sun so rapidly that an observer on Mercury's equator would see a
1419 sunrise only every 175.97 Earth days, i.e., a Mercury year is 0.5 of a
1421 Venus is more complicated, partly because its rotation is slightly
1422 <a href="https://en.wikipedia.org/wiki/Retrograde_motion">retrograde</a>:
1423 its year is 1.92 of its days.
1424 Gas giants like Jupiter are trickier still, as their polar and
1425 equatorial regions rotate at different rates, so that the length of a
1426 day depends on latitude.
1427 This effect is most pronounced on Neptune, where the day is about 12
1428 hours at the poles and 18 hours at the equator.
1432 Although the <code><abbr>tz</abbr></code> database does not support
1433 time on other planets, it is documented here in the hopes that support
1434 will be added eventually.
1438 Sources for time on other planets:
1443 Michael Allison and Robert Schmunk,
1444 "<a href="https://www.giss.nasa.gov/tools/mars24/help/notes.html">Technical
1445 Notes on Mars Solar Time as Adopted by the Mars24 Sunclock</a>"
1450 <em><a href="https://mitpress.mit.edu/books/making-time-mars">Making
1451 Time on Mars</a></em>, MIT Press (March 2020), ISBN 978-0262043854.
1455 "<a href="https://www.latimes.com/archives/la-xpm-2004-jan-14-sci-marstime14-story.html">Workdays
1456 Fit for a Martian</a>", <cite>Los Angeles Times</cite>
1457 (2004-01-14), pp A1, A20–A21.
1461 "<a href="https://www.theatlantic.com/technology/archive/2015/02/jet-lag-is-worse-on-mars/386033/">Jet
1462 Lag Is Worse on Mars</a>", <cite>The Atlantic</cite> (2015-02-26)
1466 "<a href="https://www.universetoday.com/37481/days-of-the-planets/">How
1467 long is a day on the other planets of the solar system?</a>"
1475 This file is in the public domain, so clarified as of 2009-05-17 by