- tz - lists.iana.org

A discussion on POSIX on new timezone related C APIs
by Random832 Nov. 13, 2015

Nov. 13, 2015

http://austingroupbugs.net/view.php?id=1005 The proposals seem to be: 1. Something involving integrating timezone information into locale_t 2. Something along the lines of NetBSD's _z/_rz functions The fact that no-one seems to have a reference C implementation of localizing time zone names/abbreviations also came up. I don't know how open they are to outsiders sticking their noses into the process though, just thought some people here might be interested. Also proposed is the addition of "tm_nsec" to struct tm, and the new APIs in one of the proposals use a struct timespec rather than time_t.

4 4

Re: [tz] efele.net/tz maps updated
by Steve Jones Nov. 11, 2015

Nov. 11, 2015

At 08:13 PM 11/10/2015, Chris Walton wrote: >There is one (and only one) community in the >western half of the PRRD;Â it is the isolated >community of Fort Ware (also known as >Kwadacha).Â It has a population of about 270. There is another - Tsay Keh Deme, about 86km to the south (56.892014, -124.963373). Population 245. I imagine it also observes Vancouver time. I have tried to reach out to the RCMP detachment office there via the "contact us" thingy on the RCMP website, but have not received any response. I didn't feel like making an international call. Their detachment webpage is: www.rcmp-grc.gc.ca/detach/en/d/212. If it's a free call for you, it may be worth giving them a ring. There area of coverage is basically the whole western part of the PRRD. I imagine they could speak with some authority about whatever time zone observances may be practiced throughout the area. But it has to be Pacific time - nothing else makes sense regardless of the opinion of the GIS folks at the PRRD. Their primary economic ties will be outside the PRRD to their south and west. Regards, Steve Jones Emacs! 14h pg

2 1

Re: [tz] efele.net/tz maps updated
by Steve Jones Nov. 11, 2015

Nov. 11, 2015

Sorry, typo, it's Dene, not Deme. Tsay Keh Dene. >There is another - Tsay Keh Deme, about 86km to the south \ Regards, Steve Jones Emacs!

1 0

Re: [tz] America/Santa_Isabel
by Steve Jones Nov. 10, 2015

Nov. 10, 2015

At 01:19 PM 11/9/2015, Matt Johnson (PNP) wrote: >Our research concluded that there is no known location in the state >of Baja California that observes Mexican DST rules. All are on US >DST rules. The 20km boundary mentioned in the law applies in the >other Mexican states, but not in Baja California due to the >exclusions mentioned in the same law. One more data point in support of your conclusion: I reached out to a contact on Cedros Island, which nearly touches the southern border of Baja. He indicates that clocks on the island adjusted on November 1. Regards, Steve Jones Emacs!

1 0

Re: [tz] America/Santa_Isabel
by Matt Johnson (PNP) Nov. 9, 2015

Nov. 9, 2015

WRT my previous comment: "I've got a parallel discussion going within Microsoft about this zone, and will share the findings of our research. " Our research concluded that there is no known location in the state of Baja California that observes Mexican DST rules. All are on US DST rules. The 20km boundary mentioned in the law applies in the other Mexican states, but not in Baja California due to the exclusions mentioned in the same law. This agrees with ADO’s observations about San Quintin. Also, the Wikipedia information about San Quintin becoming its own municipality is outdated. The decision was overturned by the Mexican supreme court due to jurisdictional issues. See: http://www2.scjn.gob.mx/juridica/engroses/cerrados/Publico/13000980.009-268… Thanks, Matt Johnson Microsoft ________________________________ Date: Sun, 1 Nov 2015 20:32:17 -0500 From: arthurdavidolson(a)gmail.com<mailto:arthurdavidolson@gmail.com> To: tz(a)iana.org<mailto:tz@iana.org> Subject: Re: [tz] America/Santa_Isabel And in today's final listen, XEQIN announced "cinco de la tarde con veinticuatro minutos" (literally "5 of the afternoon with 24 minutes") at about 8:24 PM USEST--the same as 5:24 PM USPST. At least this year, San Quintín seems to have followed has followed US rules. So much for that bookmark. @dashdashado On Sat, Oct 31, 2015 at 10:12 PM, Arthur David Olson <arthurdavidolson(a)gmail.com<mailto:arthurdavidolson@gmail.com>> wrote: This evening's next-to-last listen to XEQIN included advice to listeners to "atrarse una hora de su reloj" ("delay your clock by an hour") tonight before going to bed. The end-of-broadcast-day national anthem, due at 7:00 p. m. local time, came shortly after 10:00 p.m. USEDT, consistent with San Quintín still being on daylight saving time. The last listen should be tomorrow or Monday evening. @dashdashado On Tue, Oct 27, 2015 at 3:44 PM, Matt Johnson <mj1856(a)hotmail.com<mailto:mj1856@hotmail.com>> wrote: Acknowledged. Thanks for the clarification. -Matt ________________________________ From: Paul Eggert<mailto:eggert@cs.ucla.edu> Sent: ‎10/‎27/‎2015 10:31 AM To: Matt Johnson<mailto:mj1856@hotmail.com> Cc: Time Zone Mailing List<mailto:tz@iana.org> Subject: Re: [tz] America/Santa_Isabel Matt Johnson wrote: > there is a corresponding entry in Windows, "Pacific Standard Time (Mexico) |(UTC-08:00) Baja California" I'm afraid this will have to be adjusted on the Microsoft end. There needs to be a procedure for removing time zones that turn out to be mistakes. America/Santa_Isabel is a mistake, and I suppose we can view this as an opportunity for ironing out any procedural glitches while removing it. We do have a backward compatibility link on the tzdata side; perhaps that will help for the Microsoft side. > Also, there's a chance that San Quintin could indeed start using Mexican DST rules in 2017, We can create a zone America/San_Quintin if that happens.

1 0

No surprises in Trans-Pacific Partnership
by Arthur David Olson Nov. 5, 2015

Nov. 5, 2015

At least if the Washington Post's search... https://www.washingtonpost.com/news/the-fix/wp/2015/11/05/we-made-president… ...is to be believed, the Trans-Pacific Partnership mega-treaty has nothing to say about any "time zone"--phew. @dashdashado

1 0

Utility to dump raw values from binary zoneinfo files
by Andrew Paprocki Nov. 4, 2015

Nov. 4, 2015

I just released a little public domain utility I wrote to dump the raw values from zic-compiled binary files. This mainly aids anyone writing a zoneinfo reader who wants to easily view the binary structure and/or anyone creating contextual diffs from binary files. I'm posting it here in case anyone else finds it useful or wants to include it in tzcode. -Andrew https://github.com/apaprocki/zid Example output: $ ./zid /usr/share/zoneinfo/Etc/GMT 32: tzh: magic 'TZif' 32: tzh: version 2 32: tzh: ttisgmtcnt 1 32: tzh: ttisstdcnt 1 32: tzh: leapcnt 0 32: tzh: timecnt 0 32: tzh: typecnt 1 32: tzh: charcnt 4 32: gmtoff 0 isdst 0 abbrind 0 (GMT) 32: ttisstdcnt[0]: 0 32: ttisgmtcnt[0]: 0 64: tzh: magic 'TZif' 64: tzh: version 2 64: tzh: ttisgmtcnt 1 64: tzh: ttisstdcnt 1 64: tzh: leapcnt 0 64: tzh: timecnt 0 64: tzh: typecnt 1 64: tzh: charcnt 4 64: offset 0 isdst 0 abbrind 0 (GMT) 64: ttisstdcnt[0]: 0 64: ttisgmtcnt[0]: 0 TZ: 'GMT0'

1 0

Leap second coverage in mainstream media
by Arthur David Olson Nov. 4, 2015

Nov. 4, 2015

>From The Economist: "Why time may be running out for the leap second" http://www.economist.com/blogs/economist-explains/2015/11/economist-explain… @dashdashado

2 1

Fallback rules?
by Alfie John Nov. 4, 2015

Nov. 4, 2015

Hi TZers, Sorry if this has been asked before, but I couldn't find any documentation on the rules of what should happen for the following circumstances. Here's the definition for Australia/Brisbane: # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S Rule AQ 1971 only - Oct lastSun 2:00s 1:00 D Rule AQ 1972 only - Feb lastSun 2:00s 0 S Rule AQ 1989 1991 - Oct lastSun 2:00s 1:00 D Rule AQ 1990 1992 - Mar Sun>=1 2:00s 0 S Rule Holiday 1992 1993 - Oct lastSun 2:00s 1:00 D Rule Holiday 1993 1994 - Mar Sun>=1 2:00s 0 S Zone Australia/Brisbane 10:12:08 - LMT 1895 10:00 Aus AE%sT 1971 10:00 AQ AE%sT As the zone ends with AQ, and AQ does not have a "max" ending date, what should Australia/Brisbane fall back to after 1992? I thought maybe in this instance perhaps "Aus" as it's the one above it, but "Aus" seem to oddly be specific to both World Wars. I also don't think that it should then fall back to a GMTOFF = "10:12:08". That sounds awkward. Thanks, Alfie -- Alfie John alfie(a)fastmail.com

5 9

Re: [tz] News article: Turkey’s plan to ignore daylight saving time has been foiled by smartphones
by Matt Johnson Nov. 2, 2015

Nov. 2, 2015

Thanks Deborah! Is theee any way you could get Apple to post this somewhere official? Preferably there would be a page somewhere with each OS version and corresponding tzdb version, that gets maintained. Thanks, Matt ________________________________ From: Deborah Goldsmith<mailto:goldsmit@apple.com> Sent: ‎10/‎27/‎2015 12:40 PM To: Paul Eggert<mailto:eggert@CS.UCLA.EDU> Cc: Time zone mailing list<mailto:tz@iana.org> Subject: Re: [tz] News article: Turkey’s plan to ignore daylight saving time has been foiled by smartphones FWIW, 2015g was included in iOS 9.1, watchOS 2.0.1, and OS X 10.11.1. Deborah > On Oct 27, 2015, at 10:19 AM, Paul Eggert <eggert(a)CS.UCLA.EDU> wrote: > > Paul_Koning(a)dell.com wrote: >> Clearly it isn't sufficient for governments to announce this sort of thing with just 3 or 4 weeks lead time. > > It's not just computers. The Diyanet (the Presidency of Religious Affairs) warned that preprinted calendars contain Muslim prayer times that are now incorrect, and asked the faithful to manually add an hour. <http://www.dailysabah.com/nation/2015/10/25/turkey-to-end-daylight-savings-…>. > > My cell phone (running Android 6.0 Marshmallow, released 2015-10-05) still has the old rules for Asia/Istanbul. My work desktop (Fedora 21) and my home desktop (Ubuntu 15.10) do have the current rules. Other operating system distributions that are up to date include Arch Linux <https://www.archlinux.org/packages/core/any/tzdata/>, CentOS <https://lists.centos.org/pipermail/centos-announce/2015-October/021425.html>, Debian <https://packages.debian.org/sid/tzdata>, Linux From Scratch <http://wiki.linuxfromscratch.org/lfs/ticket/3845>, Oracle Linux <https://oss.oracle.com/pipermail/el-errata/2015-October/005451.html>, Red Hat Enterprise Linux <https://rhn.redhat.com/errata/RHEA-2015-1863.html>, and Scientific Linux <http://ftp.scientificlinux.org/linux/scientific/7.1/x86_64/updates/fastbugs…>. Apologies if I missed your favorite OS distribution and if it is up to date. Obviously many operating systems are not yet updated, though; it's not just Android. > > This weekend's confusion probably deserves a comment, which the attached proposed patch adds. > <0001-europe-Add-comment-re-EEST-in-Turkey.patch>

4 6

Michiana area (Indiana + Michigan)
by Matt Johnson Nov. 2, 2015

Nov. 2, 2015

My wife lived in Granger, Indiana, in the early 1990s, and still has family both there and in Indianapolis. We were discussing the mess that is history of Indiana time zones, and she recalls a period when there was an hour difference when calling between the two cities. I was trying to find it in the history, but it would appear that Granger, being in St. Joseph Country, is considered to be in the America/Indiana/Indianapolis zone - which would preclude them ever being different. >From her recollection, she believes the entire Michiana area (https://en.wikipedia.org/wiki/Michiana) may have been in the same time zone. I've been digging but haven't found any historical reference to back that up. I did see from the comments in the northamerica file that cover two counties in the Michiana area - LaPorte and Starke. LaPorte county is considered by tzdb to be in Western Indiana and covered by America/Chicago, and the Starke county has its own unique history in America/Indiana/Knox. Since both of those are in the Michiana area, is it possible that America/Indiana/Knox is actually applicable in the other counties of Michiana? If so, it would be quite interesting since Michiana includes the Berrien and Cass counties of Michigan, and we don't currently have those separated from America/Detroit. Of course, I'm not suggesting making a change without the appropriate evidence, but I'm wondering if anyone here has come across this in their research? Thanks, Matt

3 2

Re: [tz] America/Santa_Isabel
by Matt Johnson Nov. 2, 2015

Nov. 2, 2015

Acknowledged. Thanks for the clarification. -Matt ________________________________ From: Paul Eggert<mailto:eggert@cs.ucla.edu> Sent: ‎10/‎27/‎2015 10:31 AM To: Matt Johnson<mailto:mj1856@hotmail.com> Cc: Time Zone Mailing List<mailto:tz@iana.org> Subject: Re: [tz] America/Santa_Isabel Matt Johnson wrote: > there is a corresponding entry in Windows, "Pacific Standard Time (Mexico) |(UTC-08:00) Baja California" I'm afraid this will have to be adjusted on the Microsoft end. There needs to be a procedure for removing time zones that turn out to be mistakes. America/Santa_Isabel is a mistake, and I suppose we can view this as an opportunity for ironing out any procedural glitches while removing it. We do have a backward compatibility link on the tzdata side; perhaps that will help for the Microsoft side. > Also, there's a chance that San Quintin could indeed start using Mexican DST rules in 2017, We can create a zone America/San_Quintin if that happens.

2 2

Daylight time: Politics makes for strange time zones
by Brian Inglis Nov. 1, 2015

Nov. 1, 2015

http://www.cbc.ca/news/world/time-zones-daylight-time-1.3296797?cmp=rss Perpetual daylight time, 30-minute offsets are some of the world's time-keeping quirks Talks about Saskatchewan and other zones without DST; Newfoundland and other zones on 00:30 offsets; Europe's adoption of German time during WWII, and desire for a time change in Catalonia back to GMT, like Portugal. Related links in sidebars and body links to http://www.cbc.ca/news/canada/end-of-daylight-saving-time-2015-6-eye-openin… which mentions Atikokan and health effects with a number of sidebar and body links to other articles; looks like it has been updated regularly since Mar 2013. -- Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada

1 0

News: Alaska modifies SB 6, holds working session
by Ray at the Time Zone Report Oct. 31, 2015

Oct. 31, 2015

Just when I thought all DST bills for the 2015 US state legislatures were way past their shelf life. Alaska is revising its SB 6 and holding a working session this morning pm the bill. Details from the AK legislature web site here: http://www.legis.state.ak.us/basis/get_complete_bill.asp?session=29 <http://www.legis.state.ak.us/basis/get_complete_bill.asp?session=29&bill=SB 6> &bill=SB6 There appear to be two provisions of SB 6: 1. Exempt the entire state from DST, regardless of time zone boundaries 2. Petition the US Department of Transportation to move all of Alaska into the Pacific time zone. The effective date listed is January 1, 2017. Plenty of time - if this goes through. Ray ----- Ray Harwood Time Zone Report, a service of GoodClix, Inc. in the interest of public health <http://www.TimeZoneReport.com> http://www.TimeZoneReport.com and Twitter: @TimeZoneReport Office: (480) 378-3221 - Email: Ray(a)TimeZoneReport.com

2 1

News article: Turkey’s plan to ignore daylight saving time has been foiled by smartphones
by Philip Newton Oct. 31, 2015

Oct. 31, 2015

http://qz.com/533126/turkeys-plan-to-ignore-daylight-saving-time-has-been-f… Turkey, like other countries on Eastern European Time (EET), was supposed to turn its clocks back by one hour on Sunday (Oct. 25) for the annual end of daylight saving time—or, as it’s known in Europe, “summer time.” But this year, the Turkish government told citizens to wait until Nov. 1 to allow more hours of light for voting in the nation’s upcoming parliamentary elections, reports the BBC. However, anyone who uses internet- or radio-connected mobile devices or computers to tell time knows that those clocks sync automatically to international standards. Today, Oct. 26, many people in Turkey with the latest Apple, Android, and Windows devices woke up with their clocks an hour slower than the government’s. The hashtag #saatkac (“what’s the time”) is trending among Twitter users in the country. The government has done this before for national elections and university entrance exams, in each case delaying the clock changes by one day. This year, however, Turkey has asked citizens to wait 14 days before turning their clocks back. For the next two weeks, then, the country is on “Erdogan Time,” a national joke and a nod to president Recep Tayyip Erdogan’s paternalistic reputation.

7 6

Re: [tz] timezone change for antarctica\casey [SEC=UNCLASSIFIED]
by Michael Chinn Oct. 31, 2015

Oct. 31, 2015

| Can we now assume, do you think, that Casey will be switching back and forth between +0800 and +1100 in accordance with Australian summer time shifts for the forseeable future (that 3 hour offset change will break assumptions made in some code I expect!) The TZ changes, particularly at Casey, are highly variable, they are based around resupply and flights. They will rarely line up with DST changes, this year there were early flights (leaving tomorrow) and an early insertion of ground crew into Wilkins | What's useful to know is when it was first founded/settled, what times were in use locally, when they altered (as accurately as possible) and what the prediction is for the future. I will endeavour to go through the database and pull out the TZ changes, should be able to get 3-4 years. Regarding predictions, sometimes there are only a couple of days notice on changes much to the anguish of a few wintering expiditioners. All changes are visible live on the external site (it's the same database) ___________________________________________________________________________ Australian Antarctic Division - Commonwealth of Australia IMPORTANT: This transmission is intended for the addressee only. If you are not the intended recipient, you are notified that use or dissemination of this communication is strictly prohibited by Commonwealth law. If you have received this transmission in error, please notify the sender immediately by e-mail or by telephoning +61 3 6232 3209 and DELETE the message. Visit our web site at http://www.antarctica.gov.au/ ___________________________________________________________________________

6 6

timezone database isn't everywhere yet
by Arthur David Olson Oct. 30, 2015

Oct. 30, 2015

I serve on the board of elections for Takoma Park, Maryland, USA.. Early voting in our city election started on 2015-10-28; election day is 2015-11-03. The city uses county-provided electronic poll books to track who has voted. Two were in use on the 28th. I noticed that they were showing times that differed by an hour. I inquired about it; I was told it was deliberate, and was because daylight saving time was in effect when voting started whereas standard time will be in effect when voting ends. If only we had a way for gizmos to automatically switch the clock when daylight saving time ends.-S @dashdashado

1 0

tzdata isstd question
by Ken Murchison Oct. 30, 2015

Oct. 30, 2015

All, Given the following zone: Zone America/Atikokan -6:06:28 - LMT 1895 -6:00 Canada C%sT 1940 Sep 29 -6:00 1:00 CDT 1942 Feb 9 2:00s -6:00 Canada C%sT 1945 Sep 30 2:00 -5:00 - EST I'm wondering why the transition on Feb 9, 1942 isn't flagged as being observed at standard time in the compiled tzfile. Does the time suffix in UNTIL not get applied to the following transition? -- Kenneth Murchison Principal Systems Software Engineer Carnegie Mellon University

4 5

include implementations of str{f,p}time{,_l} in tzcode
by christos＠zoulas.com Oct. 30, 2015

Oct. 30, 2015

Hello, I think that it is an omission for those functions not to be included in tzcode. They are tightly bound to the rest of the time code implementation and although standardized, the implementations that the most popular operating systems provide vary in behavior. Yes, I understand that there are portability considerations (locale for example) that need to be handled, but not including them is taking the easy way out. Best, christos

4 3

timezone change for antarctica\casey [SEC=UNCLASSIFIED]
by Michael Chinn Oct. 30, 2015

Oct. 30, 2015

Hi, Im not too sure on the correct formatting for submitting changes to tzdata, or if the Antarctic tzdata is still being maintained? There is a change for the Australian Antarctic zone Antarctica/Casey went to +1100 for Summer operations on Sunday 04 October at 2am. Source: http://www.antarctica.gov.au/living-and-working/stations/weather and internal emails Also, there is an location for Wilkins Aerodrome that technically can have its own timeone separate from Casey. If it is useful I can get some historical dates? Michael Chinn Information Communications & Technology Section Australian Antartic Division 203 Channel Hwy Kingston TAS 7050 Ph 03 6232 3289 Fax 0362832574 ___________________________________________________________________________ Australian Antarctic Division - Commonwealth of Australia IMPORTANT: This transmission is intended for the addressee only. If you are not the intended recipient, you are notified that use or dissemination of this communication is strictly prohibited by Commonwealth law. If you have received this transmission in error, please notify the sender immediately by e-mail or by telephoning +61 3 6232 3209 and DELETE the message. Visit our web site at http://www.antarctica.gov.au/ ___________________________________________________________________________

2 1

Official definition and origin of the term "Central European Time"
by Even Scharning Oct. 29, 2015

Oct. 29, 2015

Is there an official definition of the term "Central European Time" (CET), specifically which UTC offset(s) it refers to? If there is no official definition, where did the term originate from? I asked the information desk at the EU (Europe Direct) to help me find links to relevant legislature, but their reply was that term CET is "not regulated at a EU level, but internationally". All credible sources I have found, say that CET is a constant UTC+1. The closest thing I have found to an authoritative document, is the German Time Act from 1978. It says: "Legal time is Central European Time. It is defined as Coordinated Universal Time plus one hour." There is no reference to the origin of this definition, and no information of whether anyone other than the Germans agree on this definition. http://www.cl.cam.ac.uk/~mgk25/time/zeitgesetz.en.html On the EU's web pages (for instance http://europa.eu/contact/index_en.htm) times are usually given in CET, not CEST (Central European Summer Time), even though daylight saving time is currently observed. Google gives a few hundred thousand results for CET on europa.eu, and numerous of these results are hits on dates in the daylight saving time period. I think it makes more sense to define CET as alternating between UTC+1 (standard time) and UTC+2 (when DST is observed). (See http://time.is/CET for a more elaborate definition.) This is in line with for instance the US time zones Pacific Time and Eastern Time, which refer to the time currently observed, whether it is standard time or daylight saving time. Even Scharning Time.is - exact time for any time zone http://time.is/

6 6

Re: [tz] America/Santa_Isabel
by Greg Slayden Oct. 27, 2015

Oct. 27, 2015

The largest population center in Baja California State south of Ensenada is San Quintín, 190 km south of Ensenada. A new municipality of that name will separate from the large Ensenada municipality on December 1, 2016, and although there are other large localities such as Lazaro Cardenas in that area, San Quintin seems to be the central administrative town. See https://en.wikipedia.org/wiki/San_Quint%C3%ADn,_Baja_California I see conflicting information online about when DST ends there, either November or October. Making contact with locals in that area is likely the best way to find out

5 11

unsubscribe
by Mehdi.Erroussafi＠inwi.ma Oct. 27, 2015

Oct. 27, 2015

[cid:image001.jpg@01D110B0.3DB0CAE0] Elmehdi Erroussafi Manager Core Voice/Data/Vas Direction opérations et maintenance mobile : 0600 000 140 mail : mehdi.erroussafi(a)inwi.ma<mailto:mehdi.erroussafi@inwi.ma> site : www.inwi.ma<http://www.inwi.ma/> imprimer utile préserve l'environnement ________________________________ « Ce message et toutes les pièces y jointes sont susceptibles de contenir des informations confidentielles ou privilégiées, lesquelles ne doivent être reproduites, diffusées ou exploitées sans autorisation. L'intégrité des messages électroniques n'étant pas garantie, WANA CORPORATE décline toute responsabilité dans le cas où ce message aurait été altéré, déformé ou falsifié. Ce message est établi à l'attention exclusive de ses destinataires. Si vous avez reçu ce message par erreur, veuillez le signaler à l'expéditeur et le détruire y compris les pièces jointes. Merci. » -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- « This message and its attachments may contain confidential or privileged information that should not be copied, distributed or used without authorization. As the integrity of emails may not be guaranteed, WANA CORPORATE is not liable for messages that have been modified, changed or falsified. If you have received this email in error, please notify the sender and delete this message and its attachments. Thank you. »

1 0

Re: [tz] Version in zoneinfo files?
by Paul Ganssle Oct. 25, 2015

Oct. 25, 2015

I don't think anyone's stopping you from using old zic to compile newer data. My understanding is that the release version is in the Makefile of all version-2 tzdata files, so nothing would have to change about the tzdata files themselves. You may well have metadata-incompatible zoneinfo files for some time, however, if compiling them with older versions of zic, but I think there's always going to be some lag between the introduction of a new versioning system and reliability in its adoption. On 10/25/2015 01:51 PM, Tim Parenti wrote: > > On 25 October 2015 at 13:26, Paul G <paul(a)ganssle.io <mailto:paul@ganssle.io>> wrote: > > > I think, as Paul E suggests, keeping the version external is a good choice here. It encourages people to use whatever the platform provides to find out the exact, local version information. If anyone knows how to find out what version of tzdata is installed on this macbook I’d love to know :) > > I think this perfectly illustrates the reason why the versioning should be at the zic level and not downstream. > > > There's also the separate matter that tzcode and tzdata are separate and largely independent. Although there have been a few breaking changes to the data format over the years, in principle, there's little reason someone couldn't use a 2014j zic to compile 2015g data, and I suspect that sort of thing actually happens somewhat often. > > -- > Tim Parenti

1 0

Re: [tz] Version in zoneinfo files?
by Guy Harris Oct. 25, 2015

Oct. 25, 2015

On Oct 25, 2015, at 10:37 AM, Paul G <paul(a)ganssle.io> wrote: > On 10/25/2015 01:35 PM, Guy Harris wrote: > >> So where would zic get the version information to put into the binary files? > > My understanding is that it's the VERSION= line of the Makefile in the tzdata tarballs. I.e., you're proposing that the value of the VERSION value in the Makefile be passed as an argument to zic, and zic would put the value of that argument into a field in the files? So, if the goal is to provide a "reliable way to get the version information in a cross-platform way", what would ensure that this field be set properly, for some value of "properly", everywhere that binary tzdata files are generated? I.e., what is to guarantee that 1) nobody downstream change VERSION if they aren't changing the zic files from what's distributed and 2) people change it "properly" if the files don't correspond to a particular version of files that were distributed (e.g., if some downstream supplier takes 2015q and adds some changes mentioned on the list to it before 2015r comes out with those changes plus other changes, they should call the resulting files something other than "2015q" or "2015r")?

1 0