- tz - lists.iana.org

Morocco to fall back on Oct. 25, 2015
by Sterling Udell Sept. 19, 2015

Sept. 19, 2015

It looks like the Morocco World News did get it wrong. Here's an official page (in Arabic) noting that the correct date remains Oct 25: http://www.mmsp.gov.ma/ar/actualites.aspx?id=904

2 1

New Abbreviation needed for Asia/Pyongyang
by Greg Slayden Sept. 17, 2015

Sept. 17, 2015

Hi, I noticed in the 2015f tz database that the new Asia/Pyongyang zone for North Korea (now UTC+8:30) has the same abbreviation as Asia/Seoul. Both are noted as KST. Should the abbreviation for Asia/Pyongyang by changed to NKST, perhaps? Pardon if this issue has already been discussed, or if a new abbreviation has already been decided. I am new to the forum, thanks for understanding. --Greg Slayden Microsoft Corporation

2 1

[PROPOSED PATCH] Declare timezone, daylight, altzone in private.h
by Paul Eggert Sept. 17, 2015

Sept. 17, 2015

* NEWS: Document this. * private.h (timezone, daylight) [USG_COMPAT]: (altzone) [ALTZONE]: Declare as extern if not defined as a macro. This may avoid problems in environments where these variables are not declared by the system <time.h>. >From a suggestion by Kees Dekker in: http://mm.icann.org/pipermail/tz/2015-August/022627.html --- NEWS | 5 ++++- private.h | 13 +++++++++++++ 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/NEWS b/NEWS index 37f461f..2bbb967 100644 --- a/NEWS +++ b/NEWS @@ -13,10 +13,13 @@ Unreleased, experimental changes (Thanks to Bradley White for reporting the bug.) The localtime module allows the variables 'timezone', 'daylight', - and 'altzone' to be in common storage shared with other modules. + and 'altzone' to be in common storage shared with other modules, + and declares them in case the system <time.h> does not. + (Problems reported by Kees Dekker.) On platforms with tm_zone, strftime.c now assumes it is not NULL. This simplifies the code and is consistent with zdump.c. + (Problem reported by Christos Zoulas.) Changes affecting documentation diff --git a/private.h b/private.h index 184da8f..1c176e6 100644 --- a/private.h +++ b/private.h @@ -383,12 +383,25 @@ void tzset(void); ** Some time.h implementations don't declare asctime_r. ** Others might define it as a macro. ** Fix the former without affecting the latter. +** Similarly for timezone, daylight, and altzone. */ #ifndef asctime_r extern char * asctime_r(struct tm const *restrict, char *restrict); #endif +#ifdef USG_COMPAT +# ifndef timezone +extern long timezone; +# endif +# ifndef daylight +extern int daylight; +# endif +#endif +#if defined ALTZONE && !defined altzone +extern long altzone; +#endif + /* ** The STD_INSPIRED functions are similar, but most also need ** declarations if time_tz is defined. -- 2.1.4

2 1

[PROPOSED PATCH] * asia (Azer): Update URL for 1997-03-17 Azeri resolution.
by Paul Eggert Sept. 17, 2015

Sept. 17, 2015

--- asia | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/asia b/asia index 78eed56..5467024 100644 --- a/asia +++ b/asia @@ -131,7 +131,8 @@ Zone Asia/Yerevan 2:58:00 - LMT 1924 May 2 # Azerbaijan # From Rustam Aliyev of the Azerbaijan Internet Forum (2005-10-23): # According to the resolution of Cabinet of Ministers, 1997 -# Resolution available at: http://aif.az/docs/daylight_res.pdf +# From Paul Eggert (2015-09-17): It was Resolution No. 21 (1997-03-17). +# http://code.az/files/daylight_res.pdf # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S Rule Azer 1997 max - Mar lastSun 4:00 1:00 S Rule Azer 1997 max - Oct lastSun 5:00 0 - -- 2.1.4

1 0

Duplicate Links
by Daniel Schultze Sept. 15, 2015

Sept. 15, 2015

All, While working on a project related to timezone conversion I found that there are a couple of Links that have multiple mappings. I found the problem in tzdata2015f. I have attatched a patch to the checklinks.awk file which lists links with more than one mapping. If I have misunderstood Links please advise me. Thank you, Daniel Schultze

2 2

[PROPOSED PATCH] 'make check_links' now checks duplicate links etc.
by Paul Eggert Sept. 14, 2015

Sept. 14, 2015

* checklinks.awk: Check for duplicate links and zones. Prompted by a problem report from Daniel Schultze in: http://mm.icann.org/pipermail/tz/2015-September/022650.html --- checklinks.awk | 39 ++++++++++++++++++++++++++++++++++----- 1 file changed, 34 insertions(+), 5 deletions(-) diff --git a/checklinks.awk b/checklinks.awk index f57f736..f0ff726 100644 --- a/checklinks.awk +++ b/checklinks.awk @@ -2,14 +2,43 @@ # Contributed by Paul Eggert. -/^Link/ { used[$2] = 1 } -/^Zone/ { defined[$2] = 1 } +BEGIN { + # Special marker indicating that the name is defined as a Zone. + # It is a newline so that it cannot match a valid name. + # It is not null so that its slot does not appear unset. + Zone = "\n" +} -END { - status = 0 +/^Zone/ { + if (defined[$2]) { + if (defined[$2] == Zone) { + printf "%s: Zone has duplicate definition\n", $2 + } else { + printf "%s: Link with same name as Zone\n", $2 + } + status = 1 + } + defined[$2] = Zone +} +/^Link/ { + if (defined[$3]) { + if (defined[$3] == Zone) { + printf "%s: Link with same name as Zone\n", $3 + } else if (defined[$3] == $2) { + printf "%s: Link has duplicate definition\n", $3 + } else { + printf "%s: Link to both %s and %s\n", $3, defined[$3], $2 + } + status = 1 + } + used[$2] = 1 + defined[$3] = $2 +} + +END { for (tz in used) { - if (!defined[tz]) { + if (defined[tz] != Zone) { printf "%s: Link to non-zone\n", tz status = 1 } -- 2.1.4

1 0

[PROPOSED PATCH] Remove synopsis from tzfile man page
by Paul Eggert Sept. 14, 2015

Sept. 14, 2015

* tzfile.5: Remove SYNOPSIS section, as tzfile.h is not intended to be public and should not be copied to /usr/include. Problem reported by kaduk in: https://github.com/eggert/tz/pull/13 --- tzfile.5 | 3 --- 1 file changed, 3 deletions(-) diff --git a/tzfile.5 b/tzfile.5 index 0f2b2ea..b7e6d45 100644 --- a/tzfile.5 +++ b/tzfile.5 @@ -1,9 +1,6 @@ .TH TZFILE 5 .SH NAME tzfile \- time zone information -.SH SYNOPSIS -.B -#include <tzfile.h> .SH DESCRIPTION The time zone information files used by .IR tzset (3) -- 2.1.0

1 0

South Australia may move clocks 30 minutes
by Paul Eggert Sept. 10, 2015

Sept. 10, 2015

The South Australian government is considering changing their clocks by 30 minutes. The government would prefer advancing clocks by 30 minutes to match the time on the east coast. The public is split. Much of the resistance is coming from the west part of the state, where locals worry about their children leaving for school in the dark, and the government is prepared to consider changing school hours to address this. Another possibility is retarding clocks by 30 minutes; this would move closer to solar time, and would align better with Asia. This change would probably affect Broken Hill as well. Northern Territory will be consulted but seems less inclined to change. The consultation process will take at least two months. The government's website crashed as citizens went online to give their opinion. Sources: Nicholson M. South Australia has the most ridiculous time zone, and might be about to change it. Junkee 2015-02-06. http://junkee.com/south-australia-has-the-most-ridiculous-time-zone-and-mig… Ball G, Clift J, Coote G. Broken Hill urged to air views on proposed time zone change for South Australia. ABC News (Australia), 2015-02-06. http://www.abc.net.au/news/2015-02-06/broken-hill-community-urged-air-views… SA considers changing time zone to match eastern states. 9NEWS 2015-02-05. http://www.9news.com.au/national/2015/02/05/12/15/sa-considers-changing-tim… South Australians remain divided on whether it is time for the time to change. 9NEWS 2015-02-06. http://www.9news.com.au/national/2015/02/06/21/03/south-australians-remain-…

2 3

Elsie.nci.nih.gov--a bituary.
by Arthur David Olson Sept. 10, 2015

Sept. 10, 2015

Elsie.nci.nih.gov, best known as the long-time hostess of the time zone database, reached end of life on 2015-09-10; Elsie was 34 (or 457 in human years). Around the time of Elsie's 1981 arrival, the Laboratory of Carcinogen Metabolism at the National Cancer Institute (NCI) in Bethesda, Maryland, was renamed the Laboratory of Experimental Carcinogenesis; Elsie was named for the initials common to the old and new lab names. Elsie began life as a Digital Equipment VAX running BSD UNIX, reincarnating once as a Sun SPARC running Solaris and then again (in declining years) as a Dell Optiplex running Windows. Elsie's early career was in image processing; a software package for analysis of images of two-dimensional polyacrylamide gels bore Elsie's name. As the first computer in the lab, Elsie was the family head of (and file server for) a herd of several dozen computers, all named after cows; they most notably included Daisy, Flower, Millie (adopted from Australia) and Abunga (adopted from Hawaiʻi). Elsie was an early participant in the dial-up UUCP network, net news, and the Internet, as well as a prototype HTTP server within NCI. After an unprovoked 2011 attack, Elsie withdrew behind a wall, but remained active in that more limited sphere. The lab having been disbanded, there are no survivors; since there is no longer a server, there will be no service. @dashdashado

2 1

Binary File Format
by Bill Seymour Sept. 10, 2015

Sept. 10, 2015

I have three questions about the format of the binary files. 1. Re the array of tzh_timecnt transitions: my kneejerk reaction was to think that these were similar to the values in the UNTIL column in the source files, in which case they'd be open upper bounds (in other words, the struct ttinfo that they eventually link to would be for the period immediately before the transition); but it occurs to me that they might be closed lower bounds instead (the corresponding ttinfo is for the period after the transition). tzfile(5) doesn't say explicitly. 2. Basically the same question for the array of leap seconds. The reason it occurs to me that the time_t value could be the point after the leap second is that that's the way that the USNO’s list of leap seconds works. 3. Re the version-'2' array of leap seconds: each array element is a time_t value followed by a count of the number of leap seconds after that time. tzfile(5) says that eight bytes are used for each "transition time" and "leap second time"; but since the count of leap seconds is neither of those, is the count still just four bytes? (I'm thinking that I can't just have a two-dimensional array of int64_t, but rather need a one-dimensional array of struct { int64_t, int32_t }. Is that right?) Thanks, --Bill Seymour

2 1