Module Name: src
Committed By: christos
Date: Sat Oct 27 22:29:24 UTC 2018
Modified Files:
src/lib/libc/time: NEWS theory.html tz-how-to.html tz-link.html
tzfile.5 version zdump.8 zic.8 zic.c
Log Message:
Welcome tzcode-2018g
Changes to code
When generating TZif files with leap seconds, zic no longer uses a
format that trips up older 32-bit clients, fixing a bug introduced
in 2018f. (Reported by Daniel Fischer.) Also, the zic workaround
for QTBUG-53071 now also works for TZif files with leap seconds.
The translator to rearguard format now rewrites the line
"Rule Japan 1948 1951 - Sep Sat>=8 25:00 0 S" to
"Rule Japan 1948 1951 - Sep Sun>=9 1:00 0 S".
This caters to zic before 2007 and to Oracle TZUpdater 2.2.0
and earlier. (Reported by Christos Zoulas.)
Changes to documentation
tzfile.5 has new sections on interoperability issues.
To generate a diff of this commit:
cvs rdiff -u -r1.25 -r1.26 src/lib/libc/time/NEWS
cvs rdiff -u -r1.4 -r1.5 src/lib/libc/time/theory.html
cvs rdiff -u -r1.5 -r1.6 src/lib/libc/time/tz-how-to.html
cvs rdiff -u -r1.2 -r1.3 src/lib/libc/time/tz-link.html
cvs rdiff -u -r1.26 -r1.27 src/lib/libc/time/tzfile.5
cvs rdiff -u -r1.8 -r1.9 src/lib/libc/time/version
cvs rdiff -u -r1.18 -r1.19 src/lib/libc/time/zdump.8
cvs rdiff -u -r1.30 -r1.31 src/lib/libc/time/zic.8
cvs rdiff -u -r1.71 -r1.72 src/lib/libc/time/zic.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
Modified files:
Index: src/lib/libc/time/NEWS
diff -u src/lib/libc/time/NEWS:1.25 src/lib/libc/time/NEWS:1.26
--- src/lib/libc/time/NEWS:1.25 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/NEWS Sat Oct 27 18:29:24 2018
@@ -1,5 +1,40 @@
News for the tz database
+Release 2018g - 2018-10-26 22:22:45 -0700
+
+ Briefly:
+ Morocco switches to permanent +01 on 2018-10-27.
+
+ Changes to future timestamps
+
+ Morocco switches from +00/+01 to permanent +01 effective 2018-10-27,
+ so its clocks will not fall back on 2018-10-28 as previously scheduled.
+ (Thanks to Mohamed Essedik Najd and Brian Inglis.)
+
+ Changes to code
+
+ When generating TZif files with leap seconds, zic no longer uses a
+ format that trips up older 32-bit clients, fixing a bug introduced
+ in 2018f. (Reported by Daniel Fischer.) Also, the zic workaround
+ for QTBUG-53071 now also works for TZif files with leap seconds.
+
+ The translator to rearguard format now rewrites the line
+ "Rule Japan 1948 1951 - Sep Sat>=8 25:00 0 S" to
+ "Rule Japan 1948 1951 - Sep Sun>=9 1:00 0 S".
+ This caters to zic before 2007 and to Oracle TZUpdater 2.2.0
+ and earlier. (Reported by Christos Zoulas.)
+
+ Changes to past time zone abbreviations
+
+ Change HDT to HWT/HPT for WWII-era abbreviations in Hawaii.
+ This reverts to 2011h, as the abbreviation change in 2011i was
+ likely inadvertent.
+
+ Changes to documentation
+
+ tzfile.5 has new sections on interoperability issues.
+
+
Release 2018f - 2018-10-18 00:14:18 -0700
Briefly:
Index: src/lib/libc/time/theory.html
diff -u src/lib/libc/time/theory.html:1.4 src/lib/libc/time/theory.html:1.5
--- src/lib/libc/time/theory.html:1.4 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/theory.html Sat Oct 27 18:29:24 2018
@@ -407,7 +407,7 @@ in decreasing order of importance:
EST/EDT/EWT/EPT/EDDT Eastern [North America],
EET/EEST Eastern European,
GST Guam,
- HST/HDT Hawaii,
+ HST/HDT/HWT/HPT Hawaii,
HKT/HKST Hong Kong,
IST India,
IST/GMT Irish,
Index: src/lib/libc/time/tz-how-to.html
diff -u src/lib/libc/time/tz-how-to.html:1.5 src/lib/libc/time/tz-how-to.html:1.6
--- src/lib/libc/time/tz-how-to.html:1.5 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/tz-how-to.html Sat Oct 27 18:29:24 2018
@@ -459,8 +459,8 @@ given alphabetic string.</li>
<p>An example of a specific amount of time is:</p>
<pre>
#Zone NAME GMTOFF RULES FORMAT [UNTIL]
-Zone Pacific/Honolulu ... 1933 Apr 30 2:00
- -10:30 1:00 HDT 1933 May 21 2:00
+Zone Pacific/Honolulu ... 1933 Apr 30 2:00
+ -10:30 1:00 HDT 1933 May 21 12:00
...
</pre>
@@ -559,7 +559,7 @@ the true offset is undefined.
Rule US 1918 1919 - Oct lastSun 2:00 0 S
Rule US 1942 only - Feb 9 2:00 1:00 W # War
Rule US 1945 only - Aug 14 23:00u 1:00 P # Peace
-Rule US 1945 only - Sep 30 2:00 0 S
+Rule US 1945 only - Sep lastSun 2:00 0 S
</pre>
</td></tr></table></td>
</tr>
@@ -570,7 +570,7 @@ Rule US 1945 only - Sep 30 2
<td colspan="6" align="center"><table><tr><td>
<pre>
#Zone NAME GMTOFF RULES FORMAT [UNTIL]
-Zone Pacific/Honolulu -10:31:26 - LMT 1900 Jan 1 12:00
+Zone Pacific/Honolulu -10:31:26 - LMT 1896 Jan 13 12:00
-10:30 - HST 1933 Apr 30 2:00
-10:30 1:00 HDT 1933 May 21 2:00
-10:30 US H%sT 1947 Jun 8 2:00
@@ -598,7 +598,7 @@ Zone Pacific/Honolulu -10:31:26 - LM
<td>—</td>
<td>LMT</td>
<td>local mean time</td>
- <td>1900-01-01</td>
+ <td>1896-01-13</td>
<td>12:00</td>
</tr>
<tr align="center">
@@ -607,7 +607,7 @@ Zone Pacific/Honolulu -10:31:26 - LM
<td>HST</td>
<td>Hawaii standard time</td>
<td>1933-04-30</td>
- <td rowspan="3">02:00</td>
+ <td>02:00</td>
</tr>
<tr align="center">
<td>−9:30</td>
@@ -615,6 +615,7 @@ Zone Pacific/Honolulu -10:31:26 - LM
<td>HDT</td>
<td>Hawaii daylight time</td>
<td>1933-05-21</td>
+ <td>12:00</td>
</tr>
<tr align="center">
<td>−10:30¹</td>
@@ -622,6 +623,7 @@ Zone Pacific/Honolulu -10:31:26 - LM
<td>HST¹</td>
<td>Hawaii standard time</td>
<td>1942-02-09</td>
+ <td>02:00</td>
</tr>
<tr align="center">
<td rowspan="2">−9:30</td>
Index: src/lib/libc/time/tz-link.html
diff -u src/lib/libc/time/tz-link.html:1.2 src/lib/libc/time/tz-link.html:1.3
--- src/lib/libc/time/tz-link.html:1.2 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/tz-link.html Sat Oct 27 18:29:24 2018
@@ -524,10 +524,11 @@ clock (intclock)</a> is a clock that dis
under the <abbr>GPL</abbr>.</li>
<li>Microsoft Windows 8.1
and later has <code><abbr>tz</abbr></code> data and <abbr>CLDR</abbr>
-data (mentioned <a href="#CLDR">below</a>) used by
-<a href="https://en.wikipedia.org/wiki/Windows_Runtime";>Windows Runtime</a>
-classes such as <a
-href="https://msdn.microsoft.com/en-us/library/windows/apps/windows.globalization.datetimeformatting.datetimeformatter.aspx";><code>DateTimeFormatter</code></a>.
+data (mentioned <a href="#CLDR">below</a>) used by the
+<a href="https://en.wikipedia.org/wiki/Windows_Runtime";>Windows Runtime</a> /
+<a href="https://en.wikipedia.org/wiki/Universal_Windows_Platform";>Universal Windows Platform</a> classes
+<a href="https://docs.microsoft.com/uwp/api/Windows.Globalization.DateTimeFormatting.DateTimeFormatter";><code>DateTimeFormatter</code></a> and
+<a href="https://docs.microsoft.com/uwp/api/windows.globalization.calendar";><code>Calendar</code></a>.
<a id="System.TimeZoneInfo"
href="https://blogs.msdn.microsoft.com/bclteam/2007/06/07/exploring-windows-time-zones-with-system-timezoneinfo-josh-free/";>Exploring
Windows Time Zones with <code>System.TimeZoneInfo</code></a> describes
@@ -540,6 +541,8 @@ Tzid table</a> or <a
href="https://unicode.org/repos/cldr/trunk/common/supplemental/windowsZones.xml";><abbr>XML</abbr>
file</a> of the <abbr>CLDR</abbr> data maps proprietary zone IDs
to <code><abbr>tz</abbr></code> names.
+These mappings can be performed programmatically via the <a href="https://github.com/mj1856/TimeZoneConverter";>TimeZoneConverter</a> .NET library,
+or the ICU Java and C++ libraries mentioned <a href="#ICU">above</a>.
<li><a
href="https://www.oracle.com/java/index.html";>Oracle
Java</a> contains a copy of a subset of a recent
@@ -742,6 +745,14 @@ zones & daylight saving time</a>.</d
<dd>The Hydrographic and Oceanographic Service of the Chilean Navy publishes a
<a href="http://www.horaoficial.cl/historia_hora.html"; hreflang="es">history of
Chile's official time (in Spanish)</a>.</dd>
+<dt>China</dt>
+<dd>The Hong Kong Observatory maintains a
+<a href="https://www.hko.gov.hk/gts/time/Summertime.htm";>history of
+ summer time in Hong Kong</a>,
+and Macau's Meteorological and Geophysical Bureau maintains a <a
+href="http://www.smg.gov.mo/smg/geophysics/e_t_Summer%20Time.htm";>similar
+history for Macau</a>.
+Unfortunately the latter is incomplete and has errors.</dd>
<dt>Czech Republic</dt>
<dd><a href="https://kalendar.beda.cz/kdy-zacina-a-konci-letni-cas";
hreflang="cs">When daylight saving time starts and ends (in Czech)</a>
@@ -758,10 +769,6 @@ hreflang="he">announcements (in Hebrew)<
<dd>The National Institute of Metrological Research maintains a
<a href="http://oldsite.inrim.it/res/tf/ora_legale_i.shtml";>table of civil time
(in Italian)</a>.</dd>
-<dt>Macau</dt>
-<dd>The Meteorological and Geophysical Bureau maintains a
-<a href="http://www.smg.gov.mo/smg/geophysics/e_t_Summer%20Time.htm";>history
-of summer time</a>. Unfortunately it is incomplete and has errors.</dd>
<dt>Malaysia</dt>
<dd>See Singapore <a href="#Singapore">below</a>.</dd>
<dt>Mexico</dt>
Index: src/lib/libc/time/tzfile.5
diff -u src/lib/libc/time/tzfile.5:1.26 src/lib/libc/time/tzfile.5:1.27
--- src/lib/libc/time/tzfile.5:1.26 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/tzfile.5 Sat Oct 27 18:29:24 2018
@@ -1,8 +1,8 @@
-.\" $NetBSD: tzfile.5,v 1.26 2018/10/19 23:05:35 christos Exp $
+.\" $NetBSD: tzfile.5,v 1.27 2018/10/27 22:29:24 christos Exp $
.\"
.\" This file is in the public domain, so clarified as of
.\" 1996-06-05 by Arthur David Olson ([email protected]).
-.Dd October 19, 2018
+.Dd October 27, 2018
.Dt TZFILE 5
.Os
.Sh NAME
@@ -174,8 +174,202 @@ from 0 through 24.
Second, DST is in effect all year if it starts
January 1 at 00:00 and ends December 31 at 24:00 plus the difference
between daylight saving and standard time.
+.Ss Interoperability considerations
.Pp
+Version 1 files are considered a legacy format and
+should be avoided, as they do not support transition
+times after the year 2038.
+Readers that only understand Version 1 must ignore
+any data that extends beyond the calculated end of the version
+1 data block.
+.PP
+Writers should generate a version 3 file if
+TZ string extensions are necessary to accurately
+model transition times.
+Otherwise, version 2 files should be generated.
+.Pp
+The sequence of time changes defined by the version 1
+header and data block should be a contiguous subsequence
+of the time changes defined by the version 2+ header and data
+block, and by the footer.
+This guideline helps obsolescent version 1 readers
+agree with current readers about timestamps within the
+contiguous subsequence. It also lets writers not
+supporting obsolescent readers use a
+.Dv tzh_timecnt
+of zero
+in the version 1 data block to save space.
+.Pp
+Time zone designations should consist of at least three (3)
+and no more than six (6) ASCII characters from the set of
+alphanumerics,
+.Dq \&- ,
+and
+.Dq + .
+This is for compatibility with POSIX requirements for
+time zone abbreviations.
+.Pp
+When reading a version 2 or 3 file, readers
+should ignore the version 1 header and data block except for
+the purpose of skipping over them.
+.Pp
+Readers should calculate the total lengths of the
+headers and data blocks and check that they all fit within
+the actual file size, as part of a validity check for the file.
+.Ss Common interoperability issues
+This section documents common problems in reading or writing TZif files.
+Most of these are problems in generating TZif files for use by
+older readers.
+The goals of this section are:
+.Bl -bullet
+.It
+to help TZif writers output files that avoid common
+pitfalls in older or buggy TZif readers,
+.It
+to help TZif readers avoid common pitfalls when reading
+files generated by future TZif writers, and
+.It
+to help any future specification authors see what sort of
+problems arise when the TZif format is changed.
+.El
+.Pp
++When new versions of the TZif format have been defined, a
+design goal has been that a reader can successfully use a TZif
+file even if the file is of a later TZif version than what the
+reader was designed for.
+When complete compatibility was not achieved, an attempt was
+made to limit glitches to rarely-used timestamps, and to allow
+simple partial workarounds in writers designed to generate
+new-version data useful even for older-version readers.
+This section attempts to document these compatibility issues and
+workarounds, as well as to document other common bugs in
+readers.
+.Pp
+Interoperability problems with TZif include the following:
+.Bl -bullet
+.It
+Some readers examine only version 1 data.
+As a partial workaround, a writer can output as much version 1
+data as possible.
+However, a reader should ignore version 1 data, and should use
+version 2+ data even if the reader's native timestamps have only
+32 bits.
+.It
+Some readers designed for version 2 might mishandle
+timestamps after a version 3 file's last transition, because
+they cannot parse extensions to POSIX in the TZ-like string.
+As a partial workaround, a writer can output more transitions
+than necessary, so that only far-future timestamps are
+mishandled by version 2 readers.
+.It
+Some readers designed for version 2 do not support
+permanent daylight saving time, e.g., a TZ string
+.Dq EST5EDT,0/0,J365/25
+denoting permanent Eastern Daylight Time (\&-04).
+As a partial workaround, a writer can substitute standard time
+for the next time zone east, e.g.,
+.Dq AST4
++for permanent Atlantic Standard Time (\&-04).
+.It
+Some readers ignore the footer, and instead predict future
+timestamps from the time type of the last transition.
+As a partial workaround, a writer can output more transitions
+than necessary.
+.It
+Some readers do not use time type 0 for timestamps before
+the first transition, in that they infer a time type using a
+heuristic that does not always select time type 0.
+As a partial workaround, a writer can output a dummy (no-op)
+first transition at an early time.
+.It
+Some readers mishandle timestamps before the first
+transition that has a timestamp not less than -2**31.
+Readers that support only 32-bit timestamps are likely to be
+more prone to this problem, for example, when they process
+64-bit transitions only some of which are representable in 32
+bits.
+As a partial workaround, a writer can output a dummy
+transition at timestamp \&-2**31.
+.It
+Some readers mishandle a transition if its timestamp has
+the minimum possible signed 64-bit value.
+Timestamps less than \&-2**59 are not recommended.
+.It
+Some readers mishandle POSIX-style TZ strings that
+contain
+.Dq <
+or
+.Dq > .
+As a partial workaround, a writer can avoid using
+.Dq <
+or
+.Dq >
+for time zone abbreviations containing only alphabetic
+characters.
+.Pp
+Many readers mishandle time zone abbreviations that contain
+non-ASCII characters.
+These characters are not recommended.
+.Pp
+Some readers may mishandle time zone abbreviations that
+contain fewer than 3 or more than 6 characters, or that
+contain ASCII characters other than alphanumerics,
+.Dq \&- .
+and
+.Dq + .
+These abbreviations are not recommended.
+.It
+Some readers mishandle TZif files that specify
+daylight-saving time UT offsets that are less than the UT
+offsets for the corresponding standard time.
+These readers do not support locations like Ireland, which
+uses the equivalent of the POSIX TZ string
+.Dq IST\&-1GMT0,M10.5.0,M3.5.0/1 ,
+observing standard time
+(IST, +01) in summer and daylight saving time (GMT, +00) in winter.
+As a partial workaround, a writer can output data for the
+equivalent of the POSIX TZ string
+.Dq GMT0IST,M3.5.0/1,M10.5.0 ,
+thus swapping standard and daylight saving time.
+Although this workaround misidentifies which part of the year
+uses daylight saving time, it records UT offsets and time zone
+abbreviations correctly.
+.El
+.Pp
+Some interoperability problems are reader bugs that
+are listed here mostly as warnings to developers of readers.
+.Bl -bullet
+.It
+Some readers do not support negative timestamps.
+Developers of distributed applications should keep this
+in mind if they need to deal with pre-1970 data.
+.It
+Some readers mishandle timestamps before the first
+transition that has a nonnegative timestamp.
+Readers that do not support negative timestamps are likely to
+be more prone to this problem.
+.It
++Some readers mishandle time zone abbreviations like
+.Dq \&-08
+that contain
+.Dq + ,
+.Dq \&- ,
+or digits.
+.It
+Some readers mishandle UT offsets that are out of the
+traditional range of \*-12 through +12 hours, and so do not
+support locations like Kiritimati that are outside this
+range.
+.It
+Some readers mishandle UT offsets in the range [\*-3599, \*-1]
+seconds from UT, because they integer-divide the offset by
+3600 to get 0 and then display the hour part as
+.Dq +00 .
+.It
+Some readers mishandle UT offsets that are not a multiple
+of one hour, or of 15 minutes, or of 1 minute.
Future changes to the format may append more data.
+.El
.Sh SEE ALSO
.Xr ctime 3 ,
.Xr localtime 3 ,
Index: src/lib/libc/time/version
diff -u src/lib/libc/time/version:1.8 src/lib/libc/time/version:1.9
--- src/lib/libc/time/version:1.8 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/version Sat Oct 27 18:29:24 2018
@@ -1 +1 @@
-2018f
+2018g
Index: src/lib/libc/time/zdump.8
diff -u src/lib/libc/time/zdump.8:1.18 src/lib/libc/time/zdump.8:1.19
--- src/lib/libc/time/zdump.8:1.18 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/zdump.8 Sat Oct 27 18:29:24 2018
@@ -1,5 +1,5 @@
-.\" $NetBSD: zdump.8,v 1.18 2018/10/19 23:05:35 christos Exp $
-.Dd October 19, 2018
+.\" $NetBSD: zdump.8,v 1.19 2018/10/27 22:29:24 christos Exp $
+.Dd October 27, 2018
.Dt ZDUMP 8
.Os
.Sh NAME
@@ -147,13 +147,14 @@ tabbed columns line up.)
.Bd -literal
TZ="Pacific/Honolulu"
.Ed
-.Bl -column "XXXX-XX-XX" "HH:MM:SS" "-HH:MM:SS" "TZT" "X" -compact
-.It - Ta - Ta -10:31:26 Ta LMT Ta
-.It 1896-01-13 Ta 12:01:26 Ta -10:30 Ta HST Ta
-.It 1933-04-30 Ta 03 Ta -09:30 Ta HDT Ta 1
-.It 1933-05-21 Ta 11 Ta -10:30 Ta HST Ta
-.It 1942-02-09 Ta 03 Ta -09:30 Ta HDT Ta 1
-.It 1945-09-30 Ta 01 Ta -10:30 Ta HST Ta
+.Bl -column "XXXX-XX-XX" "HH:MM:SS" "-HHMMSS" "TZT" "X" -compact
+.It - Ta - Ta -103126 Ta LMT Ta
+.It 1896-01-13 Ta 12:01:26 Ta -1030 Ta HST Ta
+.It 1933-04-30 Ta 03 Ta -0930 Ta HDT Ta 1
+.It 1933-05-21 Ta 11 Ta -1030 Ta HST Ta
+.It 1942-02-09 Ta 03 Ta -0930 Ta HDT Ta 1
+.It 1945-08-14 Ta 13:30 Ta -0930 Ta HPT Ta 1
+.It 1945-09-30 Ta 01 Ta -1030 Ta HST Ta
.It 1947-06-08 Ta 02:30 Ta -10 Ta HST Ta
.El
.Pp
@@ -174,7 +175,7 @@ Here are excerpts from another example:
TZ="Europe/Astrakhan"
.Ed
.Bl -column "XXXX-XX-XX" "HH:MM:SS" "-HH:MM:SS" "TZT" "X" -compact
-.It - Ta - Ta +03:12:12 Ta LMT Ta
+.It - Ta - Ta +031212 Ta LMT Ta
.It 1924-04-30 Ta 23:47:48 Ta +03 Ta Ta
.It 1930-06-21 Ta 01 Ta +04 Ta Ta
.It 1981-04-01 Ta 01 Ta +05 Ta Ta 1
Index: src/lib/libc/time/zic.8
diff -u src/lib/libc/time/zic.8:1.30 src/lib/libc/time/zic.8:1.31
--- src/lib/libc/time/zic.8:1.30 Fri Oct 19 19:05:35 2018
+++ src/lib/libc/time/zic.8 Sat Oct 27 18:29:24 2018
@@ -1,4 +1,4 @@
-.\" $NetBSD: zic.8,v 1.30 2018/10/19 23:05:35 christos Exp $
+.\" $NetBSD: zic.8,v 1.31 2018/10/27 22:29:24 christos Exp $
.Dd October 19, 2018
.Dt ZIC 8
.Os
@@ -295,6 +295,10 @@ or
if the given time is universal time;
in the absence of an indicator,
wall clock time is assumed.
+These forms ignore leap seconds; for example,
+if a leap second occurs at 00:59:60 local time,
+.q "1:00"
+stands for 3601 seconds after local midnight instead of the usual 3600 seconds.
The intent is that a rule line describes the instants when a
clock/calendar set to the type of time specified in the
.Em AT
Index: src/lib/libc/time/zic.c
diff -u src/lib/libc/time/zic.c:1.71 src/lib/libc/time/zic.c:1.72
--- src/lib/libc/time/zic.c:1.71 Fri May 4 11:51:00 2018
+++ src/lib/libc/time/zic.c Sat Oct 27 18:29:24 2018
@@ -1,4 +1,4 @@
-/* $NetBSD: zic.c,v 1.71 2018/05/04 15:51:00 christos Exp $ */
+/* $NetBSD: zic.c,v 1.72 2018/10/27 22:29:24 christos Exp $ */
/*
** This file is in the public domain, so clarified as of
** 2006-07-17 by Arthur David Olson.
@@ -10,7 +10,7 @@
#include <sys/cdefs.h>
#ifndef lint
-__RCSID("$NetBSD: zic.c,v 1.71 2018/05/04 15:51:00 christos Exp $");
+__RCSID("$NetBSD: zic.c,v 1.72 2018/10/27 22:29:24 christos Exp $");
#endif /* !defined lint */
#include "private.h"
@@ -73,6 +73,11 @@ typedef int_fast64_t zic_t;
static ptrdiff_t const PTRDIFF_MAX = MAXVAL(ptrdiff_t, TYPE_BIT(ptrdiff_t));
#endif
+/* The minimum alignment of a type, for pre-C11 platforms. */
+#if __STDC_VERSION__ < 201112
+# define _Alignof(type) offsetof(struct { char a; type b; }, b)
+#endif
+
/* The type for line numbers. Use PRIdMAX to format them; formerly
there was also "#define PRIdLINENO PRIdMAX" and formats used
PRIdLINENO, but xgettext cannot grok that. */
@@ -199,7 +204,9 @@ enum { PERCENT_Z_LEN_BOUND = sizeof "+99
QTBUG-53071 <https://bugreports.qt.io/browse/QTBUG-53071>. This
workaround will no longer be needed when Qt 5.6.1 and earlier are
obsolete, say in the year 2021. */
+#ifndef WORK_AROUND_QTBUG_53071
enum { WORK_AROUND_QTBUG_53071 = 1 };
+#endif
static int charcnt;
static bool errors;
@@ -441,6 +448,16 @@ size_product(size_t nitems, size_t items
return nitems * itemsize;
}
+static ATTRIBUTE_PURE size_t
+align_to(size_t size, size_t alignment)
+{
+ size_t aligned_size = size + alignment - 1;
+ aligned_size -= aligned_size % alignment;
+ if (aligned_size < size)
+ memory_exhausted(_("alignment overflow"));
+ return aligned_size;
+}
+
#if !HAVE_STRDUP
static char *
strdup(char const *str)
@@ -1761,12 +1778,6 @@ atcomp(const void *avp, const void *bvp)
return (a < b) ? -1 : (a > b);
}
-static bool
-is32(const zic_t x)
-{
- return INT32_MIN <= x && x <= INT32_MAX;
-}
-
static void
writezone(const char *const name, const char *const string, char version)
{
@@ -1781,7 +1792,11 @@ writezone(const char *const name, const
zic_t one = 1;
zic_t y2038_boundary = one << 31;
ptrdiff_t nats = timecnt + WORK_AROUND_QTBUG_53071;
- zic_t *ats = zic_malloc(size_product(nats, sizeof *ats + 1));
+
+ /* Allocate the ATS and TYPES arrays via a single malloc,
+ as this is a bit faster. */
+ zic_t *ats = zic_malloc(align_to(size_product(nats, sizeof *ats + 1),
+ _Alignof(zic_t)));
void *typesptr = ats + nats;
unsigned char *types = typesptr;
@@ -1834,18 +1849,6 @@ writezone(const char *const name, const
types[i] = attypes[i].type;
}
- /* Work around QTBUG-53071 for time stamps less than y2038_boundary - 1,
- by inserting a no-op transition at time y2038_boundary - 1.
- This works only for timestamps before the boundary, which
- should be good enough in practice as QTBUG-53071 should be
- long-dead by 2038. */
- if (WORK_AROUND_QTBUG_53071 && timecnt != 0
- && ats[timecnt - 1] < y2038_boundary - 1 && strchr(string, '<')) {
- ats[timecnt] = y2038_boundary - 1;
- types[timecnt] = types[timecnt - 1];
- timecnt++;
- }
-
/*
** Correct for leap seconds.
*/
@@ -1857,6 +1860,22 @@ writezone(const char *const name, const
break;
}
}
+
+ /* Work around QTBUG-53071 for timestamps less than y2038_boundary - 1,
+ by inserting a no-op transition at time y2038_boundary - 1.
+ This works only for timestamps before the boundary, which
+ should be good enough in practice as QTBUG-53071 should be
+ long-dead by 2038. Do this after correcting for leap
+ seconds, as the idea is to insert a transition just before
+ 32-bit time_t rolls around, and this occurs at a slightly
+ different moment if transitions are leap-second corrected. */
+ if (WORK_AROUND_QTBUG_53071 && timecnt != 0
+ && ats[timecnt - 1] < y2038_boundary - 1 && strchr(string, '<')) {
+ ats[timecnt] = y2038_boundary - 1;
+ types[timecnt] = types[timecnt - 1];
+ timecnt++;
+ }
+
/*
** Figure out 32-bit-limited starts and counts.
*/
@@ -1864,22 +1883,22 @@ writezone(const char *const name, const
timei32 = 0;
leapcnt32 = leapcnt;
leapi32 = 0;
- while (timecnt32 > 0 && !is32(ats[timecnt32 - 1]))
+ while (0 < timecnt32 && INT32_MAX < ats[timecnt32 - 1])
--timecnt32;
- while (timecnt32 > 0 && !is32(ats[timei32])) {
+ while (1 < timecnt32 && ats[timei32] < INT32_MIN
+ && ats[timei32 + 1] <= INT32_MIN) {
+ /* Discard too-low transitions, except keep any last too-low
+ transition if no transition is exactly at INT32_MIN.
+ The kept transition will be output as an INT32_MIN
+ "transition" appropriate for buggy 32-bit clients that do
+ not use time type 0 for timestamps before the first
+ transition; see below. */
--timecnt32;
++timei32;
}
- /*
- ** Output an INT32_MIN "transition" if appropriate; see below.
- */
- if (timei32 > 0 && ats[timei32] > INT32_MIN) {
- --timei32;
- ++timecnt32;
- }
- while (leapcnt32 > 0 && !is32(trans[leapcnt32 - 1]))
+ while (0 < leapcnt32 && INT32_MAX < trans[leapcnt32 - 1])
--leapcnt32;
- while (leapcnt32 > 0 && !is32(trans[leapi32])) {
+ while (0 < leapcnt32 && trans[leapi32] < INT32_MIN) {
--leapcnt32;
++leapi32;
}
