[jdk8u] RFR: 8339637: (tz) Update Timezone Data to 2024b [v2]
Andrew John Hughes
andrew at openjdk.org
Thu Mar 6 01:51:57 UTC 2025
On Wed, 5 Mar 2025 21:46:24 GMT, Dmitry Cherepanov <dcherepanov at openjdk.org> wrote:
>> jdk/test/java/util/TimeZone/TimeZoneData/aliases.txt line 56:
>>
>>> 54: Link America/Mexico_City Mexico/General
>>> 55: Link Pacific/Auckland NZ
>>> 56: Link Asia/Ulaanbaatar Asia/Choibalsan
>>
>> I have found one suspicious change. For the current release there are a few modifications were reverted after the 'make rearguard_tarballs' except for the line above. However, I believe the following change should not be reverted:
>>> -- Link Asia/Urumqi Antarctica/Vostok
>>
>> It was part of 2023d, see change in the [tzdata/backward
>> ](https://github.com/openjdk/jdk8u/commit/15e2c0b58c74e723d7686abc6ec244351508aaf6#diff-b903e3514a1322a34f2e2bac40a50dd97bedb9fa7e38ba2d91ff79ae245aefcbL231)
>>
>> For example aliases.txt in [jdk21u-dev](https://github.com/openjdk/jdk21u-dev/commit/d5b8c283ba6d58eb4c3af9709c7c3ad60dc11366#diff-489ad8ece4848331545099e194ea7c84998bdad0888877ba5b55a926cebdfa23L151) was updated properly.
>
> I regenerated aliases.txt using [Makefile](https://github.com/openjdk/jdk8u/blob/master/jdk/test/java/util/TimeZone/tools/share/Makefile) and this also created aliases.txt with the alias removed. Updated aliases.txt. Thanks.
Thanks for catching this. It looks like an omission in the 8u backport of 2023d (https://github.com/openjdk/jdk8u-dev/pull/414) rather than an issue with the initial version of this patch. I didn't review that one at the time, but doing so retroactively the omission is clear from `diffstat` but difficult to spot in a `diff`:
~~~
$ diffstat -p0 ${TMPDIR}/8322725.11u
b/make/data/tzdata/VERSION | 2 +-
b/make/data/tzdata/africa | 7 -------
b/make/data/tzdata/antarctica | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++++++--
b/make/data/tzdata/asia | 6 ++++--
b/make/data/tzdata/australasia | 8 +++++++-
b/make/data/tzdata/backward | 1 -
b/make/data/tzdata/europe | 29 ++++++++++++++++++++++-------
b/make/data/tzdata/iso3166.tab | 17 +++++++++++------
b/make/data/tzdata/leapseconds | 8 ++++----
b/make/data/tzdata/northamerica | 3 ++-
b/make/data/tzdata/southamerica | 6 ++++++
b/make/data/tzdata/zone.tab | 24 ++++++++++++------------
b/test/jdk/java/util/TimeZone/TimeZoneData/VERSION | 2 +-
b/test/jdk/java/util/TimeZone/TimeZoneData/aliases.txt | 1 -
14 files changed, 125 insertions(+), 46 deletions(-)
$ diffstat -p0 ${TMPDIR}/8322725.8u
b/jdk/make/data/tzdata/VERSION | 2 +-
b/jdk/make/data/tzdata/africa | 7 -------
b/jdk/make/data/tzdata/antarctica | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++++++--
b/jdk/make/data/tzdata/asia | 6 ++++--
b/jdk/make/data/tzdata/australasia | 8 +++++++-
b/jdk/make/data/tzdata/backward | 1 -
b/jdk/make/data/tzdata/europe | 29 ++++++++++++++++++++++-------
b/jdk/make/data/tzdata/iso3166.tab | 17 +++++++++++------
b/jdk/make/data/tzdata/leapseconds | 8 ++++----
b/jdk/make/data/tzdata/northamerica | 2 +-
b/jdk/make/data/tzdata/southamerica | 6 ++++++
b/jdk/make/data/tzdata/zone.tab | 24 ++++++++++++------------
b/jdk/test/java/util/TimeZone/TimeZoneData/VERSION | 2 +-
b/jdk/test/sun/util/calendar/zi/tzdata/VERSION | 2 +-
b/jdk/test/sun/util/calendar/zi/tzdata/africa | 7 -------
b/jdk/test/sun/util/calendar/zi/tzdata/antarctica | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++++++--
b/jdk/test/sun/util/calendar/zi/tzdata/asia | 6 ++++--
b/jdk/test/sun/util/calendar/zi/tzdata/australasia | 8 +++++++-
b/jdk/test/sun/util/calendar/zi/tzdata/backward | 1 -
b/jdk/test/sun/util/calendar/zi/tzdata/europe | 29 ++++++++++++++++++++++-------
b/jdk/test/sun/util/calendar/zi/tzdata/iso3166.tab | 17 +++++++++++------
b/jdk/test/sun/util/calendar/zi/tzdata/leapseconds | 8 ++++----
b/jdk/test/sun/util/calendar/zi/tzdata/northamerica | 2 +-
b/jdk/test/sun/util/calendar/zi/tzdata/southamerica | 6 ++++++
b/jdk/test/sun/util/calendar/zi/tzdata/zone.tab | 24 ++++++++++++------------
25 files changed, 247 insertions(+), 89 deletions(-)
~~~
It's a small change so it gets lost in all the additional test data (something I would like to look at removing from 8u if feasible)
-------------
PR Review Comment: https://git.openjdk.org/jdk8u/pull/69#discussion_r1982441955
More information about the jdk8u-dev
mailing list