timezone data is duplicated over at least three packages in main

[Matthias Klose]

timezone data is duplicated over at least three packages in main, these are:

  - belocs-locales
  - icu
  - python-tz

probably not something to solve before dapper.

[Jeff Bailey (jbailey)]

tzdata has now been split into a separate package that things can depend on directly.

belocs-locales-data is no longer in main.

I'll add a task for this for icu and python-tz.

[Matthias Klose (doko)]

pytz 2007c now references tzdata; other packages with duplicated tzdata:

 - gcj-4.1 (will be fixed in 4.3 or the generices backport)
 - evolution-data-server

[Martin Pitt (pitti)]

 postgresql-8.2 (8.2.3-2) experimental; urgency=low
 .
   * debian/control: Add Perl dependency to p-8.2-plperl, to ensure that
     creating plperl functions works (as opposed to plperlu, which only needs
     libperl). (see bug #412135)
   * debian/control: Do not mention nor suggest 'pgdocs' any more in p-doc's
     description since pgdocs is only available for 7.4. (see bug #405097)
   * debian/patches/04-timezone-symlinks.patch:
     - Use the timezone database from the system tzdata instead of shipping our
       own. Towards a single authoritative time zone database in Debian and
       Ubuntu... :) (LP: #41159)
     - Drop previous hardlink-to-symlink patch to zic, since that is irrelevant
       now.
     - debian/control: Add tzdata dependency.
   * Add debian/patches/12-vacuum-cycle-hang.patch: Properly release our
     semaphore lock before erroring out wit elog() to prevent deadlocks on
     vacuum errors. Thanks to Heikki Linnakangas!
   * debian/rules: Have a test suite failure fail the build again. Let's ignore
     the old kernels on the Debian mips[el] buildds for now.

[Matthias Klose (doko)]

gcj-4.2 now reads the system timezone data.

[Steve Kowalik (stevenk)]

I don't think icu actually includes any timezone data files. Would someone else mind double checking?

[C de-Avillez (hggdh2)]

As of Evolution 2.21.4 (Hardy), Evo now uses the system tzdata, so I am marking this piece of the bug as fix released.

[Black_Diamond (black-diamond-deactivatedaccount)]

Both ICU4C and ICU4J include their own copy of the tzdata in a more usable format. Since the existing POSIX and Linux APIs insufficiently expose all of the timezone data, and because the CLDR metazone information is tightly bound to a version of the timezone data, ICU can't use the system tzdata. Also ICU is a cross platform library, and not all operating system use the same type of timezone data. This is a common problem for environments that want to detect and use all the rules for the timezone data, like Java, vmware, ICU and others. There is no way to reliably get access to the Olson timezone ID.

ICU already provides instructions for updating the timezone data, and those instructions are located here: http://www.icu-project.org/userguide/dateTimezone.html#tzupdate

I recommend using a similar approach to what Apple and other operating system vendors do with ICU and its timezone data. This basically means using the --with-data-packaging=archive configure option, and defining the ICU_DATA_DIR macro in CPPFLAGS before configuring ICU. Defining a default value for ICU_DATA_DIR allows ICU to find its data. An example can be found on Apple's web site: http://www.opensource.apple.com/darwinsource/Current/ICU-8.11.1/makefile

ICU's User Guide also provides more detailed information about managing ICU's data, if you're interested. http://www.icu-project.org/userguide/icudata.html

If you have further questions, the people on the icu-support mailing list might be able to help.
http://www.icu-project.org/contacts.html

[ranzhe (rzz)]

Looks like problem is still actual.
Latest icu update (52.1-3ubuntu0.2) contains outdated timezone information.

[Dimitri John Ledkov (xnox)]

I wonder if we should change the packaging of tzdata to always build the stuff that icu needs/wants. Such that an update of tzdata ripples through to icu as well. Somehow.