- tz - lists.iana.org

[PROPOSED PATCH] Distribute .tzs file for regression testing
by Paul Eggert Aug. 22, 2016

Aug. 22, 2016

* NEWS: Document this. * Makefile (TZS_YEAR, TZS, TZS_NEW, TZS_DEPS): New macros. (ENCHILADA): Add TZS. ($(TZS_NEW), $(TZS), force_tzs, check_tzs): New targets. (check): Add check_tzs. (clean): Remove *.dir. (maintainer-clean): Remove $(TZS). (set-timestamps.out): Set $(TZS)'s time stamp. (check_public, check_time_t_alternatives): Use a temporary directory name ending in .dir, to make it easier to clean these directories out systematically. (tzdata$(VERSION).tar.gz): Add $(TZS). (.PHONY): Add force_tzs, check_tzs. --- Makefile | 71 ++++++++++++++++++++++++++++++++++++++++++++++------------------ NEWS | 12 +++++++---- 2 files changed, 59 insertions(+), 24 deletions(-) diff --git a/Makefile b/Makefile index bbe1ba8..fc18819 100644 --- a/Makefile +++ b/Makefile @@ -378,7 +378,12 @@ DATA= $(YDATA) $(NDATA) backzone $(TABDATA) \ leap-seconds.list yearistype.sh AWK_SCRIPTS= checklinks.awk checktab.awk leapseconds.awk MISC= $(AWK_SCRIPTS) zoneinfo2tdf.pl -ENCHILADA= $(COMMON) $(DOCS) $(SOURCES) $(DATA) $(MISC) +TZS_YEAR= 2050 +TZS= to$(TZS_YEAR).tzs +TZS_NEW= to$(TZS_YEAR)new.tzs +TZS_DEPS= $(PRIMARY_YDATA) asctime.c localtime.c \ + private.h tzfile.h zdump.c zic.c +ENCHILADA= $(COMMON) $(DOCS) $(SOURCES) $(DATA) $(MISC) $(TZS) # And for the benefit of csh users on systems that assume the user # shell should be used to handle commands in Makefiles. . . @@ -477,6 +482,27 @@ posix_packrat: zones: $(REDO) +$(TZS_NEW): $(TDATA) zdump zic + mkdir -p tzs.dir + $(zic) -d tzs.dir $(TDATA) + $(AWK) '/^Link/{print $$1 "\t" $$2 "\t" $$3}' \ + $(TDATA) | LC_ALL=C sort >$@.out + zones=$$($(AWK) -v wd="$$(pwd)" \ + '/^Zone/{print wd "/tzs.dir/" $$2}' $(TDATA) \ + | LC_ALL=C sort) && \ + ./zdump -i -c $(TZS_YEAR) $$zones >>$@.out + sed 's,^TZ=".*tzs\.dir/,TZ=",' $@.out >$@ + rm -fr tzs.dir $@.out + +# If $(TZS) does not already exist (e.g., old-format tarballs), create it. +# If it exists but 'make check_tzs' fails, a maintainer should inspect the +# failed output and fix the inconsistency, perhaps by running 'make force_tzs'. +$(TZS): + $(MAKE) force_tzs + +force_tzs: $(TZS_NEW) + cp $(TZS_NEW) $(TZS) + libtz.a: $(LIBOBJS) $(AR) ru $@ $(LIBOBJS) $(RANLIB) $@ @@ -496,7 +522,7 @@ tzselect: tzselect.ksh chmod +x $@ check: check_character_set check_white_space check_links check_sorted \ - check_tables check_web + check_tables check_tzs check_web check_character_set: $(ENCHILADA) LC_ALL=en_US.utf8 && export LC_ALL && \ @@ -532,6 +558,9 @@ check_tables: checktab.awk $(PRIMARY_YDATA) $(ZONETABLES) || exit; \ done +check_tzs: $(TZS) $(TZS_NEW) + diff -u $(TZS) $(TZS_NEW) + check_web: $(WEB_PAGES) $(VALIDATE_ENV) $(VALIDATE) $(VALIDATE_FLAGS) $(WEB_PAGES) @@ -539,12 +568,12 @@ clean_misc: rm -f core *.o *.out \ date tzselect version.h zdump zic yearistype libtz.a clean: clean_misc - rm -fr tzpublic + rm -fr *.dir $(TZS_NEW) maintainer-clean: clean @echo 'This command is intended for maintainers to use; it' @echo 'deletes files that may need special tools to rebuild.' - rm -f leapseconds $(MANTXTS) *.asc *.tar.gz + rm -f leapseconds $(MANTXTS) $(TZS) *.asc *.tar.gz names: @echo $(ENCHILADA) @@ -591,6 +620,7 @@ set-timestamps.out: $(ENCHILADA) touch -cmr `ls -t $$file workman.sh | sed 1q` $$file.txt || \ exit; \ done + touch -cmr $$(ls -t $(TZS_DEPS) | sed 1q) $(TZS) touch $@ # The zics below ensure that each data file can stand on its own. @@ -599,12 +629,12 @@ set-timestamps.out: $(ENCHILADA) check_public: $(MAKE) maintainer-clean $(MAKE) "CFLAGS=$(GCC_DEBUG_FLAGS)" ALL - mkdir tzpublic + mkdir -p public.dir for i in $(TDATA) ; do \ - $(zic) -v -d tzpublic $$i 2>&1 || exit; \ + $(zic) -v -d public.dir $$i 2>&1 || exit; \ done - $(zic) -v -d tzpublic $(TDATA) - rm -fr tzpublic + $(zic) -v -d public.dir $(TDATA) + rm -fr public.dir # Check that the code works under various alternative # implementations of time_t. @@ -616,15 +646,15 @@ check_time_t_alternatives: fi && \ zones=`$(AWK) '/^[^#]/ { print $$3 }' <zone1970.tab` && \ for type in $(TIME_T_ALTERNATIVES); do \ - mkdir -p tzpublic/$$type && \ + mkdir -p time_t.dir/$$type && \ $(MAKE) clean_misc && \ - $(MAKE) TOPDIR=`pwd`/tzpublic/$$type \ + $(MAKE) TOPDIR=$$(pwd)/time_t.dir/$$type \ CFLAGS='$(CFLAGS) -Dtime_tz='"'$$type'" \ REDO='$(REDO)' \ install && \ diff $$quiet_option -r \ - tzpublic/int64_t/etc/zoneinfo \ - tzpublic/$$type/etc/zoneinfo && \ + time_t.dir/int64_t/etc/zoneinfo \ + time_t.dir/$$type/etc/zoneinfo && \ case $$type in \ int32_t) range=-2147483648,2147483647;; \ uint32_t) range=0,4294967296;; \ @@ -633,14 +663,14 @@ check_time_t_alternatives: *) range=-10000000000,10000000000;; \ esac && \ echo checking $$type zones ... && \ - tzpublic/int64_t/etc/zdump -V -t $$range $$zones \ - >tzpublic/int64_t.out && \ - tzpublic/$$type/etc/zdump -V -t $$range $$zones \ - >tzpublic/$$type.out && \ - diff -u tzpublic/int64_t.out tzpublic/$$type.out \ + time_t.dir/int64_t/etc/zdump -V -t $$range $$zones \ + >time_t.dir/int64_t.out && \ + time_t.dir/$$type/etc/zdump -V -t $$range $$zones \ + >time_t.dir/$$type.out && \ + diff -u time_t.dir/int64_t.out time_t.dir/$$type.out \ || exit; \ done - rm -fr tzpublic + rm -fr time_t.dir tarballs: tzcode$(VERSION).tar.gz tzdata$(VERSION).tar.gz @@ -652,7 +682,7 @@ tzcode$(VERSION).tar.gz: set-timestamps.out tzdata$(VERSION).tar.gz: set-timestamps.out LC_ALL=C && export LC_ALL && \ - tar $(TARFLAGS) -cf - $(COMMON) $(DATA) $(MISC) | \ + tar $(TARFLAGS) -cf - $(COMMON) $(DATA) $(MISC) $(TZS) | \ gzip $(GZIPFLAGS) > $@ signatures: tzcode$(VERSION).tar.gz.asc tzdata$(VERSION).tar.gz.asc @@ -688,7 +718,8 @@ zic.o: private.h tzfile.h version.h .PHONY: ALL INSTALL all .PHONY: check check_character_set check_links .PHONY: check_public check_sorted check_tables -.PHONY: check_time_t_alternatives check_web check_white_space clean clean_misc +.PHONY: check_time_t_alternatives check_tzs check_web check_white_space +.PHONY: clean clean_misc force_tzs .PHONY: install install_data maintainer-clean names .PHONY: posix_only posix_packrat posix_right .PHONY: public right_only right_posix signatures tarballs typecheck diff --git a/NEWS b/NEWS index 3d01581..023883b 100644 --- a/NEWS +++ b/NEWS @@ -2,18 +2,18 @@ News for the tz database Unreleased, experimental changes - Changes affecting future time stamps + Changes to future time stamps New leap second 2016-12-31 23:59:60 UTC as per IERS Bulletin C 52. (Thanks to Tim Parenti.) - Changes affecting past time stamps + 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. - Changes affecting code + Changes to code zic no longer generates binary files containing POSIX TZ-like strings that disagree with the local time type after the last @@ -28,7 +28,11 @@ Unreleased, experimental changes (Thanks to Jon Skeet for suggesting that an option was needed, and thanks to Tim Parenti and Chris Rovick for further comments.) - Changes affecting documentation and commentary + The tzdata distribution has a new file to2050.tzs that contains + what should be the output of 'zdump -i -c 2050' on primary zones. + 'make check' now checks that zdump generates this output. + + Changes to documentation and commentary tzfile.5 now documents the new restriction on POSIX TZ-like strings that is now implemented by zic. -- 2.5.5

1 0

[PROPOSED PATCH] Fix 1948-1966 transitions for Los Angeles
by Paul Eggert Aug. 20, 2016

Aug. 20, 2016

These changes came from research I did to check the historical background behind Assembly Bill 385, which would abolish daylight saving time in California, and which has passed the CA Senate Appropriations Committee's suspense file. See: White JB. Bill letting California end Daylight Saving Time clears key committee. Sacramento Bee 2016-08-11. http://www.sacbee.com/news/politics-government/capitol-alert/article9511673… * NEWS, northamerica (CA): For America/Los_Angeles, correct spring-forward transition times from 02:00 to 02:01 in 1948, from 02:00 to 01:00 in 1950-1966. Correct the fall-back transition date for early 1949 from 1949-01-01 to 1949-01-16. --- NEWS | 7 +++++++ northamerica | 35 ++++++++++++++++++++++++++++++++--- 2 files changed, 39 insertions(+), 3 deletions(-) diff --git a/NEWS b/NEWS index ba26a49..a8479eb 100644 --- a/NEWS +++ b/NEWS @@ -7,6 +7,13 @@ Unreleased, experimental changes New leap second 2016-12-31 23:59:60 UTC as per IERS Bulletin C 52. (Thanks to Tim Parenti.) + Changes affecting 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. The fall-back transition date for early 1949 has been + corrected from 1949-01-01 to 1949-01-16. + Changes affecting code zic no longer generates binary files containing POSIX TZ-like diff --git a/northamerica b/northamerica index 6256f97..c9c2061 100644 --- a/northamerica +++ b/northamerica @@ -413,11 +413,40 @@ Zone America/Denver -6:59:56 - LMT 1883 Nov 18 12:00:04 # north of the Salmon River, and the towns of Burgdorf and Warren), # Nevada (except West Wendover), Oregon (except the northern 3/4 of # Malheur county), and Washington + +# From Paul Eggert (2016-08-19): +# In 1948 California had an electricity shortage. To save electricity, +# in early February PG&E changed power frequency from 60 to 59.5 Hz during +# daylight hours, causing electric clocks to lose six minutes per day. +# (This did not change legal time, and is not part of the data here.) +# See: Ross SA. An energy crisis from the past: Northern California in 1948. +# Working Paper No. 8, Institute of Governmental Studies, UC Berkeley, 1973-11. +# http://escholarship.org/uc/item/8x22k30c +# +# In another measure to save electricity, DST was instituted from 1948-03-14 +# at 02:01 to 1949-01-16 at 02:00, with the governor having the option to move +# the fallback transition earlier. See pages 3-4 of: +# http://clerk.assembly.ca.gov/sites/clerk.assembly.ca.gov/files/archive/Stat… +# Apparently the governor did not exercise the option. +# +# In response: +# +# Governor Warren received a torrent of objecting mail, and it is not too much +# to speculate that the objections to Daylight Saving Time were one important +# factor in the defeat of the Dewey-Warren Presidential ticket in California. +# -- Ross, p 25 +# +# Despite the controversy, in 1949 California voters approved Proposition 12, +# which established DST from April's last Sunday at 01:00 until September's +# last Sunday at 02:00. This was amended by 1962's Proposition 6, which changed +# the fall-back date to October's last Sunday. See: +# http://repository.uchastings.edu/cgi/viewcontent.cgi?article=1501&context=c… +# http://repository.uchastings.edu/cgi/viewcontent.cgi?article=1636&context=c… # # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER -Rule CA 1948 only - Mar 14 2:00 1:00 D -Rule CA 1949 only - Jan 1 2:00 0 S -Rule CA 1950 1966 - Apr lastSun 2:00 1:00 D +Rule CA 1948 only - Mar 14 2:01 1:00 D +Rule CA 1949 only - Jan 16 2:00 0 S +Rule CA 1950 1966 - Apr lastSun 1:00 1:00 D Rule CA 1950 1961 - Sep lastSun 2:00 0 S Rule CA 1962 1966 - Oct lastSun 2:00 0 S # Zone NAME GMTOFF RULES FORMAT [UNTIL] -- 2.7.4

3 4

Use of <> delimiters in TZ string is not documented
by Alexander Belopolsky Aug. 19, 2016

Aug. 19, 2016

The Asia/Almaty tzfile generated from recent Github tzdata snapshot contains the following TZ string at the end of the file $ git describe 2016f-7-ge0a7616 $ strings etc/zoneinfo/Asia/Almaty | tail -1 <+06>-6 This format is not documented in newtzset.3: When TZ is used directly as a specification of the time conversion information, it must have the following syntax (spaces inserted for clarity): stdoffset[dst[offset][,rule]] Where: std and dst Three or more bytes that are the designation for the standard (std) or summer (dst) time zone. Only std is required; if dst is missing, then summer time does not apply in this locale. Upper- and lowercase letters are explicitly allowed. Any characters except a leading colon (:), digits, comma (,), ASCII minus (-), ASCII plus (+), and NUL bytes are allowed.

3 3

Egypt DST Abolition Decree
by Matt Johnson Aug. 10, 2016

Aug. 10, 2016

For those interested, I've finally been able to locate the published decree associated with Egypt's abolition of DST. Published in the Official Gazette #29, Year 2016, Legislation # 62. Date of publication 2016-07-24. http://www.cc.gov.eg/Images/L/369859.pdf I believe it simply reflects the permanent cancellation of DST, but a more direct translation or summary would be appreciated. (Translation tools are failing due to the nature of how the Arabic text is embedded in the PDF.) Thanks, Matt

4 3

Kuwait considers DST
by Matt Johnson Aug. 9, 2016

Aug. 9, 2016

Apparently Kuwait is considering introducing DST next year. It has never done so before. I can't find a more official reference at the moment, but worth keeping an eye on: https://twitter.com/jahrawie/status/762362762760744961 Any Arabic native speakers able to read the text in the images? If so, a summary would be appreciated. Thanks, Matt

2 3

New leap second
by Meno Hochschild Aug. 8, 2016

Aug. 8, 2016

Dear users of tzdb, IERS has announced a new leap second at the end of year 2016, see https://hpiers.obspm.fr/iers/bul/bulc/bulletinc.dat Another point: I have myself written a time library in Java called Time4J which contains a tz compiler based on your data and which also evaluates the leap second file contained in tz-data. Hence this library could be listed in the file "tz-link" within the section "Other tz compilers" maybe with this description: <a href="https://github.com/MenoData/Time4J/">Time4J – Advanced date, time and interval library for Java</a> , uses the class <code>net.time4j.tool.TimezoneRepositoryCompiler.java</code> located in TZTool-module that compiles <code><abbr>tz</abbr></code> source into a Time4J-specific binary format. Time4J is available under a LGPLv2.1-licence. Cheers and thanks for your very useful work Meno

3 3

POSIX-TZ-environment-variable-style string rules
by Alexander Belopolsky Aug. 5, 2016

Aug. 5, 2016

The tzfile.5 manual page describes the TZ string found at the end of the version 2+ tzfiles as follows: "After the second header and data comes a newline-enclosed, POSIX-TZ-environment-variable-style string for use in handling instants after the last transition time stored in the file." Consider the case when the last transition time (T) stored in the file comes a few weeks before the first POSIX rule transition P > T. What local time type should be used for the the times after T, but before P? The strict reading of the manual page suggests that one should use the the local time type derived from the POSIX rule for all times t >= T, but it would be more reasonable to continue using the type that corresponds to T for times T <= t < P. It turns out the two alternative logics are implemented in real life. The glibc library switches [1] to the POSIX rule when t >= P, while the tzcode implementation uses POSIX rule to supply additional transitions with the first POSIX-derived transition at P. This leads to a discrepancy between glibc and tzcode results that is best described on the Ubuntu bug tracker (bug #1587128 [2]). While tzcode behavior is more logical, glibc code seems to better match the documentation and it will probably be an uphill battle to get a fix into glibc. [3] I think a better solution would be to make zic add a redundant transition when the local time type corresponding to the last transition does not match the POSIX rule. This will ensure that tzcode and glibc produces the same results and systems affected by this issue can be fixed by simply upgrading the zoneinfo database. [1]: https://sourceware.org/git/?p=glibc.git;a=blob;f=time/tzfile.c;h=9049878399… [2]: https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/1587128 [3]: https://sourceware.org/ml/libc-help/2016-04/msg00000.html

2 4

Morocco DST change happened on 4/June/2016 instead of 5/June/2016
by Gopal . Aug. 4, 2016

Aug. 4, 2016

Hi Team, TZupdater updated time zone 1 day before scheduled time, it was supposed to change time on 5 June 2016 for Morocco Casablanca and but application time changed on 4 June 2016. Could you please let us know as we are facing some problems in application due to this. Regards Gopal This e-mail and all material transmitted with it are for the use of the intended recipient(s) ONLY and contains confidential and/or privileged information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies and the original message. Any unauthorized review, use, disclosure, dissemination, forwarding, printing or copying of this email or any action taken pursuant to the contents of the present e-mail is strictly prohibited and is unlawful. The recipient acknowledges that Comviva Technologies Limited or its management or directors, are unable to exercise control or ensure the integrity over /of the contents of the information contained in e-mail. Any views expressed herein are those of the individual sender only and no binding nature of the contents shall be implied or assumed unless the sender does so expressly with due authority of Comviva Technologies Limited. E-mail and any contents transmitted with it are prone to viruses and related defects despite all efforts to avoid such by Comviva Technologies Limited.

4 6

Regarding IANA update for Novosibirsk Time Zone
by Shruthi Venkataramanan -X (shruthve - WIPRO LIMITED at Cisco) Aug. 2, 2016

Aug. 2, 2016

Hi Iana Team, We have already requested for an update regarding Novosibirsk Time Zone , please find the email chain attached. Currently we want to know, when can you give us an updated Version, so that we can do the changes in our software. Thanks & Regards, Shruthi.

2 1

why does mktime binary search?
by enh July 29, 2016

July 29, 2016

is there a more detailed explanation of why mktime does binary search? ** Adapted from code provided by Robert Elz, who writes: ** The "best" way to do mktime I think is based on an idea of Bob ** Kridle's (so its said...) from a long time ago. ** It does a binary search of the time_t space. Since time_t's are ** just 32 bits, its a max of 32 iterations (even at 64 bits it ** would still be very reasonable). this came up on Android because we had some code that did "errno = 0; mktime(...); if (errno == EOVERFLOW) ..." rather than checking the return value of mktime. this failed for us because the tzcode mktime sets errno even on non-failure. obviously i can fix this by saving errno around the call to mktime_tzname in mktime, but i'm curious to know why there's a binary search at all. i also notice that, of the BSDs, NetBSD seems to have fixed this locally, FreeBSD has an old enough copy of tzcode that it's not setting errno at all, and OpenBSD is in the same boat as Android. here's the NetBSD change, which is does the errno save-restore in time1 instead: revision 1.90 date: 2014-10-16 10:53:32 -0700; author: christos; state: Exp; lines: +24 -36; commitid: 4woGqYWkKe0i6sUx; - don't leak errno in mktime() - when we load a new timezone, don't change anything unless the load succeeded. http://cvsweb.netbsd.org/bsdweb.cgi/src/lib/libc/time/localtime.c.diff?r1=1…

4 3