Home | History | Annotate | Download | only in data
      1 # Copyright (C) 2010, International Business Machines Corporation and others.
      2 # All Rights Reserved.                  
      3 #
      4 # Commands for regenerating ICU4C locale data (.txt files) from CLDR.
      5 #
      6 # The process requires local copies of
      7 #    - CLDR (the source of most of the data, and some Java tools)
      8 #    - ICU4J  (used by the conversion tools)
      9 #    - ICU4C  (the destination for the new data, and the source for some of it)
     10 #
     11 # For an official CLDR data integration into ICU, these should be clean, freshly
     12 # checked-out. For released CLDR sources, an alternative to checking out sources
     13 # for a given version is downloading the zipped sources for the common (core.zip)
     14 # and tools (tools.zip) directory subtrees from the Data column in
     15 # [http://cldr.unicode.org/index/downloads].
     16 #
     17 # The versions of each of these must match. Included with the release notes for
     18 # ICU is the version number and/or a CLDR svn tag name for the revision of CLDR
     19 # that was the source of the data for that release of ICU.
     20 #
     21 # Note: Some versions of the OpenJDK will not build the CLDR java utilities.
     22 #   If you see compilation errors complaining about type incompatibilities with
     23 #   functions on generic classes, try switching to the Sun JDK.
     24 #
     25 # Besides a standard JDK, the process also requires ant
     26 # (http://ant.apache.org/),
     27 # plus the xml-apis.jar from the Apache xalan package
     28 # (http://xml.apache.org/xalan-j/downloads.html).
     29 #
     30 # Note: Enough things can (and will) fail in this process that it is best to
     31 #   run the commands separately from an interactive shell. They should all
     32 #   copy and paste without problems.
     33 #
     34 # It is often useful to save logs of the output of many of the steps in this
     35 # process. The commands below put log files in /tmp; you may want to put them
     36 # somewhere else.
     37 #
     38 #----
     39 #
     40 # There are several environment variables that need to be defined.
     41 #
     42 # a) Java- and ant-related variables
     43 #
     44 # JAVA_HOME:     Path to JDK (a directory, containing e.g. bin/java, bin/javac,
     45 #                etc.); on many systems this can be set using
     46 #                `/usr/libexec/java_home`.
     47 #
     48 # XML_APIS_JAR:  Complete path to xml-apis.jar (at the end of the path)
     49 #
     50 # ANT_OPTS:      Two important options you may want to set:
     51 #
     52 #                -Xmx1024m, to give Java more memory; otherwise it may run out
     53 #                 of heap.
     54 #
     55 #                -DCLDR_DTD_CACHE=/tmp/cldrdtd, to set up a temp directory for
     56 #                 caching the CLDR xml dtd, to speed up data generation.
     57 #
     58 # b) CLDR-related variables
     59 #
     60 # CLDR_DIR:      Path to root of CLDR sources, below which are the common and
     61 #                tools directories. Two other variables will be defined based
     62 #                on this:
     63 # CLDR_MAIN, CLDR_JAR are defined relative to CLDR_DIR.
     64 #
     65 # c) ICU-related variables:
     66 #
     67 # ICU4C_DIR:     Path to root of ICU4C sources, below which is the source dir.
     68 #
     69 # ICU4J_ROOT:    Path to root of ICU4J sources, below which is the main dir.
     70 #                Two other variables will be defined based on this:
     71 # ICU4J_JAR, UTILITIES_JAR are defined relative to ICU4J_ROOT.
     72 #
     73 #----
     74 #
     75 # If you are adding or removing locales, or specific kinds of locale data,
     76 # there are some xml files in the ICU sources that need to be updated (these xml
     77 # files are used in addition to the CLDR files as inputs to the CLDR data build
     78 # process for ICU):
     79 #
     80 #    icu/trunk/source/data/icu-config.xml - Update <locales> to add or remove
     81 #                CLDR locales for inclusion in ICU. Update <paths> to prefer
     82 #                alt forms for certain paths, or to exclude certain paths; note
     83 #                that <paths> items can only have draft or alt attributes.
     84 #
     85 #    icu/trunk/source/data/build.xml - If you are adding or removing break
     86 #                iterators, you need to update  <fileset id="brkitr" ...> under
     87 #                <target name="clean" ...> to clean the correct set of files.
     88 #
     89 #    icu/trunk/source/data/xml/      - If you are adding a new locale, break
     90 #                iterator, collattion tailoring, or rule-based number formatter,
     91 #                you need to add a corresponding xml file in (respectively) the
     92 #                main/, brkitr/, collation/, or rbnf/ subdirectory here.
     93 #
     94 #----
     95 #
     96 # For an official CLDR data integration into ICU, there are some additional
     97 # considerations:
     98 #
     99 # a) Don't commit anything in ICU sources (and possibly any changes in CLDR
    100 #    sources, depending on their nature) until you have finished testing and
    101 #    resolving build issues and test failures for both ICU4C and ICU4J.
    102 #
    103 # b) There are version numbers that may need manual updating in CLDR (other
    104 #    version numbers get updated automatically, based on these):
    105 #
    106 #    common/dtd/ldml.dtd                            - update cldrVersion
    107 #    common/dtd/ldmlBCP47.dtd                       - update cldrVersion
    108 #    common/dtd/ldmlSupplemental.dtd                - update cldrVersion
    109 #    tools/java/org/unicode/cldr/util/CLDRFile.java - update GEN_VERSION
    110 #
    111 # c) After everything is committed, you will need to tag the CLDR, ICU4J, and
    112 #    ICU4C sources that ended up being used for the integration; see step 17
    113 #    below.
    114 #
    115 ################################################################################
    116 
    117 # 1a. Java and ant variables, adjust for your system
    118 # (here XML_APIS_JAR is set as on Mac OSX).
    119 
    120 export JAVA_HOME=`/usr/libexec/java_home`
    121 export XML_APIS_JAR=/Library/Java/Extensions/xml-apis.jar
    122 mkdir /tmp/cldrdtd
    123 export ANT_OPTS="-Xmx1024m -DCLDR_DTD_CACHE=/tmp/cldrdtd"
    124 
    125 # 1b. CLDR variables, adjust for your setup; with cygwin it might be e.g.
    126 # CLDR_DIR=`cygpath -wp /build/cldr`
    127 
    128 export CLDR_DIR=$HOME/cldr/trunk
    129 export CLDR_MAIN=$CLDR_DIR/common/main
    130 export CLDR_JAR=$CLDR_DIR/tools/java/cldr.jar
    131 #export CLDR_CLASSES=$CLDR_DIR/tools/java/classes
    132 
    133 # 1c. ICU variables
    134 
    135 export ICU4C_DIR=$HOME/icu/icu/trunk
    136 export ICU4J_ROOT=$HOME/icu/icu4j/trunk
    137 export ICU4J_JAR=$ICU4J_ROOT/icu4j.jar
    138 export UTILITIES_JAR=$ICU4J_ROOT/out/cldr_util/lib/utilities.jar
    139 #export ICU4J_CLASSES=$ICU4J_ROOT/out/cldr_util/bin
    140 
    141 # 2. Build ICU4J, including the cldr utilities.
    142 
    143 cd $ICU4J_ROOT
    144 ant clean all jar cldrUtil
    145 
    146 # 3. Build the CLDR Java tools
    147 
    148 cd $CLDR_DIR/tools/java
    149 ant jar
    150 
    151 # 4. Configure ICU4C, build and test without new data first, to verify that
    152 # there are no pre-existing errors (configure shown here for MacOSX, adjust
    153 # for your platform).
    154 
    155 cd $ICU4C_DIR/source
    156 ./runConfigureICU MacOSX
    157 make all 2>&1 | tee /tmp/icu4c-oldData-makeAll.txt
    158 make check 2>&1 | tee /tmp/icu4c-oldData-makeCheck.txt
    159 
    160 # 5. Build the new ICU4C data files; these include .txt files and .mk files.
    161 # These new files will replace whatever was already present in the ICU4C sources.
    162 # This process uses ant with ICU's data/build.xml and data/icu-config.xml to
    163 # operate (via CLDR's ant/CLDRConverterTool.java and ant/CLDRBuild.java) the
    164 # necessary CLDR tools including LDML2ICUConverter, ConvertTransforms, etc.
    165 # This process will take several minutes (CLDR_DTD_CACHE helps reduce this).
    166 # Keep a log so you can investigate anything that looks suspicious.
    167 
    168 cd $ICU4C_DIR/source/data
    169 ant clean
    170 ant all 2>&1 | tee /tmp/cldrNN-buildLog.txt
    171 
    172 # 6. Check which data files have modifications, which have been added or removed
    173 # (if there are no changes, you may not need to proceed further). Make sure the
    174 # list seems reasonable.
    175 
    176 svn status
    177 
    178 # 7. Fix any errors, investigate any warnings. Some warnings are expected,
    179 # including  warnings for missing versions in locale names which specify some
    180 # collationvariants, e.g.
    181 #   [cldr-build] WARNING (ja_JP_TRADITIONAL): No version #??
    182 #   [cldr-build] WARNING (zh_TW_STROKE): No version #??
    183 # and warnings for some empty collation bundles, e.g.
    184 #   [cldr-build] WARNING (en):  warning: No collations found. Bundle will ...
    185 #   [cldr-build] WARNING (to):  warning: No collations found. Bundle will ...
    186 #
    187 # Fixing may entail modifying CLDR source data or tools - for example,
    188 # updating the validSubLocales for collation data (file a bug f appropriate).
    189 # Repeat steps 5-7 until there are no build errors and no unexpected
    190 # warnings.
    191 
    192 # 8. Now rebuild ICU4C with the new data and run make check tests.
    193 # Again, keep a log so you can investigate the errors.
    194 
    195 cd $ICU4C_DIR/source
    196 make check 2>&1 | tee /tmp/icu4c-newData-makeCheck.txt
    197 
    198 # 9. Investigate each test case failure. The first run processing new CLDR data
    199 # from the Survey Tool can result in thousands of failures (in many cases, one
    200 # CLDR data fix can resolve hundres of test failures). If the error is caused
    201 # by bad CLDR data, then file a CLDR bug, fix the data, and regenerate from
    202 # step 5a. If the data is OK but the testcase needs to be updated because the
    203 # data has legitimately changed, then update the testcase. You will check in
    204 # the updated testcases along with the new ICU data at the end of this process.
    205 # Repeat steps 5-8 until there are no errors.
    206 
    207 # 10. Now run the make check tests in exhaustive mode:
    208 
    209 cd $ICU4C_DIR/source
    210 export INTLTEST_OPTS="-e"
    211 export CINTLTST_OPTS="-e"
    212 make check 2>&1 | tee /tmp/icu4c-newData-makeCheckEx.txt
    213 
    214 # 11. Again, investigate each failure, fixing CLDR data or ICU test cases as
    215 # appropriate, and repeating steps 5-7 and 10 until there are no errors.
    216 
    217 # 12. Now with ICU4J, build and test without new data first, to verify that
    218 # there are no pre-existing errors (or at least to have the pre-existing errors
    219 # as a base for comparison):
    220 
    221 cd $ICU4J_ROOT
    222 ant all 2>&1 | tee /tmp/icu4j-oldData-antAll.txt
    223 ant check 2>&1 | tee /tmp/icu4j-oldData-antCheck.txt
    224 
    225 # 13. Now build the new data for ICU4J
    226 
    227 cd $ICU4C_DIR/source/data
    228 make icu4j-data-install
    229 
    230 # 14. Now rebuild ICU4J with the new data and run tests:
    231 # Keep a log so you can investigate the errors.
    232 
    233 cd $ICU4J_ROOT
    234 ant check 2>&1 | tee /tmp/icu4j-newData-antCheck.txt
    235 
    236 # 15. Investigate test case failures; fix test cases and repeat from step 14,
    237 # or fix CLDR data and repeat from step 5, as appropriate, unitl; there are no
    238 # more failures in ICU4C or ICU4J (except failures that were present before you
    239 # began testing the new CLDR data).
    240 
    241 # 16. Check the file changes; then svn add or svn remove as necessary, and
    242 # commit the changes.
    243 
    244 cd $ICU4C_DIR/source/
    245 svn status
    246 # add or remove as necessary, then commit
    247 
    248 cd $ICU4J_ROOT
    249 svn status
    250 # add or remove as necessary, then commit
    251 
    252 # 17. For an official CLDR data integration into ICU, now tag the CLDR, ICU4J,
    253 # and ICU4C sources with an appropriate CLDR milestone (you can check previous
    254 # tags for format), e.g.:
    255 
    256 svn copy svn+ssh://unicode.org/repos/cldr/trunk \
    257 svn+ssh://unicode.org/repos/cldr/tags/release-NNN \
    258 --parents -m "cldrbug nnnn: tag cldr sources for NNN"
    259 
    260 svn copy svn+ssh://source.icu-project.org/repos/icu/icu4j/trunk \
    261 svn+ssh://source.icu-project.org/repos/icu/icu4j/tags/cldr-NNN \
    262 --parents -m 'ticket:mmmm: tag the version used for integrating CLDR NNN'
    263 
    264 svn copy svn+ssh://source.icu-project.org/repos/icu/icu/trunk \
    265 svn+ssh://source.icu-project.org/repos/icu/icu/tags/cldr-NNN \
    266 --parents -m 'ticket:mmmm: tag the version used for integrating CLDR NNN'
    267 
    268