- tz - lists.iana.org

const cleanup for scheck
by roland＠redhat.com Jan. 10, 2006

Jan. 10, 2006

Please consider including these changes, which reduce some ugly warnings compiling scheck.c for zic. Thanks, Roland Index: private.h =================================================================== RCS file: /cvs/glibc/libc/timezone/private.h,v retrieving revision 1.11 diff -u -r1.11 private.h --- private.h 10 Jan 2006 07:55:37 -0000 1.11 +++ private.h 10 Jan 2006 09:46:12 -0000 @@ -218,7 +218,7 @@ void * irealloc P((void * pointer, int size)); void icfree P((char * pointer)); void ifree P((char * pointer)); -char * scheck P((const char *string, char *format)); +const char *scheck P((const char *string, const char *format)); /* ** Finally, some convenience items. Index: scheck.c =================================================================== RCS file: /cvs/glibc/libc/timezone/scheck.c,v retrieving revision 1.3 diff -u -r1.3 scheck.c --- scheck.c 6 Aug 2005 05:40:57 -0000 1.3 +++ scheck.c 10 Jan 2006 09:46:12 -0000 @@ -8,20 +8,19 @@ #include "private.h" -char * +const char * scheck(string, format) const char * const string; -char * const format; +const char * const format; { register char * fbuf; register const char * fp; register char * tp; register int c; - register char * result; + register const char * result; char dummy; - static char nada; - result = &nada; + result = ""; if (string == NULL || format == NULL) return result; fbuf = imalloc((int) (2 * strlen(format) + 4)); @@ -53,7 +52,7 @@ *tp++ = 'c'; *tp = '\0'; if (sscanf(string, fbuf, &dummy) != 1) - result = (char *) format; + result = format; ifree(fbuf); return result; }

1 0

Re: New Brunswick switches at 00:01, not 02:00; new Zone needed
by Paul Eggert Jan. 2, 2006

Jan. 2, 2006

[I'm sending this on behalf of Clive D.W. Feather, who had problems with sending email to <tz(a)lecserver.nci.nih.gov>.] ----- From: "Clive D.W. Feather" <clive(a)demon.net> Date: Wed, 28 Dec 2005 09:34:14 +0000 Paul Eggert said: > I found a copy of New Brunswick's Time Definition Act consolidated to > 2005-07-15 <http://www.gnb.ca/0062/PDF-acts/t-06.pdf> and it confirms > that New Brunswick switches to and from DST at 00:01 rather than > 02:00. > Page 2 of the above reference says "1965, c.7, s.1; 1973, c.74, s.75; > 1993, c.9, s.1; 2005, c.7, s.83." I assume this refers to dates when > the law changed, It looks to me like references to the legislation making the change. In the UK, the first of these would be "section 1 of chapter 7 of the laws of 1965"; chapter 7 is, in turn, the 7th Public Act passed in that year. Since the cited work says "This Act is consolidated to <date>", I presume it's a publication that applies all the various amending Acts up to that date. Hmm; 2005 c.7 is "An Act Respecting Rural Communities" <http://www.gnb.ca/0062/acts/BBA-2005/Chap-7.pdf>. Section 83 is titled "Time Definition Act" and, found on page 123, reads: 83 Subsection 1(1) of the French version of the Time Definition Act, chapter T-6 of the Revised Statutes, 1973, is amended by striking out "municipal". It looks to me as if your original document cites all the bits of legislation you need to look up to determine the changes. Hmm, on second thoughts, if one change cancels a previous one, the previous one might not be shown. Sorry. -- Clive D.W. Feather | Work: <clive(a)demon.net> | Tel: +44 20 8495 6138 Internet Expert | Home: <clive(a)davros.org> | Fax: +44 870 051 9937 Demon Internet | WWW: http://www.davros.org | Mobile: +44 7973 377646 Thus plc | |

1 0

Leap seconds in the news
by Olson, Arthur David (NIH/NCI) [E] Dec. 27, 2005

Dec. 27, 2005

The Washington Post's Monday, December 26, 2005 edition includes, on page A17, the story "Added Ticktock of the Clock Restarts Time Debate." The story is precipitated by the upcoming leap second. (My guess is that author Guy Gugliotta wrote it well in advance rather than on Christmas Day.) The story is available from http://washingtonpost.com; free registration is required. --ado

2 1

Tzcode2005r.tar.gz and tzdata2005r.tar.gz
by Olson, Arthur David (NIH/NCI) [E] Dec. 27, 2005

Dec. 27, 2005

The files... ftp://elsie.nci.nih.gov/pub/tzcode2005r.tar.gz ...and... ftp://elsie.nci.nih.gov/pub/tzdata2005r.tar.gz ...are now available. Most importantly, these reflect the Canadian changes circulated last week on the time zone mailing list by Paul Eggert. They also add "<pre>" directives to time zone data files and reflect changes to warning message logic in "zdump.c" (but with calls to "gettext" kept unbundled at the suggestion of Ken Pizzini). With providence, that's it for 2005! --ado

1 0

Zdump.c
by Olson, Arthur David (NIH/NCI) [E] Dec. 23, 2005

Dec. 23, 2005

Here are proposed changes to zdump.c, reflecting suggestions from Robbin Kawabata. --ado ------- zdump.c ------- *** /tmp/geta11896 Thu Dec 22 10:20:54 2005 --- /tmp/getb11896 Thu Dec 22 10:20:54 2005 *************** *** 1,4 **** ! static char elsieid[] = "@(#)zdump.c 7.72"; /* ** This code has been made independent of the rest of the time --- 1,4 ---- ! static char elsieid[] = "@(#)zdump.c 7.73"; /* ** This code has been made independent of the rest of the time *************** *** 212,222 **** while (isascii((unsigned char) *cp) && isalpha((unsigned char) *cp)) ++cp; if (cp - abbrp == 0) ! wp = _("lacks alphabetic at start"); ! if (cp - abbrp < 3) ! wp = _("has fewer than 3 alphabetics"); ! if (cp - abbrp > 6) ! wp = _("has more than 6 alphabetics"); if (wp == NULL && (*cp == '+' || *cp == '-')) { ++cp; if (isascii((unsigned char) *cp) && --- 212,222 ---- while (isascii((unsigned char) *cp) && isalpha((unsigned char) *cp)) ++cp; if (cp - abbrp == 0) ! wp = "lacks alphabetic at start"; ! else if (cp - abbrp < 3) ! wp = "has fewer than 3 alphabetics"; ! else if (cp - abbrp > 6) ! wp = "has more than 6 alphabetics"; if (wp == NULL && (*cp == '+' || *cp == '-')) { ++cp; if (isascii((unsigned char) *cp) && *************** *** 223,237 **** isdigit((unsigned char) *cp)) if (*cp++ == '1' && *cp >= '0' && *cp <= '4') ++cp; } - if (*cp != '\0') - wp = _("differs from POSIX standard"); if (wp == NULL) return; (void) fflush(stdout); (void) fprintf(stderr, _("%s: warning: zone \"%s\" abbreviation \"%s\" %s\n"), ! progname, zone, abbrp, wp); warned = TRUE; } --- 223,237 ---- isdigit((unsigned char) *cp)) if (*cp++ == '1' && *cp >= '0' && *cp <= '4') ++cp; + if (*cp != '\0') + wp = "differs from POSIX standard"; } if (wp == NULL) return; (void) fflush(stdout); (void) fprintf(stderr, _("%s: warning: zone \"%s\" abbreviation \"%s\" %s\n"), ! progname, zone, abbrp, _(wp)); warned = TRUE; }

3 2

Timezone mailing list
by Graham Yarbrough Dec. 22, 2005

Dec. 22, 2005

I was given this email as a source of information relating to timezones and time calculation. Was that correct? Could you please advise what is available and how I might obtain it?

1 0

New Brunswick switches at 00:01, not 02:00; new Zone needed
by Paul Eggert Dec. 21, 2005

Dec. 21, 2005

I found a copy of New Brunswick's Time Definition Act consolidated to 2005-07-15 <http://www.gnb.ca/0062/PDF-acts/t-06.pdf> and it confirms that New Brunswick switches to and from DST at 00:01 rather than 02:00. So we need a new Zone. I am thinking of using the name America/Moncton. Moncton has approximately the same population as Saint John, but the latter name is uncomfortably close to the already-existing Zone America/St_Johns. Page 2 of the above reference says "1965, c.7, s.1; 1973, c.74, s.75; 1993, c.9, s.1; 2005, c.7, s.83." I assume this refers to dates when the law changed, which would be useful for tracking down the history of America/Moncton. I did track down an older version of the consolidated document <http://www.canlii.org/nb/laws/sta/t-6/20030127/whole.html> which makes it clear that they haven't changed the rules since 1993 at least, but I can't go earlier than that in the canlii.org database. For dates before 1993 I'll probably defer to Shanks, unless someone else can track it down.

1 0

Manitoba, Ontario, P.E.I., Quebec adopt 2007 US DST change
by Paul Eggert Dec. 21, 2005

Dec. 21, 2005

I looked around for news reports of the effect on the US's 2007 daylight-saving plans on Canada, and found reports that Manitoba, Ontario, P.E.I., and Quebec are committed to switching, and other provinces are thinking of switching. My assumption is that all of Canada (save perhaps Newfoundland) will switch, but that it's too early to regard this as official. One problem is how to reflect this in the tz database. I'm inclined to propose something along the following lines, at least for now, to keep the tables simple; we should be able to fix things before 2007 if some of the provinces decline to follow the US lead. Please see the changes in commentary below to see what provinces would be affected. =================================================================== RCS file: RCS/northamerica,v retrieving revision 2005.15.1.3 diff -pu -r2005.15.1.3 northamerica --- northamerica 2005/12/20 15:04:00 2005.15.1.3 +++ northamerica 2005/12/21 21:00:48 @@ -820,18 +820,37 @@ Zone America/Menominee -5:50:27 - LMT 18 # Unless otherwise specified, the data for Canada are all from Shanks. -# From Paul Eggert (2000-10-02): +# From Paul Eggert (2005-12-21): # H. David Matthews and Mary Vincent's map -# <a href="http://www.canadiangeographic.ca/SO98/geomap.htm"> +# <a href="http://www.canadiangeographic.ca/Magazine/SO98/geomap.asp"> # "It's about TIME", _Canadian Geographic_ (September-October 1998) # </a> contains detailed boundaries for regions observing nonstandard # time and daylight saving time arrangements in Canada circa 1998. # -# INMS, the Institute for National Measurement Standards in Ottawa, has -# <a href="http://www.nrc.ca/inms/time/tze.html"> +# INMS, the Institute for National Measurement Standards in Ottawa, has <a +# href="http://inms-ienm.nrc-cnrc.gc.ca/en/time_services/daylight_saving_e.php"> # information about standard and daylight saving time zones in Canada. # </a> (updated periodically). # Its unofficial information is often taken from Matthews and Vincent. +# +# CBC News reported that Ontario and Manitoba have announced plans to +# follow the US change, and that Nova Scotia is considering it; see +# <http://www.cbc.ca/news/background/daylightsavingtime/> (2005-10-21). +# CBC news also reported that Prince Edward Island is the first +# province in Atlantic Canada to follow the US change, and that Quebec +# had agreed; see <http://www.cbc.ca/pei/story/pe_daylight_20051207.html> +# (2005-12-07). +# +# To reflect all this, the Canada and Winn rules have been adjusted to +# agree with the 2007 US change. This means we assume most of Canada +# will fall into line. However, Alberta, British Columbia, +# Newfoundland, Northwest Territories, and Yukon already have separate +# rules in our database, so for now we'll leave them alone, which +# means that we currently assume these regions will not change their +# rules and will disagree with the US starting in 2007. This +# assumption is probably incorrect, with the possible exception of +# Newfoundland. We plan to adjust the Edm, Vanc, StJohns, and NT_YK +# rules as the corresponding provinces make their announcements. # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S Rule Canada 1918 only - Apr 14 2:00 1:00 D @@ -840,8 +859,10 @@ Rule Canada 1942 only - Feb 9 2:00 1:00 Rule Canada 1945 only - Aug 14 23:00u 1:00 P # Peace Rule Canada 1945 only - Sep 30 2:00 0 S Rule Canada 1974 1986 - Apr lastSun 2:00 1:00 D -Rule Canada 1974 max - Oct lastSun 2:00 0 S -Rule Canada 1987 max - Apr Sun>=1 2:00 1:00 D +Rule Canada 1974 2006 - Oct lastSun 2:00 0 S +Rule Canada 1987 2006 - Apr Sun>=1 2:00 1:00 D +Rule Canada 2007 max - Mar Sun>=8 2:00 1:00 D +Rule Canada 2007 max - Nov Sun>=1 2:00 0 S # Newfoundland (and far southeast Labrador) @@ -1167,11 +1188,13 @@ Rule Winn 1963 only - Apr lastSun 2:00 1 Rule Winn 1963 only - Sep 22 2:00 0 S Rule Winn 1966 1986 - Apr lastSun 2:00 1:00 D Rule Winn 1966 1986 - Oct lastSun 2:00 0 S -Rule Winn 1987 max - Apr Sun>=1 2:00 1:00 D +Rule Winn 1987 2006 - Apr Sun>=1 2:00s 1:00 D # From Paul Eggert (2000-10-02): # INMS (2000-09-12) says that, since 1988 at least, Manitoba switches from # DST at 03:00 local time. For now, assume it started in 1987. -Rule Winn 1987 max - Oct lastSun 2:00s 0 S +Rule Winn 1987 2006 - Oct lastSun 2:00s 0 S +Rule Winn 2007 max - Mar Sun>=8 2:00s 1:00 D +Rule Winn 2007 max - Nov Sun>=1 2:00s 0 S # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone America/Winnipeg -6:28:36 - LMT 1887 Jul 16 -6:00 Winn C%sT

1 0

RE: latest try at backward/northamerica fix
by Olson, Arthur David (NIH/NCI) [E] Dec. 21, 2005

Dec. 21, 2005

I'd suggest making the addition shown below to the commentary; otherwise the proposed change looks great. --ado ------- northamerica ------- *** /tmp/geta893 Tue Dec 20 10:02:12 2005 --- /tmp/getb893 Tue Dec 20 10:02:12 2005 *************** *** 1,4 **** ! # @(#)northamerica 7.85 # <pre> # also includes Central America and the Caribbean --- 1,4 ---- ! # @(#)northamerica 7.86 # <pre> # also includes Central America and the Caribbean *************** *** 136,141 **** --- 136,143 ---- # From Arthur David Olson, 2005-12-19 # We generate the files specified below to guard against old files with # obsolete information being left in the time zone binary directory. + # We limit the list to names that have appeared in previous versions of + # this time zone package. # We do these as separate Zones rather than as Links to avoid problems if # a particular place changes whether it observes DST. # We put these specifications here in the northamerica file both to

2 1

Re: latest try at backward/northamerica fix
by Paul Eggert Dec. 19, 2005

Dec. 19, 2005

Arthur David Olson <olsona(a)elsie.nci.nih.gov> writes: > Here's my lastest try at avoiding problems with zones such as CST5CDT; The patch to 'backward' looks fine, but the patch to 'northamerica' has some problems. First, it removes support for TZ='EST', TZ='MST', TZ='HST', which have been in the tz database since 1987. Second, it creates new Zones AST4ADT, YST9YDT, HST10HDT that are not needed for backwards compatibility, as they have never been supported by the tz data. Third, it creates new Zones AST4, EST5, CST6, MST7, PST8, YST9, HST10 that also are not needed for backward compatibility (for the same reason as above), and furthermore don't change how TZ strings are evaluated (as they are equivalent to their POSIX interpretations). Here's a proposed patch to 'northamerica' to address these issues. --- northamerica 2005/11/28 15:52:28 2005.15 +++ northamerica 2005/12/19 21:52:52 2005.15.1.2 @@ -131,6 +131,24 @@ Rule US 1987 2006 - Apr Sun>=1 2:00 1:00 Rule US 2007 max - Mar Sun>=8 2:00 1:00 D Rule US 2007 max - Nov Sun>=1 2:00 0 S +# From Arthur David Olson, 2005-12-19 +# We generate the files specified below to guard against old files with +# obsolete information being left in the time zone binary directory. +# We do these as separate Zones rather than as Links to avoid problems if +# a particular place changes whether it observes DST. +# We put these specifications here in the northamerica file both to +# increase the chances that they'll actually get compiled and to +# avoid the need to duplicate the US rules in another file. + +# Zone NAME GMTOFF RULES FORMAT [UNTIL] +Zone EST -5:00 - EST +Zone MST -7:00 - MST +Zone HST -10:00 - HST +Zone EST5EDT -5:00 US E%sT +Zone CST6CDT -6:00 US C%sT +Zone MST7MDT -7:00 US M%sT +Zone PST8PDT -8:00 US P%sT + # From Bob Devine (1988-01-28): # ...Alaska (and Hawaii) had the timezone names changed in 1967. # old new

1 0