- tz - lists.iana.org

Android (bionic) tzcode changes
by enh Aug. 9, 2012

Aug. 9, 2012

i've been maintaining Android's copy of the tzdata for several years now and have more recently inherited maintenance of Android's C library, bionic. i was looking at automating updates of tzcode (which hasn't been tracking upstream at all) but noticed that we have several extensions that i can't see upstream (in tzcode2012e): 1. Android concatenates all the data files into one big file that's indexed by a separate .idx file. the idea being that you binary search the index (by Olson id) to find the offset into the data of that zone's data. 2. Android implements the glibc strftime extensions (search for "an optional flag" in http://www.gnu.org/software/libc/manual/html_node/Formatting-Calendar-Time.… ). 3. Android implements strptime ( http://pubs.opengroup.org/onlinepubs/007904975/functions/strptime.html) including the glibc extensions. i might have to clean some of these up, but i was curious whether upstream would at least in theory be interested in any of this? i can see 1 being least interesting, so i might just have to factor our changes there out into other files as best i can. but 2 and 3 sound like they'd be of general interest? (also feel free to add Android to the web page's list of systems that rely on the tz database.) --elliott

1 0

Detecting a new release
by Neil Masson Aug. 3, 2012

Aug. 3, 2012

This might have been talked about recently, but I don't remember any resolution. I have a script which polls the IANA Website daily to detect if there has been any update. It parses the html to find the latest version and then checks to see if the version has changed since the last time. This solution is fragile and could easily be broken if the layout of the Website changes. I don't see any way of detecting the version from the ftp or rsync sites as the "real" filename which contains the version has been replaced with a link tzdata-latest.tar.gz. It would be better if a version file was added to the ftp and rsync sites which contained the version identifier of the release. Neil Masson Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

6 7

Re: [tz] timegm() function and state
by Paul Eggert Aug. 3, 2012

Aug. 3, 2012

I tried to reproduce that configuration under Ubuntu but still did not see the problem. Here's how I built it: make CFLAGS='-DNOID -DUSG_COMPAT -DSTD_INSPIRED -DPCTS -DTM_GMTOFF=tm_gmtoff -DTM_ZONE=tm_zone -DTZDEFRULESTRING=\",M3.2.0/2,M11.1.0/2\"' TOPDIR=/tmp/top install gcc -I. test.c /tmp/top/lib/libtz.a and here's my test program: #include <string.h> #include <errno.h> #include <stdio.h> #include <time.h> int main(int argc, char *argv[]) { struct tm localt; localt.tm_sec = 0; localt.tm_min = 0; localt.tm_hour = 0; localt.tm_mday = 1; localt.tm_year = 113; localt.tm_mon = 0; localt.tm_wday = 0; localt.tm_yday = 0; localt.tm_isdst = 0; localt.tm_gmtoff = 0; localt.tm_zone = 0; mktime(&localt); printf("asctime: %s", asctime(&localt)); printf("timegm(&localt) = %ld (errno %s)\n", timegm(&localt), strerror(errno)); printf("timegm(&localt) = %ld (errno %s)\n", timegm(&localt), strerror(errno)); printf("timegm(&localt) = %ld (errno %s)\n", timegm(&localt), strerror(errno)); return 0; } and its output: asctime: Tue Jan 1 00:00:00 2013 timegm(&localt) = 1356998400 (errno Success) timegm(&localt) = 1356998400 (errno Success) timegm(&localt) = 1356998400 (errno Success)

1 0

Re: [tz] Detecting a new release
by Neil Masson Aug. 3, 2012

Aug. 3, 2012

> You could parse the output of lftp or ncftpls in long-listing mode: Or indeed use rsync -l $ rsync -l rsync://rsync.iana.org/tz/tzdata-latest.tar.gz lrwxrwxrwx 27 2012/08/03 05:28:07 tzdata-latest.tar.gz -> releases/tzdata2012e.tar.gz Thanks for your help Neil Masson Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

1 0

tzcode2012e and tzdata2012e available
by Paul Eggert Aug. 3, 2012

Aug. 3, 2012

The 2012e release of the tz code and data is available. It reflects the following changes circulated last week on the tz mailing list: * australasia (Pacific/Fakaofo): Tokelau is UTC+13, not UTC+14. (Thanks to Steffen Thorsen.) * Use a single version number for both code and data. * .gitignore: New file. * Remove trailing white space. Here are links to the release files: http://www.iana.org/time-zones/repository/releases/tzcode2012e.tar.gz http://www.iana.org/time-zones/repository/releases/tzdata2012e.tar.gz ftp://ftp.iana.org/tz/releases/tzcode2012e.tar.gz ftp://ftp.iana.org/tz/releases/tzdata2012e.tar.gz As usual, links to the latest release files are here: http://www.iana.org/time-zones/repository/tzcode-latest.tar.gz http://www.iana.org/time-zones/repository/tzdata-latest.tar.gz ftp://ftp.iana.org/tz/tzcode-latest.tar.gz ftp://ftp.iana.org/tz/tzdata-latest.tar.gz This release corresponds to commit 60e612c3f451846ce24437e00b29974ab195b410 in the unofficial github repository at <https://github.com/eggert/tz>. Here are the GPG checksums for the release files: -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iQIcBAABAgAGBQJQG1MZAAoJEO2X6Q5iqn40u8kP/0XwaAAhJw/G6rdHD3rhFLcV FXgPJ5OmYFffYc4MxO83OZnFrfl+Wt+qP3C8XD5ggOIZ9NSpg245MQL5kNa7jH20 Wa9Gc26E7QycGjuFlTe3+1wfO2UH16A3M80cRgPGKxsziKSTy4gnbrXHWgdL68gP 5uJpW7zlQsgQQUHZcp+HObq556vhdOnZ/frS+kPWPYq2ZCJhjaJLUGrOLzgVsJfI HXIC50RQpiR9RorNWMEnQTaaSdGrzZdPICyQYesfs5AbC56G86tV4MOQC/9jpkMs qY+seV4PmEdGtvO7baUXDsUkXtxF8WGbPaz9w6ceduwoGNXC/WqaXyfiykmkTeC5 Pq0j4Ru7d4JSLWz7xgZUJ1HM2MedojOWUxZk5eurONY0HNsfq1q4Oc/L70DtHTSu HuewfXAkeU0Dr2zg/JW2FKaQrv+xK2jPyA+Je+jkgvuWTCj56S/KO/V2F8FqzKi/ e9mDmw1Cm/5GCzKWw5sr3aoviI0wcZCSnykQ+VMcq23f/2lNvcBzDjjYA5AKua5V wOwDIkk9aYNOphH/X/z5pT6U6hk3WmaIdfowOTIvEKhQK4UndByBsZiAo6hpOEIt ZSxJ3gR1f/E/3M1KHzL6Mjko0KWr01nSVc7/zsLODlihYPsFTbSdMQqVJKLRnd9e 5xmk63NXrodCs4gXvr5B =Bz0X -----END PGP SIGNATURE----- -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iQIcBAABAgAGBQJQG1MpAAoJEO2X6Q5iqn406o4P+wQ3JEmjTMhTxpVcCxpeSkko zDJnVvIE2LZoz7voCiPj6QmnEvty2yasI5EkLjnWTlmhFBISvbPDzceMMM0X01XE vEiHjeZGCHlEHgvo2wishyBzsCr7d8eoNjtkBng4DFyOc42zLjARJ8Z2elGji8JF bdoeAtYM6hFXwyPkKJ/0QrnN56q4jWFSYQxdoyx/HMhATtUeGOmubfQpWPTMswlJ ng8UWoqr6amULT0IXhlVZjyemJCY+JH3BJ4cDKYNHrllNORNnftFORvSpsNhl0dI TFRqGSsbb8CZDaXNVZ5OE/3loXWatPl+aac6HEXhIgF/tfy6uoupFmAtoHAUCByq v7ajL58iXx2HhAXGhlaBHlMVbcBRHQlPuTaw4vuDQ5G5leS5y+7Sm8bIt8Du0NHI uNx3wRKFPVzXIkAy/FSWbh43OyN5iv5cYzcE+PKvC4K6EyMzMHPc++D99J/16ZX7 C0gkrk1FJo8xmOI54/qP3pQDp99Xju0qKfb2hIugbzzNAGByENZShrPsNCRstIxU 1+N5gNjxKFuX3avr3asCRG73OK7Vx0QH2QitQ9+YBErT9X3i9hZBKKSSLchwgqPI LpOrNiMLFoW3hQ/adMqbdCGe06W3SSIRUSX8xw5lVqM4dviulXWhgfhNQB4jv51Y ZeLuMbIR2e2eV6par7+n =iuFl -----END PGP SIGNATURE-----

1 0

timegm() function and state
by Doug Bailey Aug. 3, 2012

Aug. 3, 2012

One of my users reported that the first time timegm() is called, it returns -1. Investigating, it seems that there's a problem with the gmtptr state. Take the following code: int main(int argc, char *argv[]) { struct tm localt; localt.tm_sec = 0; localt.tm_min = 0; localt.tm_hour = 0; localt.tm_mday = 1; localt.tm_year = 113; localt.tm_mon = 0; localt.tm_wday = 0; localt.tm_yday = 0; localt.tm_isdst = 0; localt.tm_gmtoff = 0; localt.tm_zone = 0; mktime(&localt); printf("asctime: %s", asctime(&localt)); printf("timegm(&localt) = %d (errno %s)\n", timegm(&localt), strerror(errno)); printf("timegm(&localt) = %d (errno %s)\n", timegm(&localt), strerror(errno)); printf("timegm(&localt) = %d (errno %s)\n", timegm(&localt), strerror(errno)); return 0; } Run it and you get: asctime: Tue Jan 1 00:00:00 2013 timegm(&localt) = -1 (errno No such process) timegm(&localt) = 1356998400 (errno No such process) timegm(&localt) = 1356998400 (errno No such process) The first time timegm() is invoked, it fails. Subsequent invocations work correctly. I didn't manage to track down the root problem, but I found if I forced gmtptr to load (as gmtsub should), then everything works correctly. time_t timegm(tmp) struct tm * const tmp; { if (!gmt_is_set) { gmt_is_set = TRUE; #ifdef ALL_STATE gmtptr = (struct state *) calloc(1, sizeof *gmtptr); if (gmtptr != NULL) #endif /* defined ALL_STATE */ gmtload(gmtptr); } if (tmp != NULL) tmp->tm_isdst = 0; return time1(tmp, gmtsub, 0L); } I just copied the gmt_is_set test from gmtsub. I realize that gmtsub gets called anyways (timegm->time1->time2->gmtsub), but maybe the context doesn't get loaded quickly enough. I haven't tested, but I suspect that timeoff will have the same problem since it also invokes time1(..,gmtsub,..) without first ensuring the gmt context is loaded. Doug

2 1

[PATCH 4/4] Use a single version number for both code and data.
by Paul Eggert July 25, 2012

July 25, 2012

* Makefile (VERSION): New macro, superseding ... (TZCODE_VERSION, TZDATA_VERSION): Remove. All uses changed. (version.h): Call the variable TZVERSION, not TZCODE_VERSION. All uses changed. (tzselect): Interpolate TZVERSION, not TZCODE_VERSION. (public): Adjust to new versioning scheme. * tzselect.ksh (TZVERSION): Rename from TZCODE_VERSION. * zdump.c, zic.c (main): Use TZVERSION, not TZCODE_VERSION. --- Makefile | 11 +++++------ tzselect.ksh | 4 ++-- zdump.c | 2 +- zic.c | 2 +- 4 files changed, 9 insertions(+), 10 deletions(-) diff --git a/Makefile b/Makefile index 270fbaa..510d77f 100644 --- a/Makefile +++ b/Makefile @@ -3,8 +3,7 @@ # 2009-05-17 by Arthur David Olson. # Version numbers of the code and data distributions. -TZCODE_VERSION = tzcode2012c -TZDATA_VERSION = tzdata2012d +VERSION = 2012d # Change the line below for your time zone (after finding the zone you want in # the time zone files, or adding it to a time zone file). @@ -327,7 +326,7 @@ INSTALL: ALL install date.1 version.h: echo >$@ \ - 'static char const TZCODE_VERSION[]="$(TZCODE_VERSION)";' + 'static char const TZVERSION[]="tz$(VERSION)";' zdump: $(TZDOBJS) $(CC) $(CFLAGS) $(LFLAGS) $(TZDOBJS) $(LDLIBS) -o $@ @@ -379,7 +378,7 @@ tzselect: tzselect.ksh sed \ -e 's|AWK=[^}]*|AWK=$(AWK)|g' \ -e 's|TZDIR=[^}]*|TZDIR=$(TZDIR)|' \ - -e 's|$TZCODE_VERSION$=.*|\1=$(TZCODE_VERSION)|' \ + -e 's|$TZVERSION$=.*|\1=tz$(VERSION)|' \ <$? >$@ chmod +x $@ @@ -437,10 +436,10 @@ public: $(ENCHILADA) set-timestamps $(AWK) -f checktab.awk $(PRIMARY_YDATA) LC_ALL=C && export LC_ALL && \ tar $(TARFLAGS) -cf - $(DOCS) $(SOURCES) $(MISC) *.[1-8].txt | \ - gzip $(GZIPFLAGS) > $(TZCODE_VERSION).tar.gz + gzip $(GZIPFLAGS) > tzcode$(VERSION).tar.gz LC_ALL=C && export LC_ALL && \ tar $(TARFLAGS) -cf - $(DATA) | \ - gzip $(GZIPFLAGS) > $(TZDATA_VERSION).tar.gz + gzip $(GZIPFLAGS) > tzdata$(VERSION).tar.gz typecheck: make clean diff --git a/tzselect.ksh b/tzselect.ksh index 3ebaaa7..9e8ceea 100644 --- a/tzselect.ksh +++ b/tzselect.ksh @@ -1,6 +1,6 @@ #! /bin/ksh -TZCODE_VERSION=see_Makefile +TZVERSION=see_Makefile # Ask the user about the time zone, and output the resulting TZ value to stdout. # Interact with the user via stderr and stdin. @@ -55,7 +55,7 @@ EOF exit 0 elif [ "$1" = "--version" ]; then cat <<EOF -tzselect $TZCODE_VERSION +tzselect $TZVERSION EOF exit 0 fi diff --git a/zdump.c b/zdump.c index cbd3338..7b96a4f 100644 --- a/zdump.c +++ b/zdump.c @@ -282,7 +282,7 @@ char * argv[]; progname = argv[0]; for (i = 1; i < argc; ++i) if (strcmp(argv[i], "--version") == 0) { - (void) printf("%s\n", TZCODE_VERSION); + (void) printf("%s\n", TZVERSION); exit(EXIT_SUCCESS); } else if (strcmp(argv[i], "--help") == 0) { usage(stdout, EXIT_SUCCESS); diff --git a/zic.c b/zic.c index 28d7bcf..4ae31a2 100644 --- a/zic.c +++ b/zic.c @@ -497,7 +497,7 @@ char * argv[]; } for (i = 1; i < argc; ++i) if (strcmp(argv[i], "--version") == 0) { - (void) printf("%s\n", TZCODE_VERSION); + (void) printf("%s\n", TZVERSION); exit(EXIT_SUCCESS); } else if (strcmp(argv[i], "--help") == 0) { usage(stdout, EXIT_SUCCESS); -- 1.7.9.5

2 2

technical details for today's code and data release
by Paul Eggert July 25, 2012

July 25, 2012

# This email documents how I built today's tzcode and tzdata release. # It mostly uses POSIX shell syntax (hence the leading '#' here). # The idea is to document what I did this time, so that more of # this stuff can be done better and/or automated next time. # Here's how I built the release, on Ubuntu 12.04 x86, # starting from the unofficial public repository. git clone -n https://github.com/eggert/tz.git (cd tz git checkout ee42236d5192a0199bbac05385fe2f613b1b30e2 make check make public ) # Here's how I imported the release into the private IANA repository. # This procedure makes sense only if one has access to this repository, # which is not currently public. (cd tz-iana cp ../tz/tzcode2012c.tar.gz ../tz/tzdata2012d.tar.gz releases ln -sf releases/tzcode2012c.tar.gz tzcode-latest.tar.gz ln -sf releases/tzdata2012d.tar.gz tzdata-latest.tar.gz (cd data && tar xf ../tzdata-latest.tar.gz) (cd code && tar xf ../tzcode-latest.tar.gz) git add releases/tzcode2012c.tar.gz releases/tzdata2012d.tar.gz git commit -am'tzcode2012c and tzdata2012d' # Oops, forgot to update '.latest'. Do that now. (echo 'data tzdata2012d.tar.gz 2012-07-20' echo 'code tzcode2012c.tar.gz 2012-07-20' ) >.latest git commit -am'tzcode2012c and tzdata2012d in .latest.' ) # Here's how I created the signatures for the distributions. (cd tz gpg --detach-sign --armor -o tzcode2012c.tar.gz.sign tzcode2012c.tar.gz gpg --detach-sign --armor -o tzdata2012d.tar.gz.sign tzdata2012d.tar.gz ) # Here are the signatures themselves, generated by: (cd tz; more tzcode2012c.tar.gz.sign tzdata2012d.tar.gz.sign) :::::::::::::: tzcode2012c.tar.gz.sign :::::::::::::: -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iQIcBAABAgAGBQJQCPX9AAoJEO2X6Q5iqn40A+EQAKquIXzcZH92dAoJeGUO3Hfr p0pdz2wXQZvOhBZn3O2MVWZtLgER+e/qRBFi7zAm7/PAjJUwP54mVIdWnDUCGvn1 ieN/qd4qoF00d5g+xHfmo8XmogXfzjiQ6hqRqN3fSz0qzp2D9+8ckPyZyvU9hAYA OfcCB/EeSvi+U2Gg4GZzx0uhq5huhFPl6SRy+SHolAfWb0o+xxYlVsLv94jBX3rv MTEBkNPgx6FXdarfa1agvaJK4j1fWUQ056/80jfmrwZDrewTfaF2hNEDS0MAM9rC VGdbsW5mqikWEt2mFL0n1fn56bgxOFPDJOQvWt2gJ23tfAR/jhSYuQKwuAlwbwTm bTUcrxOJyyh6bUkgujHrkqSOkY+5ya6laF91ejn95tzfcTHqXOQmFT2jVBlZBhey f4pP16QHb5Q7Nf+kA//1oNDSYFrDMtfLoj/9IqLlxgzlfnwNjE+udOWgEVNGS9iF 99Bh+H0GNl1sd2xq65b7Ne+4M6pHuJzOtSCJwmJ+3+owYFs/ocIH97IAEMoFkwWJ 3LPSqLhep5OC+Z2O9OgYREhVjdPuUavz1+UF6xOroEbCAg4Act9Ok8HGlZ+q7gRk FuR/ItM+5dDFBrUQrNXRuAjR9HrcoZ9zyqMtFPzV6indI7EQwMuCtZopsi9PrC6A arMfzjlKJBJ4UCCy45wW =Dfi2 -----END PGP SIGNATURE----- :::::::::::::: tzdata2012d.tar.gz.sign :::::::::::::: -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iQIcBAABAgAGBQJQCPYfAAoJEO2X6Q5iqn406TwP/RHspsy44+WUFIEbfsijFkEf VW409JP0QSC0ocRe8ir1ijorNggF+j62uAQjvSSVaNb/tLy67/j+sUR5xU1/ccw6 bWM46r5P28ubSX4rs5FHQfOmOgA7LRBP+3+C6TJ8iUzSfHszZI00bOdAZWaFcQxw /hlBK83rp7vwQj2Sk9NU8tzyv1paGOcM8Ps/aL5gsZpkF48v2GQ0kDrZwazTgODD NZ4rvMg4mlNCZz6sSyTRa02B0uDkEXsNKLf/Gf3OfPIkoGdRQT6sa7qDdYXcjM1o QdTTjg+HAnCxt3CDmpNGC4K0y7aHRmZYlge1cXt0uZOvtmO9lTlSGUQIied1FuBs EaawtZw/L3Pd9bG1FeodPN/NIVQLaTss+ERU2FsgQjiOWxnaLvzSFmfBi/TrDpkW yPqqskACwhLHGLRu3uSThjAFA2OzItCdqbwHfMMNoo0aXL85ZipUpH3CnhDwOQxn I3s43HLepBpJi7RofR4KtolnRvlu2rfEAqmhUOhWk1ACVghI4ev2papQ5boucf0N B93AUQnv6WO9WgmeWrQ7CnYPoydBIlOYdjKmiOZ9c45qFsHEX76ZfUY6sbi3sLE/ mxOsYFZLBf5my8APmyespgO+WTXFCbMHBKbeKQRuchRuXYi5/72CxVJ1TZEt+aMd 0PJmnu0eg3TKYTE6/vX8 =eKib -----END PGP SIGNATURE-----

8 17

[PATCH 3/4] * .gitignore: New file.
by Paul Eggert July 25, 2012

July 25, 2012

It's not clear to me that the .gitignore file should be part of the tzcode or tzdata distributions, as it's needed only if you've gotten the files from the git repository, so for now the Makefile is not modified to add this to one or the other distribution. --- .gitignore | 10 ++++++++++ 1 file changed, 10 insertions(+) create mode 100644 .gitignore diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..d36f26e --- /dev/null +++ b/.gitignore @@ -0,0 +1,10 @@ +*.o +*.tar.gz +*.txt +*~ +ChangeLog +tzselect +version.h +yearistype +zdump +zic -- 1.7.9.5

2 1

[PATCH 2/4] * australasia (Pacific/Fakaofo): Tokelau is UTC+13, not UTC+14.
by Paul Eggert July 25, 2012

July 25, 2012

Assume Shanks & Pottenger have been wrong by an hour since 1901. Reported by Steffen Thorsen in <http://mm.icann.org/pipermail/tz/2012-July/018168.html>. --- australasia | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/australasia b/australasia index dfadfa9..c7d5611 100644 --- a/australasia +++ b/australasia @@ -640,25 +640,25 @@ Zone Pacific/Guadalcanal 10:39:48 - LMT 1912 Oct # Honiara # # From Gwillim Law (2011-12-29) # A correspondent informed me that Tokelau, like Samoa, will be skipping -# December 31 this year, thereby changing its time zone from UTC-10 to -# UTC+14. When I tried to verify this statement, I found a confirming -# article in Time magazine online -# <a href="http://www.time.com/time/world/article/0,8599,2103243,00.html"> -# (http://www.time.com/time/world/article/0,8599,2103243,00.html) -# </a> +# December 31 this year ... # -# From Jonathan Leffler (2011-12-29) -# Information from the BBC to the same effect: -# <a href="http://www.bbc.co.uk/news/world-asia-16351377"> -# http://www.bbc.co.uk/news/world-asia-16351377 -# </a> +# From Steffen Thorsen (2012-07-25) +# ... we double checked by calling hotels and offices based in Tokelau asking +# about the time there, and they all told a time that agrees with UTC+13.... +# Shanks says UTC-10 from 1901 [but] ... there is a good chance the change +# actually was to UTC-11 back then. # -# Patch supplied by Tim Parenti (2011-12-29) +# From Paul Eggert (2012-07-25) +# A Google Books snippet of Appendix to the Journals of the House of +# Representatives of New Zealand, Session 1948, +# <http://books.google.com/books?id=ZaVCAQAAIAAJ>, page 65, says Tokelau +# was "11 hours slow on G.M.T." Go with Thorsen and assume Shanks & Pottenger +# are off by an hour starting in 1901. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Pacific/Fakaofo -11:24:56 - LMT 1901 - -10:00 - TKT 2011 Dec 30 # Tokelau Time - 14:00 - TKT + -11:00 - TKT 2011 Dec 30 # Tokelau Time + 13:00 - TKT # Tonga # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S -- 1.7.9.5

1 0