New CLDR Developers

Here is a quick overview of what you need to know to do development work on CLDR.

First, you need to have accounts set up for you on:

    1. Jira — for getting and handling bug reports

    2. GitHub — for submitting code / data. (An account is not needed to clone (download) the repository)

    3. cldr-dev — for discussions of issues, questions, etc. — if you will be joining the TC.

    4. Google Docs — to view/edit the CLDR agenda and internal documents — if you will be joining the TC.

    5. Google Sites — only if you are going to edit this website

If you don't get emails about these, contact Rick or other CLDR contacts. It is handy, though not necessary, for you to use a gmail account for the last two of these. Many people use a different account than their internal company email address; you just have to link them with https://accounts.google.com/SignUp.

Warning: some of these pages get stale. Ask questions on cldr-dev if you run into problems; you or the responder should also fix the stale page.

Next, get your Eclipse environment set up properly.

  1. http://cldr.unicode.org/development/eclipse-setup

  2. http://cldr.unicode.org/development/running-survey-tool/eclipse

Run the CLDR tests to be sure they pass before beginning work:

Command line:

    1. Be at root of the cloned CLDR repository

    2. cd tools/java

    3. ant all

    4. cd ../cldr-unittest

    5. ant check

    6. If you see test errors, for instance TestBasic/TestDtdComparison fails, run only the failing test like so:

    7. ant -Druncheck.arg="-v TestBasic/TestDtdComparison" check

    8. The -v tells test script to show stack trace at the test failure for debugging.

    9. To get all parameters that could be passed at runcheck.arg, run

    10. ant -Druncheck.arg="-?" check

Via eclipse:

Once you are all set up, be sure to read the development process, for how to handle tickets, when you can't make changes, etc.

at: http://cldr.unicode.org/development/development-process (TBD update for migration to Jira/Github)

The table below points to documentation for various tasks.

Other useful pages are under CLDR Development Site; you can also use the search box.

UTS #35: Unicode Locale Data Markup Language (LDML) is the specification of the XML format used for CLDR data, including the interpretation of the CLDR data.