- tz - lists.iana.org

Fiji DST 2018/19
by Raymond Kumar July 15, 2018

July 15, 2018

Hi Paul, There is a DST time change in Fiji island from November 4th. http://www.fijitimes.com/government-approves-2018-daylight-saving/ The attached order states: Members of the public are advised to move their clocks one hour ahead at 2am on Sunday November 4. The daylight saving period will end at 3am on Sunday January 13, 2019. Fiji rules in tzdata2018e.tar.gz: There is a need to change the NON DST time from 14th Jan to 13th Jan. Can a new tzdata be generated for Fiji Islands. Rule Fiji 2014 max - Nov Sun>=1 2:00 1:00 - Rule Fiji 2015 max - Jan Sun>=14 3:00 0 - Thanks, Raymond Kumar Head of Billing & Products +679-7015315 ________________________________ Notice of Confidentiality: The information contained in this communication is intended solely for the use of the individual or entity to whom it is addressed and others authorized to receive it. It may contain confidential or legally privileged information. If you are not the intended recipient you are hereby notified that any disclosure, copying, distribution or taking any action in reliance on the contents of this information is strictly prohibited and may be unlawful. If you have received this communication in error, please notify us immediately by responding to this email and then delete it from your system.

4 9

Re: [tz] EU Public Consultation on summertime arrangements
by Pierpaolo Bernardi July 11, 2018

July 11, 2018

Not time related. Only for pedantry sake: at the border between San Marino and Italy there *are* controls, by the Italian police. This is due to different tax laws in the two countries, which would make shopping in San Marino very convenient for Italy residents, if they weren't controlled when exiting from RSM for any good bought in RSM, and forced to pay due taxes. FWIW Il giorno 11 luglio 2018, alle ore 09:08, John Wilcock <john(a)tradoc.fr> ha scritto: Le 11/07/2018 à 00:05, Brian Inglis a écrit : > Most people are probably not that aware of European geography [...] Taking the world as a whole, you are no doubt correct, but most of the people at the EU who will be reading Paul's submission will be perfectly aware of European geography! >> I don't see the relevance of "Schengen area" - all current and potential >> Schengen area countries are either EU or EEA/EFTA members. > The Schengen Area also includes non-EU/EEA/EFTA countries Monaco, San Marino, > and the Vatican, which have not signed the Schengen agreement, but have no > borders, so are de facto included. While these three countries are indeed de facto within the Schengen area due to the lack of border *controls* — they do still have borders ;-) — it still seems (at least to me as a European) very odd to mention Schengen purely in order to include its *non*-signatories, especially as the next item in the list is in any case the all-inclusive "other countries not in the EU". -- John

1 0

(no subject)
by Ashish Ahmed July 10, 2018

July 10, 2018

Confirm

1 0

tz-art suggestion
by Steve Jones July 6, 2018

July 6, 2018

Not sure how this was missed, having passed its third birthday, but probably deserves a mention in tz-art: "Daylight Saving Time: How is this still a thing?" https://www.youtube.com/watch?v=br0NW9ufUUw Regards, Steve Jones [image: Emacs!]

3 2

strftime and nonnormalized dates/times
by scs＠eskimo.com July 5, 2018

July 5, 2018

I have a question about the strftime function. This is not related to time zones per se, but there are a lot of date/time programming experts here, and of course there's a reference implementation of strftime in tzcode. [Full disclosure: I am posting the same question to StackOverflow: https://stackoverflow.com/questions/51145727/ .] Most definitions of the %H, %M, and %S specifiers (including those in the ANSI/ISO C Standards) contain language like %H is replaced by the hour (24-hour clock) as a decimal number (00-23). Now, that little notation "(00-23)" at the end tells you the range of numbers you're going to see *if the struct tm you're printing was just created by localtime or gmtime*. My question concerns the presence or absence of a requirement that the struct tm being referenced by strftime *must* have been one just generated by localtime or gmtime. I don't think there's such a requirement, but if there isn't, those notations "(00-23)" and "(00-59)" don't really have any normative implications for strftime -- they're more of a misplaced requirement on the values generated by localtime and gmtime. But if strftime doesn't restrict the ranges of the various struct tm fields to their "normal" values, it makes me wonder how much latitude a push-the-boundaries programmer actually has when calling strftime. Are the values as unlimited as they are when calling, say, mktime? To make the question more concrete, it came to me when I found myself writing this code: time_t dt = t2 - t1; struct tm *tmp = gmtime(&dt); if(dt > 86400) tmp->tm_hour += dt / 86400 * 24; strftime(etbuf, sizeof(etbuf), "elapsed: %H:%M:%S", tmp); That is, I'm subtracting two times, and letting gmtime and strftime convert the difference to HH:MM:SS format for me. But if the time delta was more than a day, I'm not printing it as days; I'm just lumping it in with the hours (which, it's true, might end up being a very large number if there were months or years between t1 and t2). So, language lawyer question: Is this legal and portable? When calling strftime, are tm_hour, tm_min, and the rest limited to their normal ranges, or not? (Yes, I've tried it, and nonnormalized values work -- not at all surprisingly -- as expected under the popular implementations, but that's not the question.)

4 6

African Timezones
by Robin Roos July 4, 2018

July 4, 2018

Hi Folks, I am using a Java platform (hosted on Linux and on Windows) in which I find some expected timezones to be missing. I write to enquire whether they are defined at IANA and, if not, whether they can be added. The timezones in question are as follows: Africa/Praia (capital city of Cabo Verde, also known also as Cape Verde) = UTC-1 CVT (Cabo Verde Time) = UTC-1 (one hour West of Greenwich) WAT (West Africa Time) = UTC+1 (one hour East of Greenwich) CAT (Central Africa Time) = UTC+2 (two hours East of Greenwich) EAT (East Africa Time) = UTC+3 (three hours East of Greenwich) My source for this information is: https://www.worldtimezone.com/time-africa24.php Many thanks, Robin.

2 2

"time zone" vs "timezone" in documentation
by Paul Eggert July 4, 2018

July 4, 2018

[I am cc'ing this to the tz mailing list (and changing the subject line) to give tz readers a heads-up about the terminology issue. The question here is: should tzdb man pages say "time zone" or "timezone"? For context please see Michael Kerrisk's email <https://marc.info/?m=152964652828320>, which says: > When I inherited the project, the pages used a mixture of "time zone" > and "timezone", with the former predominant. I nevertheless > standardized on the latter, and although I don't recall for sure, I > suspect it was because that is the spelling used in POSIX. (As an > aside, there's an argument that--because POSIX--tzdb might want to > consider switching spellings.) I'm not religious about the particular > choice (although I have naturally now got used to the particular > choice I made some years ago), but I did make that choice because I > want consistency within the project, and I'm reluctant to introduce > inconsistency. and my response <https://marc.info/?m=152964676331902>, which says: > I deferred to POSIX for "timestamp" versus "time stamp", but > "timezone" is a bridge too far for me. ] On second thought, perhaps I was too hasty. We could distinguish "time zone" in the usual English-language sense (a set of geographic locations that currently share the same standard time offset from UTC) from "timezone" in the POSIX sense (a history and predicted future of UTC offsets, abbreviations and isdst flags). If so, the tzdb documentation could be more careful about using "time zone" for the former and "timezone" for the latter, and this would make for fewer changes to the GNU/Linux man-pages for tzdb. I can look into this and propose an updated set of tzdb-related patches accordingly. If I were designing the terminology from scratch, I wouldn't specify two nearly-identical phrases "time zone" and "timezone" to mean such different things. However, the phrases do have the advantage of existing practice (common English usage and POSIX, respectively).

9 11

[PROPOSED] * tz-link.html (Precision timekeeping): Mention Huygens.
by Paul Eggert June 29, 2018

June 29, 2018

--- tz-link.html | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/tz-link.html b/tz-link.html index 548fbba..7ed1efd 100644 --- a/tz-link.html +++ b/tz-link.html @@ -813,11 +813,15 @@ title="Network Time Protocol">NTP</abbr>: The Network Time Protocol</a> (Internet <abbr>RFC</abbr> 5905) discusses how to synchronize clocks of Internet hosts.</li> +<li>The <a href="https://www.usenix.org/system/files/conference/nsdi18/nsdi18-geng.pdf"><font style="font-variant: small-caps">Huygens</font></a> +family of software algorithms can achieve accuracy to a few tens of +nanoseconds in scalable server farms without special hardware.</li> <li>The <a href="https://www.nist.gov/intelligent-systems-division/ieee-1588">Precision Time Protocol</a> (<abbr title="Institute of Electrical and Electronics Engineers">IEEE</abbr> 1588) -can achieve submicrosecond clock accuracy on a local area network.</li> +can achieve submicrosecond clock accuracy on a local area network +with special-purpose hardware.</li> <li><a href="https://tools.ietf.org/html/rfc4833">Timezone Options for <abbr title="Dynamic Host Configuration Protocol">DHCP</abbr></a> -- 2.17.1

1 0

California will vote on permanent daylight
by Steve Allen June 29, 2018

June 29, 2018

The governor approved AB 807 https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=201720… Daylight time in California was established by the voters, and must be repealed by the voters, so nothing happens until after an election decides whether to ask the state legislature to consider how permanent daylight time could be achieved. -- Steve Allen <sla(a)ucolick.org> WGS-84 (GPS) UCO/Lick Observatory--ISB 260 Natural Sciences II, Room 165 Lat +36.99855 1156 High Street Voice: +1 831 459 3046 Lng -122.06015 Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m

1 0

[PROPOSED] Make Makefile safer for parallelism
by Paul Eggert June 27, 2018

June 27, 2018

Although this still does not fully work with GNU Make, it’s better than what we had. * Makefile (TIME_T_ALTERNATIVES_HEAD) (TIME_T_ALTERNATIVES_TAIL): New macros. (TIME_T_ALTERNATIVES): Use them. (TZS_CUTOFF_FLAG): New macro. ($(ZDS)): Use it, so that submakes can use -t instead of -c cutoffs. (EIGHT_YARDS): New macro. (ENCHILADA): Use it. (INSTALL_DATA_DEPS, TZS_NEW_DEPS): New macros. (install_data, posix_only, right_only, posix_packrat, $(TZS_NEW)): Use them to avoid duplicate work with GNU make -j. ($(TZS_NEW), check_zishrink_posix, check_zishrink_right, check_public) (check_time_t_alternatives): Use rule-specific temporary directories so that other rules can run in parallel. ($(TZS_NEW)): Check for sort failure. ($(TZS)): Simply touch it if it doesn’t exist, so that it’s empty. (check_tzs): Treat empty $(TZS) as nonexistent. (check_character_set, check_white_space, check_name_lengths) (check_sorted, check_links, check_tables, check_tzs, check_web) (check_zishrink, check_public): No longer phony. (check_zishrink_posix, check_zishrink_right) (typecheck_long_long, typecheck_unsigned): New rules, so that they can be done in parallel. (clean_misc): Clean the timestamp files that are now made due to rules no longer being phony. (set-timestamps.out): Do not set timestamp of $(TZS). (set-tzs-timestamp.out): New rule. (check_time_t_alternatives): Break into subrules, one for each alternative, so that they can be done in parallel. (tarballs, rearguard_tarballs, traditional_tarballs) (signatures, rearguard_signatures, traditional_signatures): Depend on set-timestamps.out and rearguard.zi. (tzdb-$(VERSION).tar.lz): Depend on set-tzs-timestamp.out. (TYPECHECK_CFLAGS): New macro. --- Makefile | 240 +++++++++++++++++++++++++++++++++++++-------------------------- 1 file changed, 142 insertions(+), 98 deletions(-) diff --git a/Makefile b/Makefile index 2289dff..29058e5 100644 --- a/Makefile +++ b/Makefile @@ -106,8 +106,10 @@ MANDIR = $(TOPDIR)/$(USRSHAREDIR)/man LIBDIR = $(TOPDIR)/$(USRDIR)/lib -# Types to try, as an alternative to time_t. int64_t should be first. -TIME_T_ALTERNATIVES = int64_t int32_t uint32_t uint64_t +# Types to try, as an alternative to time_t. +TIME_T_ALTERNATIVES = $(TIME_T_ALTERNATIVES_HEAD) $(TIME_T_ALTERNATIVES_TAIL) +TIME_T_ALTERNATIVES_HEAD = int64_t +TIME_T_ALTERNATIVES_TAIL = int32_t uint32_t uint64_t # What kind of TZif data files to generate. # (TZif is the binary time zone data format that zic generates.) @@ -493,11 +495,14 @@ AWK_SCRIPTS= checklinks.awk checktab.awk leapseconds.awk \ ziguard.awk zishrink.awk MISC= $(AWK_SCRIPTS) zoneinfo2tdf.pl TZS_YEAR= 2050 +TZS_CUTOFF_FLAG= -c $(TZS_YEAR) TZS= to$(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) tzdata.zi +# EIGHT_YARDS is just a yard short of the whole ENCHILADA. +EIGHT_YARDS = $(COMMON) $(DOCS) $(SOURCES) $(DATA) $(MISC) tzdata.zi +ENCHILADA = $(EIGHT_YARDS) $(TZS) # Consult these files when deciding whether to rebuild the 'version' file. # This list is not the same as the output of 'git ls-files', since @@ -606,14 +611,16 @@ INSTALLARGS = \ YEARISTYPE='$(YEARISTYPE)' \ ZIC='$(ZIC)' +INSTALL_DATA_DEPS = zic leapseconds yearistype tzdata.zi + # 'make install_data' installs one set of TZif files. -install_data: zic leapseconds yearistype tzdata.zi +install_data: $(INSTALL_DATA_DEPS) $(ZIC_INSTALL) tzdata.zi -posix_only: +posix_only: $(INSTALL_DATA_DEPS) $(MAKE) $(INSTALLARGS) LEAPSECONDS= install_data -right_only: +right_only: $(INSTALL_DATA_DEPS) $(MAKE) $(INSTALLARGS) LEAPSECONDS='-L leapseconds' \ install_data @@ -640,7 +647,7 @@ posix_right: posix_only # This obsolescent rule is present for backwards compatibility with # tz releases 2014g through 2015g. It should go away eventually. -posix_packrat: +posix_packrat: $(INSTALL_DATA_DEPS) $(MAKE) $(INSTALLARGS) PACKRATDATA=backzone posix_only zones: $(REDO) @@ -651,29 +658,33 @@ ZDS = dummy.zd # Rule used only by submakes invoked by the $(TZS_NEW) rule. # It is separate so that GNU 'make -j' can run instances in parallel. $(ZDS): zdump - ./zdump -i -c $(TZS_YEAR) '$(wd)/'$$(expr $@ : '$.*$.zd') >$@ - -$(TZS_NEW): tzdata.zi zdump zic - rm -fr tzs.dir - mkdir tzs.dir - $(zic) -d tzs.dir tzdata.zi + ./zdump -i $(TZS_CUTOFF_FLAG) '$(wd)/'$$(expr $@ : '$.*$.zd') \ + >$@ + +TZS_NEW_DEPS = tzdata.zi zdump zic +$(TZS_NEW): $(TZS_NEW_DEPS) + rm -fr tzs$(TZS_YEAR).dir + mkdir tzs$(TZS_YEAR).dir + $(zic) -d tzs$(TZS_YEAR).dir tzdata.zi $(AWK) '/^L/{print "Link\t" $$2 "\t" $$3}' \ tzdata.zi | LC_ALL=C sort >$@.out wd=`pwd` && \ - set x `$(AWK) '/^Z/{print "tzs.dir/" $$2 ".zd"}' tzdata.zi \ + x=`$(AWK) '/^Z/{print "tzs$(TZS_YEAR).dir/" $$2 ".zd"}' \ + tzdata.zi \ | LC_ALL=C sort -t . -k 2,2` && \ + set x $$x && \ shift && \ ZDS=$$* && \ - $(MAKE) wd="$$wd" TZS_YEAR=$(TZS_YEAR) ZDS="$$ZDS" $$ZDS && \ - sed 's,^TZ=".*tzs\.dir/,TZ=",' $$ZDS >>$@.out - rm -fr tzs.dir + $(MAKE) wd="$$wd" TZS_CUTOFF_FLAG="$(TZS_CUTOFF_FLAG)" \ + ZDS="$$ZDS" $$ZDS && \ + sed 's,^TZ=".*\.dir/,TZ=",' $$ZDS >>$@.out + rm -fr tzs$(TZS_YEAR).dir mv $@.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 +# If $(TZS) 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 + touch $@ force_tzs: $(TZS_NEW) cp $(TZS_NEW) $(TZS) @@ -718,12 +729,14 @@ check_character_set: $(ENCHILADA) leapseconds yearistype.sh zone.tab && \ ! grep -Env $(OK_LINE) $(ENCHILADA); \ } + touch $@ check_white_space: $(ENCHILADA) patfmt=' \t|[\f\r\v]' && pat=`printf "$$patfmt\\n"` && \ ! grep -En "$$pat" $(ENCHILADA) ! grep -n '[[:space:]]$$' \ $$(ls $(ENCHILADA) | grep -Fvx leap-seconds.list) + touch $@ PRECEDES_FILE_NAME = ^(Zone|Link[[:space:]]+[^[:space:]]+)[[:space:]]+ FILE_NAME_COMPONENT_TOO_LONG = \ @@ -732,6 +745,7 @@ FILE_NAME_COMPONENT_TOO_LONG = \ check_name_lengths: $(TDATA_TO_CHECK) backzone ! grep -En '$(FILE_NAME_COMPONENT_TOO_LONG)' \ $(TDATA_TO_CHECK) backzone + touch $@ CHECK_CC_LIST = { n = split($$1,a,/,/); for (i=2; i<=n; i++) print a[1], a[i]; } @@ -744,52 +758,61 @@ check_sorted: backward backzone iso3166.tab zone.tab zone1970.tab LC_ALL=C sort -c $(AWK) '/^[^#]/ $(CHECK_CC_LIST)' zone1970.tab | \ LC_ALL=C sort -cu + touch $@ check_links: checklinks.awk $(TDATA_TO_CHECK) tzdata.zi $(AWK) -f checklinks.awk $(TDATA_TO_CHECK) $(AWK) -f checklinks.awk tzdata.zi + touch $@ check_tables: checktab.awk $(PRIMARY_YDATA) $(ZONETABLES) for tab in $(ZONETABLES); do \ $(AWK) -f checktab.awk -v zone_table=$$tab $(PRIMARY_YDATA) \ || exit; \ done + touch $@ check_tzs: $(TZS) $(TZS_NEW) - diff -u $(TZS) $(TZS_NEW) + if test -s $(TZS); then \ + diff -u $(TZS) $(TZS_NEW); \ + else \ + cp $(TZS_NEW) $(TZS); \ + fi + touch $@ # This checks only the HTML 4.01 strict page. # To check the the other pages, use <https://validator.w3.org/>. check_web: tz-how-to.html $(VALIDATE_ENV) $(VALIDATE) $(VALIDATE_FLAGS) tz-how-to.html + touch $@ # Check that zishrink.awk does not alter the data, and that ziguard.awk # preserves main-format data. -check_zishrink: zic leapseconds $(PACKRATDATA) $(TDATA) \ - $(DATAFORM).zi tzdata.zi - for type in posix right; do \ - mkdir -p time_t.dir/$$type time_t.dir/$$type-t \ - time_t.dir/$$type-shrunk && \ - case $$type in \ - right) leap='-L leapseconds';; \ - *) leap=;; \ - esac && \ - $(ZIC) $$leap -d time_t.dir/$$type $(DATAFORM).zi && \ +check_zishrink: check_zishrink_posix check_zishrink_right +check_zishrink_posix check_zishrink_right: \ + zic leapseconds $(PACKRATDATA) $(TDATA) $(DATAFORM).zi tzdata.zi + rm -fr $@.dir $(a)-t.dir $(a)-shrunk.dir + mkdir $@.dir $(a)-t.dir $(a)-shrunk.dir + case $@ in \ + *_right) leap='-L leapseconds';; \ + *) leap=;; \ + esac && \ + $(ZIC) $$leap -d $@.dir $(DATAFORM).zi && \ + $(ZIC) $$leap -d $(a)-shrunk.dir tzdata.zi && \ case $(DATAFORM) in \ main) \ - $(ZIC) $$leap -d time_t.dir/$$type-t $(TDATA) && \ + $(ZIC) $$leap -d $(a)-t.dir $(TDATA) && \ $(AWK) '/^Rule/' $(TDATA) | \ - $(ZIC) $$leap -d time_t.dir/$$type-t - \ - $(PACKRATDATA) && \ - diff -r time_t.dir/$$type time_t.dir/$$type-t;; \ - esac && \ - $(ZIC) $$leap -d time_t.dir/$$type-shrunk tzdata.zi && \ - diff -r time_t.dir/$$type time_t.dir/$$type-shrunk || exit; \ - done - rm -fr time_t.dir + $(ZIC) $$leap -d $(a)-t.dir - $(PACKRATDATA) && \ + diff -r $@.dir $(a)-t.dir;; \ + esac + diff -r $@.dir $(a)-shrunk.dir + rm -fr $@.dir $(a)-t.dir $(a)-shrunk.dir + touch $@ clean_misc: - rm -f core *.o *.out \ + rm -f *.o *.out $(TIME_T_ALTERNATIVES) \ + check_* core typecheck_* \ date tzselect version.h zdump zic yearistype libtz.a clean: clean_misc rm -fr *.dir *.zi tzdb-*/ $(TZS_NEW) @@ -826,10 +849,10 @@ $(MANTXTS): workman.sh # If git or GNU 'touch' is absent, don't bother to sync with git timestamps. # Also, set the timestamp of each prebuilt file like 'leapseconds' # to be the maximum of the files it depends on. -set-timestamps.out: $(ENCHILADA) +set-timestamps.out: $(EIGHT_YARDS) rm -f $@ if (type git) >/dev/null 2>&1 && \ - files=`git ls-files $(ENCHILADA)` && \ + files=`git ls-files $(EIGHT_YARDS)` && \ touch -md @1 test.out; then \ rm -f test.out && \ for file in $$files; do \ @@ -847,62 +870,74 @@ set-timestamps.out: $(ENCHILADA) exit; \ done touch -cmr `ls -t $(TZDATA_ZI_DEPS) | sed 1q` tzdata.zi - touch -cmr `ls -t $(TZS_DEPS) | sed 1q` $(TZS) touch -cmr `ls -t $(VERSION_DEPS) | sed 1q` version touch $@ +set-tzs-timestamp.out: $(TZS) + touch -cmr `ls -t $(TZS_DEPS) | sed 1q` $(TZS) + touch $@ # The zics below ensure that each data file can stand on its own. # We also do an all-files run to catch links to links. -check_public: - $(MAKE) maintainer-clean - $(MAKE) CFLAGS='$(GCC_DEBUG_FLAGS)' ALL - mkdir -p public.dir - for i in $(TDATA_TO_CHECK) tzdata.zi; do \ - $(zic) -v -d public.dir $$i 2>&1 || exit; \ +check_public: $(VERSION_DEPS) + rm -fr public.dir + mkdir public.dir + ln $(VERSION_DEPS) public.dir + cd public.dir && $(MAKE) CFLAGS='$(GCC_DEBUG_FLAGS)' ALL + for i in $(TDATA_TO_CHECK) public.dir/tzdata.zi; do \ + public.dir/zic -v -d public.dir/zoneinfo $$i 2>&1 || exit; \ done - $(zic) -v -d public.dir $(TDATA_TO_CHECK) + public.dir/zic -v -d public.dir/zoneinfo-all $(TDATA_TO_CHECK) rm -fr public.dir + touch $@ # Check that the code works under various alternative # implementations of time_t. -check_time_t_alternatives: - if diff -q Makefile Makefile 2>/dev/null; then \ - quiet_option='-q'; \ - else \ - quiet_option=''; \ - fi && \ +check_time_t_alternatives: $(TIME_T_ALTERNATIVES) +$(TIME_T_ALTERNATIVES_TAIL): $(TIME_T_ALTERNATIVES_HEAD) +$(TIME_T_ALTERNATIVES): $(VERSION_DEPS) + rm -fr $@.dir + mkdir $@.dir + ln $(VERSION_DEPS) $@.dir + case $@ in \ + int32_t) range=-2147483648,2147483648;; \ + u*) range=0,4294967296;; \ + *) range=-4294967296,4294967296;; \ + esac && \ wd=`pwd` && \ zones=`$(AWK) '/^[^#]/ { print $$3 }' <zone1970.tab` && \ - for type in $(TIME_T_ALTERNATIVES); do \ - mkdir -p time_t.dir/$$type && \ - $(MAKE) clean_misc && \ - $(MAKE) TOPDIR="$$wd/time_t.dir/$$type" \ - CFLAGS='$(CFLAGS) -Dtime_tz='"'$$type'" \ + if test $@ = $(TIME_T_ALTERNATIVES_HEAD); then \ + range_target=; \ + else \ + range_target=to$$range.tzs; \ + fi && \ + (cd $@.dir && \ + $(MAKE) TOPDIR="$$wd/$@.dir" \ + CFLAGS='$(CFLAGS) -Dtime_tz='"'$@'" \ REDO='$(REDO)' \ - install && \ - diff $$quiet_option -r \ - time_t.dir/int64_t/etc \ - time_t.dir/$$type/etc && \ - diff $$quiet_option -r \ - time_t.dir/int64_t/usr/share \ - time_t.dir/$$type/usr/share && \ - case $$type in \ - int32_t) range=-2147483648,2147483647;; \ - uint32_t) range=0,4294967296;; \ - int64_t) continue;; \ - *u*) range=0,10000000000;; \ - *) range=-10000000000,10000000000;; \ - esac && \ - echo checking $$type zones ... && \ - time_t.dir/int64_t/usr/bin/zdump -V -t $$range $$zones \ - >time_t.dir/int64_t.out && \ - time_t.dir/$$type/usr/bin/zdump -V -t $$range $$zones \ - >time_t.dir/$$type.out && \ - diff -u time_t.dir/int64_t.out time_t.dir/$$type.out \ - || exit; \ - done - rm -fr time_t.dir + D=$$wd/$@.dir \ + TZS_YEAR="$$range" TZS_CUTOFF_FLAG="-t $$range" \ + install $$range_target) && \ + test $@ = $(TIME_T_ALTERNATIVES_HEAD) || { \ + (cd $(TIME_T_ALTERNATIVES_HEAD).dir && \ + $(MAKE) TOPDIR="$$wd/$@.dir" \ + TZS_YEAR="$$range" TZS_CUTOFF_FLAG="-t $$range" \ + D=$$wd/$@.dir \ + to$$range.tzs) && \ + diff -u $(TIME_T_ALTERNATIVES_HEAD).dir/to$$range.tzs \ + $@.dir/to$$range.tzs && \ + if diff -q Makefile Makefile 2>/dev/null; then \ + quiet_option='-q'; \ + else \ + quiet_option=''; \ + fi && \ + diff $$quiet_option -r $(TIME_T_ALTERNATIVES_HEAD).dir/etc \ + $@.dir/etc && \ + diff $$quiet_option -r \ + $(TIME_T_ALTERNATIVES_HEAD).dir/usr/share \ + $@.dir/usr/share; \ + } + touch $@ TRADITIONAL_ASC = \ tzcode$(VERSION).tar.gz.asc \ @@ -913,7 +948,8 @@ ALL_ASC = $(TRADITIONAL_ASC) $(REARGUARD_ASC) \ tzdb-$(VERSION).tar.lz.asc tarballs rearguard_tarballs traditional_tarballs \ -signatures rearguard_signatures traditional_signatures: version +signatures rearguard_signatures traditional_signatures: \ + version set-timestamps.out rearguard.zi VERSION=`cat version` && \ $(MAKE) VERSION="$$VERSION" $@_version @@ -963,7 +999,7 @@ tzdata$(VERSION)-rearguard.tar.gz: rearguard.zi set-timestamps.out gzip $(GZIPFLAGS)) >$@.out mv $@.out $@ -tzdb-$(VERSION).tar.lz: set-timestamps.out +tzdb-$(VERSION).tar.lz: set-timestamps.out set-tzs-timestamp.out rm -fr tzdb-$(VERSION) mkdir tzdb-$(VERSION) ln $(ENCHILADA) tzdb-$(VERSION) @@ -979,14 +1015,24 @@ tzdb-$(VERSION).tar.lz.asc: tzdb-$(VERSION).tar.lz $(ALL_ASC): gpg2 --armor --detach-sign $? -typecheck: - $(MAKE) clean - for i in "long long" unsigned; \ - do \ - $(MAKE) CFLAGS="-DTYPECHECK -D__time_t_defined -D_TIME_T \"-Dtime_t=$$i\"" ; \ - ./zdump -v Europe/Rome ; \ - $(MAKE) clean ; \ - done +TYPECHECK_CFLAGS = $(CFLAGS) -DTYPECHECK -D__time_t_defined -D_TIME_T +typecheck: typecheck_long_long typecheck_unsigned +typecheck_long_long typecheck_unsigned: $(VERSION_DEPS) + rm -fr $@.dir + mkdir $@.dir + ln $(VERSION_DEPS) $@.dir + cd $@.dir && \ + case $@ in \ + *_long_long) i="long long";; \ + *_unsigned ) i="unsigned" ;; \ + esac && \ + typecheck_cflags='' && \ + $(MAKE) \ + CFLAGS="$(TYPECHECK_CFLAGS) \"-Dtime_t=$$i\"" \ + TOPDIR="`pwd`" \ + install + $@.dir/zdump -i -c 1970,1971 Europe/Rome + touch $@ zonenames: tzdata.zi @$(AWK) '/^Z/ { print $$2 } /^L/ { print $$3 }' tzdata.zi @@ -1002,9 +1048,7 @@ zic.o: private.h tzfile.h version.h .KEEP_STATE: .PHONY: ALL INSTALL all -.PHONY: check check_character_set check_links check_name_lengths -.PHONY: check_public check_sorted check_tables -.PHONY: check_time_t_alternatives check_tzs check_web check_white_space +.PHONY: check check_time_t_alternatives .PHONY: check_zishrink .PHONY: clean clean_misc dummy.zd force_tzs .PHONY: install install_data maintainer-clean names -- 2.7.4

1 0