- tz - lists.iana.org

Some extra information about the Eucla zone
by Gilmore Davidson May 17, 2019

May 17, 2019

After Australia’s clock change last weekend, the national broadcaster has published an article about the Eucla time zone: https://www.abc.net.au/news/2019-04-08/this-remote-stretch-of-desert-has-it… <https://www.abc.net.au/news/2019-04-08/this-remote-stretch-of-desert-has-it…> > Sergeant Chamberlain said when it comes to life on the Nullarbor, the custom made time zone was at the lower end of quirky. Some interesting snippets of information include a rough description of the geographical boundaries... > The time zone exists for about 340 kilometres and takes in the tiny roadhouse communities of Cocklebiddy, Madura, Madura, Eucla and Border Village. …and an indication that the zone has definitely been in existence since before the 1970 cut-off of the database: > Mr Pike has lived on the Nullarbor for 47 years > > […] > > 'Why'? > > "Good question," Mr Pike said. > > "I don't even know that, and it's been going for over 50 years.” This doesn’t change any of the existing data, but could be useful as a reference. Cheers, Gil

2 1

[PROPOSED] Ignore more non-source files
by Paul Eggert May 16, 2019

May 16, 2019

* .gitignore: Ignore *.diff, *.i, *.orig, *.rej, *.tar. Ignore check_* files, as they are created by ‘make check’. --- .gitignore | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/.gitignore b/.gitignore index 1e62d11..cf3b825 100644 --- a/.gitignore +++ b/.gitignore @@ -2,14 +2,20 @@ # This file is in the public domain. *.a *.asc +*.diff +*.i *.o +*.orig *.patch +*.rej +*.tar *.tar.* *.txt *.tzs *.zi *~ ChangeLog +check_* date leapseconds tzselect -- 2.21.0

1 0

[PROPOSED] Bring zic man page up to date
by Paul Eggert May 16, 2019

May 16, 2019

* zic.8: Don’t mention time(2), since zic does not rely on time_t any more. Mention other ways that -v generates warnings. ON fields can’t contain non-space white space characters either. Remove incorrect assertion that fractional seconds are not allowed. --- zic.8 | 60 +++++++++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 48 insertions(+), 12 deletions(-) diff --git a/zic.8 b/zic.8 index 524a113..6d893c2 100644 --- a/zic.8 +++ b/zic.8 @@ -110,9 +110,7 @@ Be more verbose, and complain about the following situations: The input specifies a link to a link. .PP A year that appears in a data file is outside the range -of years representable by -.BR time (2) -values. +of representable years. .PP A time of 24:00 or more appears in the input. Pre-1998 versions of @@ -124,6 +122,36 @@ Pre-2004 versions of .B zic prohibit this. .PP +A time zone abbreviation uses a +.B %z +format. +Pre-2015 versions of +.B zic +do not support this. +.PP +A timestamp contains fractional seconds. +Pre-2018 versions of +.B zic +do not support this. +.PP +The input contains abbreviations that are mishandled by pre-2018 versions of +.B zic +due to a longstanding coding bug. +These abbreviations include +.q L +for +.q Link , +.q mi +for +.q min , +.q Sa +for +.q Sat , +and +.q Su +for +.q Sun . +.PP The output file does not contain all the information about the long-term future of a timezone, because the future cannot be summarized as an extended POSIX TZ string. For example, as of 2019 this problem @@ -137,8 +165,15 @@ code designed for older output formats. These compatibility issues affect only timestamps before 1970 or after the start of 2038. .PP -A time zone abbreviation has fewer than 3 characters. -POSIX requires at least 3. +The output file contains more than 1200 transitions, +which may be mishandled by some clients. +The current reference client supports at most 2000 transitions; +pre-2014 versions of the reference client support at most 1200 +transitions. +.PP +A time zone abbreviation has fewer than 3 or more than 6 characters. +POSIX requires at least 3, and requires implementations to support +at least 6. .PP An output file name contains a byte that is not an ASCII letter, .q "\*-" , @@ -279,7 +314,7 @@ or a weekday name preceded by (e.g., .BR "lastSunday" ) may be abbreviated or spelled out in full. -Note that there must be no spaces within the +There must be no white space characters within the .B ON field. .TP @@ -455,7 +490,7 @@ and .TP .B UNTIL The time at which the UT offset or the rule(s) change for a location. -It takes the form of YEAR [MONTH [DAY [TIME]]]. +It takes the form of one to four fields YEAR [MONTH [DAY [TIME]]]. If this is specified, the time zone information is generated from the given UT offset and rule change until the time specified, which is interpreted using @@ -587,10 +622,10 @@ Rule EU 1979 1995 \*- Sep lastSun 1:00u 0 \*- Rule EU 1981 max \*- Mar lastSun 1:00u 1:00 S Rule EU 1996 max \*- Oct lastSun 1:00u 0 \*- .sp -.ta \w'# Zone\0\0'u +\w'Europe/Zurich\0\0'u +\w'0:34:08\0\0'u +\w'RULES\0\0'u +\w'FORMAT\0\0'u +.ta \w'# Zone\0\0'u +\w'Europe/Zurich\0\0'u +\w'0:29:45.50\0\0'u +\w'RULES\0\0'u +\w'FORMAT\0\0'u # Zone NAME UTOFF RULES FORMAT [UNTIL] Zone Europe/Zurich 0:34:08 \*- LMT 1853 Jul 16 - 0:29:46 \*- BMT 1894 Jun + 0:29:45.50 \*- BMT 1894 Jun 1:00 Swiss CE%sT 1981 1:00 EU CE%sT .sp @@ -601,9 +636,10 @@ Link Europe/Zurich Europe/Vaduz In this example, the timezone is named Europe/Zurich but it has an alias as Europe/Vaduz. This example says that Zurich was 34 minutes and 8 seconds east of UT until 1853-07-16 at 00:00, when the legal offset -was changed to 7\(de\|26\(fm\|22.50\(sd; although this works out to -0:29:45.50, the input format cannot represent fractional seconds so it -is rounded here. After 1894-06-01 at 00:00 the UT offset became one hour +was changed to 7\(de\|26\(fm\|22.50\(sd, which works out to 0:29:45.50; +.B zic +treats this by rounding it to 0:29:46. +After 1894-06-01 at 00:00 the UT offset became one hour and Swiss daylight saving rules (defined with lines beginning with .q "Rule Swiss") apply. From 1981 to the present, EU daylight saving rules have -- 2.21.0

1 0

[PROPOSED] Assume zic 2017c or later when shrinking
by Paul Eggert May 16, 2019

May 16, 2019

The tzdata.zi file was first released in 2017c, and deliberately avoided zic features that were also new to that release, to avoid problems with people mistakenly running older zic when generating this new file. After a year’s wait it is reasonable to assume that anybody using this new file is also using a new-enough zic. This saves about 900 bytes in tzdata.zi. * NEWS: Mention this. * zishrink.awk (gen_rule_name, prehash_rule_names) (process_input_line, output_saved_lines): For clarity, use the convention that parameters are on the first line of function definition, and local vars are on later lines. (process_input_line): Assume zic 2017c or later. Make line types more explicit. --- NEWS | 5 +++++ zishrink.awk | 44 +++++++++++++++++++++++--------------------- 2 files changed, 28 insertions(+), 21 deletions(-) diff --git a/NEWS b/NEWS index c939b8f..be6dd26 100644 --- a/NEWS +++ b/NEWS @@ -25,6 +25,11 @@ Unreleased, experimental changes Europe/Rome between those dates. (Thanks to Stephen Trainor and Luigi Rosa.) + Changes to build procedure + + tzdata.zi now assumes zic 2017c or later. This shrinks tzdata.zi + by a percent or so. + Release 2019a - 2019-03-25 22:01:33 -0700 diff --git a/zishrink.awk b/zishrink.awk index 8876b68..4e187ac 100644 --- a/zishrink.awk +++ b/zishrink.awk @@ -20,7 +20,8 @@ function record_hash(n, name) # Return a shortened rule name representing NAME, # and record this relationship to the hash table. -function gen_rule_name(name, n) +function gen_rule_name(name, \ + n) { # Use a simple memonic: the first two letters. n = substr(name, 1, 2) @@ -29,7 +30,8 @@ function gen_rule_name(name, n) return n } -function prehash_rule_names(name) +function prehash_rule_names( \ + name) { # Rule names are not part of the tzdb API, so substitute shorter # ones. Shortening them consistently from one release to the next @@ -148,20 +150,21 @@ function prehash_rule_names(name) } } -# Process an input line and save it for later output. +# Process the input line LINE and save it for later output. -function process_input_line(line, field, end, i, n, startdef) +function process_input_line(line, \ + field, end, i, n, startdef, \ + linkline, ruleline, zoneline) { # Remove comments, normalize spaces, and append a space to each line. sub(/#.*/, "", line) line = line " " gsub(/[\t ]+/, " ", line) - # Abbreviate keywords. Do not abbreviate "Link" to just "L", - # as pre-2017c zic erroneously diagnoses "Li" as ambiguous. - sub(/^Link /, "Li ", line) - sub(/^Rule /, "R ", line) - sub(/^Zone /, "Z ", line) + # Abbreviate keywords and determine line type. + linkline = sub(/^Link /, "L ", line) + ruleline = sub(/^Rule /, "R ", line) + zoneline = sub(/^Zone /, "Z ", line) # SystemV rules are not needed. if (line ~ /^R SystemV /) return @@ -169,7 +172,7 @@ function process_input_line(line, field, end, i, n, startdef) # Replace FooAsia rules with the same rules without "Asia", as they # are duplicates. if (match(line, /[^ ]Asia /)) { - if (line ~ /^R /) return + if (ruleline) return line = substr(line, 1, RSTART) substr(line, RSTART + 5) } @@ -179,21 +182,19 @@ function process_input_line(line, field, end, i, n, startdef) while (match(line, /:0[^:]/)) line = substr(line, 1, RSTART - 1) substr(line, RSTART + 2) - # Abbreviate weekday names. Do not abbreviate "Sun" and "Sat", as - # pre-2017c zic erroneously diagnoses "Su" and "Sa" as ambiguous. + # Abbreviate weekday names. while (match(line, / (last)?(Mon|Wed|Fri)[ <>]/)) { end = RSTART + RLENGTH line = substr(line, 1, end - 4) substr(line, end - 1) } - while (match(line, / (last)?(Tue|Thu)[ <>]/)) { + while (match(line, / (last)?(Sun|Tue|Thu|Sat)[ <>]/)) { end = RSTART + RLENGTH line = substr(line, 1, end - 3) substr(line, end - 1) } - # Abbreviate "max", "only" and month names. - # Do not abbreviate "min", as pre-2017c zic erroneously diagnoses "mi" - # as ambiguous. + # Abbreviate "max", "min", "only" and month names. gsub(/ max /, " ma ", line) + gsub(/ min /, " mi ", line) gsub(/ only /, " o ", line) gsub(/ Jan /, " Ja ", line) gsub(/ Feb /, " F ", line) @@ -221,7 +222,7 @@ function process_input_line(line, field, end, i, n, startdef) n = split(line, field) # Abbreviate rule names. - i = field[1] == "Z" ? 4 : field[1] == "Li" ? 0 : 2 + i = zoneline ? 4 : linkline ? 0 : 2 if (i && field[i] ~ /^[^-+0-9]/) { if (!rule[field[i]]) rule[field[i]] = gen_rule_name(field[i]) @@ -231,11 +232,11 @@ function process_input_line(line, field, end, i, n, startdef) # If this zone supersedes an earlier one, delete the earlier one # from the saved output lines. startdef = "" - if (field[1] == "Z") + if (zoneline) zonename = startdef = field[2] - else if (field[1] == "Li") + else if (linkline) zonename = startdef = field[3] - else if (field[1] == "R") + else if (ruleline) zonename = "" if (startdef) { i = zonedef[startdef] @@ -254,7 +255,8 @@ function process_input_line(line, field, end, i, n, startdef) output_line[nout++] = line } -function output_saved_lines(i) +function output_saved_lines( \ + i) { for (i = 0; i < nout; i++) if (output_line[i]) -- 2.21.0

1 0

[PROPOSED] zic: omit pre-2017c chatter unless -v
by Paul Eggert May 16, 2019

May 16, 2019

* zic.c (byword): Put pre-2017c warnings in the same category as other pre-2015 etc. warnings: that is, output the warnings only if -v is given. --- zic.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/zic.c b/zic.c index 4bafe5d..5b932dc 100644 --- a/zic.c +++ b/zic.c @@ -3140,8 +3140,8 @@ byword(const char *word, const struct lookup *table) else return NULL; /* multiple inexact matches */ } - /* Warn about any backward-compatibility issue with pre-2017c zic. */ - if (foundlp) { + if (foundlp && noise) { + /* Warn about any backward-compatibility issue with pre-2017c zic. */ bool pre_2017c_match = false; for (lp = table; lp->l_word; lp++) if (itsabbr(word, lp->l_word)) { -- 2.21.0

1 0

Estimate for a new tzdb release with Brazil rules.
by Pramoth Murali May 14, 2019

May 14, 2019

Wanted to request a follow-up to an earlier comment https://mm.icann.org/pipermail/tz/2019-April/027877.html <https://nam01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmm.icann.…> . Specifically, “No rush for a new tzdb release (but of course we'll want something out before October :-)”. Do we have an estimate for when the new tzdb release might be available containing the Brazil rules? Is there any way I can request this be updated sooner than later? Understand there was a patch attached in an earlier thread. My company would ideally prefer to align with a more formal release cadence to get these changes. Thanks! -- Pramoth

6 8

Spelling of Ukrainian cities #KyivNotKiev
by Vitaliy Stavropolskiy May 12, 2019

May 12, 2019

Hi! Maybe it's time to change the spelling of Ukrainian cities <https://mfa.gov.ua/en/page/open/id/5418>? I've found a comment in the tzdb: > # Admittedly English-language spelling of Ukrainian names is > # controversial, and some day "Kyiv" may become substantially more popular > in > # English; in the meantime, stick with the traditional English "Kiev" as > that > # means less disruption for our users. I think it's the right time to change it. For example, a lot of Airports and international media did it. For example,The Guardian: https://twitter.com/shaunwalker7/status/1095687682582286337 https://www.theguardian.com/guardian-observer-style-guide-k Airports in London, Toronto, Istanbul, Bucharest, etc.: 1. https://twitter.com/GamanOleksandr/status/1108371955353305095 2. https://procol-harum.livejournal.com/1091764.html?utm_source=twsharing&utm_… 3. https://www.facebook.com/UkrEmbassyDK/posts/2189053684497098 You can find a lot of other examples by using hashtag #KyivNotKiev Here is the link to the official information from the Ministry of Foreign Affairs of Ukraine: https://mfa.gov.ua/en/page/open/id/5418 Please, consider. Best regards, Vitalii Stavropolskyi

10 10

Issue with EST timezone
by Chris Woodbury May 11, 2019

May 11, 2019

Would it be "letting the cat out of the bag" to say that the timezone he's probably looking for is EST5EDT?

1 0

Re: [tz] Issue with EST timezone
by Paul Eggert May 11, 2019

May 11, 2019

On 5/7/19 3:53 AM, Manjusri Mahendran -X (manjmahe - INFOSYS LIMITED at Cisco) wrote: > > We have issues on Call Manager running on EST timezone. > > Could see that there is no replacement added for deprecated timezones > EST, HST and MST whereas other deprecated timezones such as “Brazil > East” and “Singapore” has a replacement. > > How are these three timezones different from the rest of deprecated ones.? They aren't deprecated. EST, HST, and MST are still in tzdb, in the 'northamerica' file. They've been there since 2005. It sounds like you've run into an issue that is downstream from tzdb.

8 10

Europe/Rome, 22 Sep 1866 rule
by Stephen Trainor May 7, 2019

May 7, 2019

In looking into Europe/Rome, I see there is the following rule: # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Europe/Rome 0:49:56 - LMT 1866 Sep 22 (line 1660, europe, tzdb-2019a) In reading the page referenced in this 2016 thread on historical Italy time zones (https://mm.icann.org/pipermail/tz/2016-October/024363.html <https://mm.icann.org/pipermail/tz/2016-October/024363.html>), it states: "Regio Decreto del 22 Settembre 1866 n° 3224 di Re Vittorio Emanuele II ( in vigore dal 12 Dicembre 1866 ) Il servizio dei convogli nelle ferrovie, quello dei telegrafi, delle poste, delle messaggerie e dei piroscafi postali nelle Provincie Continentali del Regno d’Italia, verrà regolato col tempo medio di Roma a datare dal giorno in cui sarà attivato l’orario delle strade ferrate per la prossima stagione invernale 1866 – 1867.” (source: http://www.ac-ilsestante.it/MERIDIANE/ora_legale/ORA_LEGALE_ESTIVA_IN_ITALI… <http://www.ac-ilsestante.it/MERIDIANE/ora_legale/ORA_LEGALE_ESTIVA_IN_ITALI…>) This implies that the law went into effect on 12 December 1866, rather than the date of the decree (22 Sep 1866). This is further supported by this source: https://web.archive.org/web/20070824155341/http://www.iav.it/planetario/did… <https://web.archive.org/web/20070824155341/http://www.iav.it/planetario/did…>, which states: "In Italy in 1866 there were 6 railway times (Torino, Verona, Firenze, Roma, Napoli, Palermo). On that year it was decided to unify them, adopting the average time of Rome (even if this city was not yet part of the kingdom). On the 12th December 1866, on the starting of the winter time table, it took effect in the railways, the post office and the telegraph, not only for the internal service but also for the public. Moreover, without any particular law, but on a free initiative of the main Italian cities and as a consequence of the practical advantages of tying the railway time to the city time, and also for patriotic reasons ( Rome was not part of kingdom) it was decided to use the Rome time in public and private life producing thus a national time. Milano set the public watches on the Rome time on the same day (12th December 1866), Torino and Bologna on the 1st January 1867, Venezia the 1st May 1880 and the last city was Cagliari in 1886.” I wonder if the rule should be modified to 1866 Dec 12, instead of 1866 Sep 22? Stephen Trainor

3 2