scuffed-code/icu4c
Steven R. Loomis f0bddb8ea1 ICU-681 hpux
X-SVN-Rev: 2739
2000-10-19 23:00:11 +00:00
..
data ICU-639 fixed D->d 2000-10-12 23:25:23 +00:00
debian ICU-678 Debian GNU/Linux packaging 2000-10-19 20:58:30 +00:00
docs ICU-449 TimeZone equivalency support 2000-09-27 16:15:21 +00:00
source ICU-681 hpux 2000-10-19 23:00:11 +00:00
.cvsignore ICU-3 support compiling the source files twice with different options 2000-10-09 19:06:57 +00:00
license.html ICU-161 IPSL version needs not to be changed. 2000-01-11 19:30:59 +00:00
readme.html ICU-653 update to readme.html regarding library name changes on win32 2000-10-14 01:38:47 +00:00

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">

<html>
  <head>
    <meta name="GENERATOR" content="HTML Tidy, see www.w3.org">
    <meta name="COPYRIGHT" value=
    "Copyright (c) IBM Corporation and others. All Rights Reserved.">
    <meta name="KEYWORDS" content=
    "ICU; International Components for Unicode; what's new; readme; read me; introduction; downloads; downloading; building; installation;">
    <meta name="DESCRIPTION" content=
    "The introduction to the International Components for Unicode with instructions on building, installation, usage and other bits of information about ICU.">
    <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">

    <title>ReadMe for ICU</title>
<style type="text/css">
      H1 {border-width: 1; border: solid; text-align: center}
      H2 {text-decoration: underline}
      H3 {text-decoration: underline}
      H4 {text-decoration: underline}
      H5 {text-decoration: underline}
      HR {height: 2; width: "100%"; text-align: center}
      PRE {margin-left: .5in; margin-bottom: .5in}
   
</style>
  </head>

  <body lang="en-US">
    <h1>International Components for Unicode<br>
     ReadMe</h1>

    <p>Version: Oct 13, 2000<br>
     Copyright &copy; 1997-2000 International Business Machines Corporation
    and others. All Rights Reserved.</p>
    <hr>

    <p><br>
    <br>
    </p>

    <h2>Contents</h2>

    <ul type="disc">
      <li><a href="#news">Late Breaking News And What Is New?</a></li>

      <li><a href="#introduction">Introduction</a></li>

      <li><a href="#WhatContain">What the International Components for
      Unicode Contain</a></li>

      <li><a href="#API">API Overview</a></li>

      <li><a href="#PlatformDependencies">Platform Dependencies</a></li>

      <li><a href="#ImportantNotes">Important Installation Notes</a></li>

      <li><a href="#HowToInstall">How to Build And Install ICU</a></li>

      <li><a href="#dataHandling">How ICU Handles Data</a></li>

      <li><a href="#CharsetConvert">Character Set Conversion
      Information</a></li>

      <li><a href="#VersionNumbers">Version Numbers In ICU</a></li>

      <li><a href="#ProgrammingNotes">Programming Notes</a></li>

      <li><a href="#WhereToFindMore">Where To Find More Information</a></li>

      <li><a href="#SubmittingComments">Submitting Comments, Requesting
      Features and Reporting Bugs</a></li>
    </ul>

    <h2><a name="#news">Late Breaking News And What Is New?</a></h2>

    <ul>
      <li><a href="#win32LibNames">Library names are changed and moved on Win32</a></li>
    
      <li><a href="#sharedLibNote">Using Shared Data Libraries</a></li>

      <li><a href="#ErrcodeChanges">Important Change Of Error Codes From
      Streaming Conversion Functions</a></li>
    </ul>
    <hr>

    <h2><a name="introduction">Introduction</a></h2>

    <p>Today's software market is a global one in which it is desirable to
    develop and maintain one application that supports a wide variety of
    national languages. International Components for Unicode provides the
    following tools to help you write language independent applications:</p>

    <ul type="disc">
      <li>UnicodeString supporting the Unicode 3.0 standard</li>

      <li>Resource bundles for storing and accessing localized
      information</li>

      <li>Number formatters for converting binary numbers into text strings
      for meaningful display</li>

      <li>Date and time formatters for converting internal time data into
      text strings for meaningful display</li>

      <li>Message formatters for putting together sequences of strings,
      numbers dates and other format to create messages</li>

      <li>Text collation supporting language sensitive comparison of
      strings</li>

      <li>Text boundary analysis for finding characters, word and sentence
      boundaries</li>

      <li>Changing simple data files rather than modifying program code
      easily localizes applications written using these tools</li>

      <li>Over 150 locales supported. Visit the <a href=
      "http://oss.software.ibm.com/developerworks/opensource/icu/localeexplorer">
      LocaleExplorer
      (http://oss.software.ibm.com/developerworks/opensource/icu/localeexplorer)</a>
      site for a demonstration and a full list of supported locales or <a
      href="docs/supp_loc.html">click here for a table of supported
      locales</a>.</li>
    </ul>

    <p>It is possible to support additional locales by adding more locale
    data files, with no code changes.</p>

    <p>Please refer to POSIX programmer's Guide for details on what the ISO
    locale ID means.</p>

    <p>Your comments are important to making this release successful. We are
    committed to fixing any bugs, and will also use your feedback to help
    plan future releases.</p>

    <p><strong><u>IMPORTANT</u>: Please make sure you understand the <a href=
    "license.html">Copyright and License information</a>.</strong></p>

    <p><br>
    </p>

    <h2><a name="WhatContain">What the International Components for Unicode
    Contain</a></h2>

    <p>There are two ways to download the ICU releases,</p>

    <ul type="disc">
      <li><strong>Official Release Snapshot:</strong><br>
       If you want to use ICU (as opposed to developing it), your best bet is
      to download an official, packaged ICU version of the ICU source code.
      These versions are tested more thoroughly than day-to-day development
      builds of the system, and they are packaged in zip and tar files for
      convenient download. These packaged files can be found at <a href=
      "http://oss.software.ibm.com/icu/download/index.html">http://oss.software.ibm.com/icu/download/index.html</a>.<br>

       If packaged snapshot is named <strong>ICUXXXXXX.zip</strong> , XXXXXX
      is the release version number.<br>
       Please unzip this file. It will re-construct the source
      directory.</li>

      <li><strong>CVS Source Repository:</strong><br>
       If you are interested in developing features, patches, or bug fixes
      for ICU, you should probably be working with the latest version of the
      ICU source code. You will need to check the code out of our CVS
      repository to ensure that you have the most recent version of all of
      the files. There are several ways to do this:</li>

      <li style="list-style: none">
        <ul type="circle">
          <li>WebCVS:<br>
           If you want to browse the code and only make occasional downloads,
          you may want to use WebCVS. It provides a convenient, web-based
          interface for browsing and downloading the latest version of the
          ICU source code and documentation. You can also view each file's
          revision history, display the differences between individual
          revisions, determine which revisions were part of which official
          release, and so on.</li>

          <li>
            WinCVS:<br>
             If you will be doing serious work on ICU, you should probably
            install a CVS client on your own machine so that you can do batch
            operations without going through the WebCVS interface. On
            Windows, we suggest the WinCVS client. The following is the
            example instruction on how to download ICU via WinCVS: 

            <ol>
              <li>Install the WinCVS client, which you can download from the
              WinCVS home page.</li>

              <li>In the WinCVS preferences, specify your CVSRoot to be
              ":pserver:anoncvs@oss.software.ibm.com:/usr/cvs/icu"<br>
               with the password "anoncvs". To enter the CVSRoot value,
              select "Preferences" from the "Cvs Admin" pull-down menu.
              Authentication should be set to "'passwd' file on the cvs
              server".</li>

              <li>To "extract" the most recent version of ICU, select
              "Checkout module" from the "Cvs Admin" menu. Specify "icu" for
              the module name.</li>
            </ol>
          </li>

          <li>CVS command line:<br>
           You can also check out the repository anonymously on UNIX using
          the following commands, after first setting your CVSROOT to point
          to the ICU repository:<br>
          <br>
           <i>export
          CVSROOT=:pserver:anoncvs@oss.software.ibm.com:/usr/cvs/icu<br>
           cvs login CVS password: anoncvs<br>
           cvs checkout icu<br>
           cvs logout</i></li>
        </ul>
      </li>
    </ul>

    <p>For more details on how to download ICU directly from the web site,
    please also see <a href=
    "http://oss.software.ibm.com/icu/download/index.html">http:/oss.software.ibm.com/icu/download/index.html</a></p>

    <p>Below, <strong>$Root</strong> is the placement of the icu directory in
    your file system, like "drive:\...\icu" in your environment. "drive:\..."
    stands for any drive and any directory on that drive that you chose to
    install icu into.</p>

    <table border="1" cellpadding="0" width="100%">
      <caption align="left">
        <strong>The following files describe the code drop</strong>
      </caption>

      <tr>
        <td width="20%">
          <p>readme.html</p>
        </td>

        <td width="80%">
          <p>Describes the International Components for Unicode (this
          file)</p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>license.html</p>
        </td>

        <td width="80%">
          <p>Contains IBM's public license</p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$Root/docs</p>
        </td>

        <td width="80%">
          <p>API documentation for the International Components for
          Unicode</p>
        </td>
      </tr>
    </table>

    <p><br>
    </p>

    <table border="1" cellpadding="0" width="100%">
      <caption align="left">
        <strong>The following directories contain source code and data
        files</strong>
      </caption>

      <tr>
        <td width="20%">
          <p>$Root/source/common/</p>
        </td>

        <td width="80%">
          <p>The utility classes, such as ResourceBundle, Unicode, Locale,
          UnicodeString. The codepage conversion library API,
          UnicodeConverter.</p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$Root/source/i18n/</p>
        </td>

        <td width="80%">
          <p>The collation source files, Collator, RuleBasedCollator and
          CollationKey.<br>
           The text boundary API, which locates character, word, sentence,
          and<br>
           line breaks.<br>
           The format API, which formats and parses data in numeric or date
          format to and from text.</p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$Root/source/test/intltest/</p>
        </td>

        <td width="80%">
          <p>A test suite including all C++ APIs. For information about
          running the test suite, see <a href=
          "docs/intltest.html">docs/intltest.html</a>.</p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$Root/source/test/cintltst/</p>
        </td>

        <td width="80%">
          <p>A test suite including all C APIs. For information about running
          the test suite, see <a href=
          "docs/cintltst.html">docs/cintltst.html.</a></p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$Root/data/</p>
        </td>

        <td width="80%">
          <p>The Unicode 3.0 data file. Please see <a href=
          "http://www.unicode.org/">http://www.unicode.org/</a> for more
          information.<br>
           This directory also contains the resource files for all
          international objects. These files are of three types:</p>

          <ul type="disc">
            <li>TXT files contain general locale data.</li>

            <li>RES files contain non-portable locale data files which are
            generated by the <strong>genrb</strong> tool.</li>

            <li>COL files are non-portable packed binary collation data files
            which are created by the <strong>gencol</strong> tool.</li>

            <li>UCM files which contain mapping tables {from,to} Unicode in
            text format</li>

            <li>CNV files are non-portable packed binary conversion data
            generated by the <strong>makeconv</strong> tool.</li>

            <li>icudata.dll file contains data files in a dynamic loadable
            library format. At this moment, this file contains CNV files,
            converter aliases, timezone data and Unicode character names.
            Please read <a href="docs/udata.html">udata.html</a> for more
            information.</li>

            <li>icudata.dat file contains data files in a memory mapped file
            format. At this moment, this file contains CNV files, converter
            aliases, timezone data and Unicode character names. Please read
            <a href="docs/udata.html">udata.html</a> for more
            information.</li>
          </ul>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$Root/source/tools</p>
        </td>

        <td width="80%">
          <p>Tools for generating the data files. Data files are generated by
          invoking $Root/source/tools/makedata.bat on Win32 or
          $Root/source/make install on Unix.</p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$Root/source/samples</p>
        </td>

        <td width="80%">
          <p>Various sample programs that use ICU</p>
        </td>
      </tr>
    </table>

    <p><br>
    </p>

    <table border="1" cellpadding="0" width="100%">
      <caption align="left">
        <strong>The following directories are populated when you've built the
        framework</strong><br>
         (on Unix, replace $Root with the value given to the "configure"
        script)
      </caption>

      <tr>
        <td width="20%">
          <p>$Root/include/</p>
        </td>

        <td width="80%">
          <p>contains all the public header files.</p>
        </td>
      </tr>

      <tr>
        <td width="20%">
          <p>$output</p>
        </td>

        <td width="80%">
          <p>contains the libraries for static/dynamic linking or executable
          programs.</p>
        </td>
      </tr>
    </table>

    <p><strong>The following shows the main directory structure of the
    International Components for Unicode</strong></p>

    <ul style='list-style-type: disc'>
      <li>
        output 

        <ul style='list-style-type: circle'>
          <li>libraries (built)</li>

          <li>programs (built)</li>
        </ul>
      </li>

      <li>
        icu-NNNN 

        <ul style='list-style-type: circle'>
          <li>
            icu 

            <ul style='list-style-type: square'>
              <li>readme.html</li>

              <li>license.html</li>

              <li>include (built)</li>

              <li>data</li>

              <li>docs</li>

              <li>
                source 

                <ul style='list-style-type: disc'>
                  <li>common</li>

                  <li>
                    extra 

                    <ul style='list-style-type: circle'>
                      <li>ustdio</li>
                    </ul>
                  </li>

                  <li>i18n</li>

                  <li>samples</li>

                  <li>
                    test 

                    <ul style='list-style-type: circle'>
                      <li>cintltst</li>

                      <li>intltest</li>
                    </ul>
                  </li>

                  <li>
                    tools 

                    <ul style='list-style-type: circle'>
                      <li>ctestfw</li>

                      <li>genrb</li>

                      <li>pkgdata</li>

                      <li>makeconv</li>

                      <li>...</li>
                    </ul>
                  </li>
                </ul>
              </li>
            </ul>
          </li>
        </ul>
      </li>
    </ul>

    <h2><a name="API">API Overview</a></h2>

    <p>In the International Components for Unicode, there are two
    categories:</p>

    <ul type="disc">
      <li>
        Low-level Unicode/Resource Attributes: (<strong>icuuc</strong>
        library) 

        <ul type="circle">
          <li><a href="docs/utilCL.html">Utility Classes</a></li>

          <li><a href="docs/conversion_interface.htm">Conversion
          Interface</a></li>
        </ul>
      </li>

      <li>
        High-level Unicode Internationalization: (<strong>icui18n</strong>
        library) 

        <ul type="circle">
          <li><a href="docs/boundCL.html">Text Boundary Classes</a></li>

          <li><a href="docs/collateCL.html">Collation Classes</a></li>

          <li><a href="docs/formatCL.html">Formatting Classes</a></li>

          <li>Transliterator Classes</li>
        </ul>
      </li>
    </ul>

    <p>See <a href=
    "http://oss.software.ibm.com/icu/develop/codestds.html">International
    Components for Unicode Coding Guidelines</a> for a discussion of code
    conventions common to all library classes.</p>

    <p>See also <a href="../icuhtml/aindex.html">../icuhtml/aindex.html</a>
    for an alphabetical index, and <a href=
    "../icuhtml/HIER.html">../icuhtml/HIER.html</a> for a hierarchical index
    to detailed API documentation.<br>
    <br>
    </p>

    <h2><a name="PlatformDependencies">Platform Dependencies</a></h2>

    <p>The platform dependencies have been isolated into the following 4
    files:</p>

    <ul type="disc">
      <li>
        <u>platform.h.in:</u> Platform-dependent typedefs and defines:<br>
        <br>
         

        <ul type="circle">
          <li>XP_CPLUSPLUS for C++ only.</li>

          <li>TRUE and FALSE, bool_t, int8_t, int16_t etc.</li>

          <li>U_EXPORT and U_IMPORT for specifying dynamic library import and
          export</li>
        </ul>
      </li>
    </ul>

    <ul type="disc">
      <li>
        <u>putil.c:</u> platform-dependent implementations of various
        functions that are platform dependent: (declared in putil.h)<br>
        <br>
         

        <ul type="circle">
          <li>icu_isNaN, icu_isInfinite(double), icu_getNaN();
          icu_getInfinity for handling special floating point values.</li>

          <li>icu_tzset, icu_timezone, icu_tzname and time for reading
          platform specific time and timezone information.</li>

          <li>icu_getDefaultDataDirectory, icu_getDefaultLocaleID for reading
          the locale setting and data directory.</li>

          <li>icu_isBigEndian for finding the endianess of the platform.</li>

          <li>icu_nextDouble is used specifically by the ChoiceFormat
          API.</li>
        </ul>
      </li>
    </ul>

    <ul type="disc">
      <li>
        <u>mutex.h and mutex.cpp</u>: Code for doing synchronization in
        multithreaded applications. If you wish to use International
        Components for Unicode in a multithreaded application, you must
        provide a synchronization primitive that the classes can use to
        protect their global data against simultaneous modifications. See <a
        href="docs/mutex.html">docs/mutex.html</a> for more information.<br>
        <br>
         

        <ul type="circle">
          <li>We supply sample implementations for WinNT, Win95, Win98,
          Sun/Solaris, RedHat/Linux, HP-UX and for AIX on an RS/6000.</li>

          <li>If you are changing the platform-dependent files, ptypes.h and
          putil.h may also be interesting, but shouldn't have to be changed.
          If you think any other files than the ones mentioned above have
          platform dependencies, please contact us.</li>

          <li>For the Intltest test suite, intltest.cpp in
          "icu/source/test/intltest/" contains the method pathnameInContext,
          which must also be adapted to any new platform.</li>
        </ul>
      </li>
    </ul>

    <ul type="disc">
      <li>
        <u>udata.h</u>: The data-accessing interface in ICU is implemented
        such that there is a lot of flexibility for reading a data file. Each
        platform can tune the performance of file accessing for its
        environment by choosing to implement one of the following
        options:<br>
        <br>
         

        <ul type="circle">
          <li>DLL</li>

          <li>Memory map</li>

          <li>Plain text</li>
        </ul>
      </li>
    </ul>

    <h2><a name="ImportantNotes">Important Installation Notes</a></h2>

    <h3><a name="ImportantNotesWin32">Win32 Platform</a></h3>

    <hr>
    <h3><a name="win32LibNames">BREAKING NEWS: Library names are changed and libraries are moved on Win32</a></h3>
    
    <p>As it was previously mentioned and proposed on <a href="http://oss.software.ibm.com/icu/archives/icu/icu.0009/msg00138.html">ICU list</a>, 
    ICU libraries on Win32 are now renamed and relocated. The following changes took place:</p>

	<ol>
	<li>in icu\lib:</li>
		<ul>
			<li>Debug\icuuc.lib -> icuuc17d.lib</li>
			<li>Release\icuuc.lib -> icuuc17.lib</li>
			<li>Debug\icui18n.lib -> icuin17d.lib</li>
			<li>Release\icui18n.lib -> icuin17.lib</li>
			<li>Debug\ustdio.lib -> icuio17d.lib</li>
			<li>Release\ustdio.lib -> icuio17.lib</li>
		</ul>
	
	<li>in icu\bin:</li>
		<ul>
			<li>Debug\icuuc.dll -> icuuc17d.dll</li>
			<li>Release\icuuc.dll -> icuuc17.dll</li>
			<li>Debug\icui18n.dll -> icuin17d.dll</li>
			<li>Release\icui18n.dll -> icuin17.dll</li>
			<li>Debug\ustdio.dll -> icuio17d.dll</li>
			<li>Release\ustdio.dll -> icuio17.dll</li>
		</ul>
	</ol>

        <p>Also, ctestfw and ex toolutil (now icutu17) libraries appeared in icu\bin
        and icu\lib dirs, but this shouldn't concern regular users of icu.</p>

<h4>What to do, how to cope?</h4>

        <p>When you first try to compile your programs with new version of icu,
        compilation will fail. The following steps are required:</p>
        
        <ol>
                <li>change your path from ...\icu\bin\debug or ...\icu\bin\release to just ...\icu\bin</li>
                <li>make the same change for MSVC executable directory setting (tools->options->directories)</li>
                <li>In all your .dsp files (project settings), on linker tab change input libraries according to the above scheme. You want to do it for both debug and release versions.</li>
        </ol>

        <p>When a new, binary incompatible version appears after 1.7, the libraries will change the version number, so you will have to rename libraries in your projects again. However, this way you can have several versions of icu on the same machine and change the libraries for different programs without having to change path settings on your machine.</p>

    <hr>
    
    <p>If you are building on the Win32 platform, it is important that you
    understand a few build details:</p>

    <p><u>DLL directories and the PATH setting:</u> As delivered, the
    International Components for Unicode build as several DLLs. These DLLs
    are placed in the directories "icu\bin\Debug" and "icu\bin\Release". You
    must add either of these directories to the PATH environment variable in
    your system, or any executables you build will not be able to access
    International Components for Unicode libraries. Alternatively, you can
    copy the DLL files into a directory already in your PATH, but we do not
    recommend this. You can wind up with multiple copies of the DLL and wind
    up using the wrong one.</p>

    <p><u>To change your PATH:</u> When you are not using the debug version,
    you will want to change the "Debug" part of the path to "Release" instead
    (the $Root is the root ICU installation directory e.g.
    drive:\installation-directory\icu).</p>

    <ul type="disk">
      <li><strong>Windows 2000</strong>: Use the System Icon in the Control
      Panel. Pick the "Advanced" tab. Select the "Environment Variables..."
      button. Select the variable PATH in the lower box, and select the lower
      "Edit..." button. In the "Variable Value" box, append the string
      ";$Root\bin\Debug" to the end of the path string. If there is nothing
      there, just type in "$Root\bin\Debug". Click the Set button, then the
      OK button.</li>

      <li><strong>Windows NT</strong>: Use the System Icon in the Control
      Panel. Pick the "Environment" tab, and select the variable PATH in the
      lower box. In the "value" box, append the string ";$Root\bin\Debug" at
      the end of the path string. If there is nothing there, just type in
      "drive:\...\icu\bin\Debug". Click the Set button, then the Ok
      button.</li>

      <li><strong>Windows 95/98/ME</strong>: Edit the autoexec.bat, and add
      the following line to the end of file, "SET
      PATH=%PATH%;$Root\bin\Debug"</li>
    </ul>

    <p><u>Link with Runtime libraries:</u> All the DLLs link with the C
    runtime library "Debug Multithreaded DLL" or "Multithreaded DLL." (This
    is changed through the Project Settings dialog, on the C/C++ tab, under
    Code Generation.) It is important that any executable or other DLL you
    build which uses the International Components for Unicode DLLs links with
    these runtime libraries as well. If you do not do this, you will
    seemingly get memory errors when you run the executable.<br>
    </p>

    <h3><a name="ImportantNotesOS390">OS/390 Platform</a></h3>

    <p>If you are building on the OS/390 UNIX System Services platform, it is
    important that you understand a few details:</p>

    <ul>
      <li>The gnu utilities gmake and gzip/gunzip are needed and can be
      obtained for OS/390 from <a href=
      "http://www.mks.com/">http://www.mks.com/</a>. Search for OS/390,
      register, and follow download directions.</li>

      <li>
        Encoding considerations: The source code assumes that it is compiled
        with codepage 1047 (to be exact, the UNIX System Services variant of
        it). The pax command converts all of the source code files from ASCII
        to codepage 1047 (USS) EBCDIC. However, some files are binary files
        and must not be converted, or must be converted back to their
        original state. Those files are: 

        <ul>
          <li>All the .brk files located in the icu/data directory
          (icu/data/*.brk)</li>

          <li>icu/source/test/testdata/uni-text.txt</li>

          <li>icu/source/test/testdata/th18057.txt</li>
        </ul>
        Such a conversion can be done using iconv:<br>
         <code>iconv -f IBM-1047 -t ISO8859-1 uni-text.txt &gt;
        uni-text.txt</code>
      </li>

      <li>
        DLL directories and the LIBPATH setting: Building and testing ICU
        needs the ICU libraries on the LIBPATH. In other words, the LIBPATH
        should contain (each path prepended with the root directory that
        contains the icu directory): 

        <ul>
          <li>icu/source/common</li>

          <li>icu/source/i18n</li>

          <li>icu/source/tools/ctestfw</li>

          <li>icu/source/tools/toolutil</li>

          <li>icu/source/extra/ustdio</li>
        </ul>
      </li>

      <li>
        <p>OS/390 supports both native S/390 hexadecimal floating point and,
        with Version 2.6 and later, IEEE binary floating point. This is a
        compile time option. Applications built with IEEE should use ICU dlls
        that are built with IEEE (and vice versa). The environment variable
        IEEE390=1 will cause the OS/390 version of ICU to be built with IEEE
        floating point. The default is native hexadecimal floating point.<br>
         <em>Important:</em> Currently (ICU 1.4.2), native floating point
        support is sufficient for codepage conversion, resource bundle and
        UnicodeString operations, but the Format APIs, especially
        ChoiceFormat, require IEEE binary floating point.</p>

        <p>Examples for configuring ICU:<br>
         Debug build: <code>IEEE390=1 ./configure</code><br>
         Release build: <code>CFLAGS=-2 IEEE390=1 ./configure</code></p>
      </li>

      <li>Since the default make on OS/390 is not gmake, pkgdata tool
      requires that the environment variable MAKE be set to path to
      gmake.</li>

      <li>The makedep executable that is used with the OS/390 ICU build
      process is not shipped with ICU. It is available at the <a href=
      "http://www.s390.ibm.com/products/oe/bpxa1ty2.html">OS/390 UNIX - Tools
      and Toys</a> site. The PATH environment variable should be updated to
      contain the location of this executable prior to build. Alternatively,
      makedep may be moved into an existing PATH directory.</li>

      <li>When running the test suite, the TZ environment variable should be
      set to export TZ="PST8PDT" so that time zone comparisons are
      correct.</li>
    </ul>

    <h3><a name="ImportantNotesOS400">OS/400 Platform</a></h3>

    <p>ICU Reference Release 1.4.0 contains partial support for the 400
    platform, but additional work by the user is currently needed to get it
    to build completely. A future release of the ICU should work
    out-of-the-box under OS/400.</p>

    <ul>
      <li>
        Requirements: 

        <ul>
          <li>QSHELL interpreter installed (install base option 30, operating
          system)</li>

          <li>QShell Utilities, PRPQ 5799-XEH</li>

          <li>ILE C++ for AS/400, PRPQ 5799-GDW</li>

          <li>GNU facilities (the gnu facilities are currently available by
          request only. Send e-mail to <a href=
          "mailto:rchasgo400@us.ibm.com">rchasgo400@us.ibm.com</a> )</li>
        </ul>
        <!-- end requirements -->
      </li>

      <li>
        Build environment setup: 

        <ol>
          <li>Create AS400 target library. This library will be the target
          for the resulting modules, programs and service programs. You will
          specify this library on the OUTPUTDIR environment variable in step
          2.</li>

          <li>
            Set up the following environment variables in your build process
            (use ADDENVVAR or WRKENVVAR CL commands) 

            <div style="margin-left: 2em">
              CC - '/usr/bin/icc'<br>
               CXX - ' /usr/bin/icc'<br>
               MAKE - '/usr/bin/gmake'<br>
               OUTPUTDIR - <i>identifies target as400 library for *module,
              *pgm and *srvpgm objects</i>
            </div>
          </li>

          <li>Add QCXXN, to your build process library list. This results in
          the resolution of CRTCPPMOD used by the icc compiler</li>

          <li>
            Configure the Makefiles (see configure below)
            <strong>Note:</strong> Verify that the mh-os400 configure file is
            used. 

            <ul>
              <li>Run 'configure --host=as400-os400'</li>

              <li>The 'clean' and 'install' targets will not work without
              changes because of symbolic links. To delete the target module,
              program, or service programs replace <tt>rm -rf</tt> with
              <strong>$(RMV)</strong>, and in the library installation
              targets (install-library) change <tt>$(INSTALL)</tt> to
              <strong><tt>$(INSTALL-S)</tt></strong>.</li>
            </ul>
          </li>

          <li>gmake -e (-e to pickup the compilers)</li>
        </ol>
        <!-- end build environment -->
      </li>
    </ul>
    <strong>Note:</strong> About the NULL pointer checks 

    <div style="margin-left: 2em">
      In common/ucnv.c and common/unistr.c (search for U_MAX_PTR), there are
      additional checks for NULL pointers. This is because pointer comparison
      works differently on the AS/400 architecture.
    </div>

    <h2><a name="HowToInstall">How To Build And Install ICU</a></h2>

    <h3><a name="HowToInstallWindows">How To Build And Install On
    Windows</a></h3>

    <p>Building International Components for Unicode requires:</p>

    <ul type="disc">
      <li>Microsoft NT 3.51 or above</li>

      <li>Microsoft Visual C++ 6.0 (Service Pack 2 is required to work with
      the release build of max speed optimization).</li>
    </ul>

    <p>The steps are:</p>

    <ol start="1" type="1">
      <li>Unzip the icu-XXXX.zip file, type "unzip -a icu-XXXX.zip -d
      drive:\directory" under command prompt or use WinZip.
      drive:\directory\icu is the root ($Root) directory (you may but don't
      need to place "icu" into another directory). If you change the root,
      you will change the project settings accordingly in EACH makefile in
      the project, updating the "include" and "library" paths.</li>

      <li>Set the environment variable <strong>ICU_DATA</strong> to the full
      pathname of the data directory. The trailing "\" is required after the
      directory name (e.g. "$Root\data\" will work, but the value
      "$Root\data" is not acceptable). This environment variable indicates
      where the locale data files and conversion mapping tables are
      located.</li>

      <li>Be sure that the ICU binary directory, $Root\bin\[Release|Debug],
      is included in the <strong>PATH</strong> environment variable. The
      tests may not work without the DLL files in the path.</li>

      <li>Set the <strong>TZ</strong> environment variable to
      <strong>PST8PDT</strong>. The tests will not work in any other
      timezone.</li>

      <li>Use Microsoft Visual C++ 6.0 to open the
      "$Root\source\allinone\allinone.dsw" workspace (This workspace includes
      all the International Components for Unicode libraries, necessary ICU
      building tools, and the intltest and cintltest test suite
      projects).</li>

      <li>Set the active Project to the "all" project. To do this: Choose
      "Project" menu, and select "Set active project". In the submenu, select
      the "all" workspace.</li>

      <li>Set the active configuration to "Win32 Debug" or "Win32 Release"
      (See note below).</li>

      <li>Choose the "Build" menu and select "Rebuild All". If you want to
      build the Debug and Release configurations at the same time, choose
      "Build" menu and select "Batch Build..." instead (and mark all
      configurations as checked), then click the button named "Rebuild All".
      The "all" workspace will build all the test programs as well as the
      tools for generating binary locale data files. The "makedata" project
      will be run automatically to convert the locale data files from text
      format into icudata.dll.</li>

      <li>Run the C++ test suite, "intltest". To do this: set the active
      project to "intltest", and press F5 to run it.</li>

      <li>Run the C test suite, "cintltst". To do this: set the active
      project to "cintltst", and press F5 to run it.</li>

      <li>Make sure that both "cintltst" and "intltest" passed without any
      errors. The return codes are non-zero when they do not pass. Visual C++
      will display the return codes in the debug tag of the output window.
      When "intltest" and "cintltest" return 0, it means that everything is
      installed correctly.</li>

      <li>Reset the <strong>TZ</strong> environment variable to its original
      value, unless you plan on testing ICU any further.</li>

      <li>You are now able to develop applications with ICU.</li>
    </ol>

    <p><strong>Note:</strong> To set the active configuration, two different
    possibilities are:</p>

    <ul type="disc">
      <li>Choose "Build" menu, select "Set Active Configuration", and select
      "Win32 Release" or "Win32 Debug".</li>

      <li>Another way is to select "Customize" in the "Tools" menu, select
      the "Toolbars" tab, enable "Build" instead of "Build Minibar", and
      click on "Close". This will bring up a toolbar which you can move aside
      the other permanent toolbars at the top of the MSVC window. The
      advantage is that you now have an easy-to-reach pop-up menu that will
      always show the currently selected active configuration. Or, you can
      drag the project and configuration selections and drop them on the menu
      bar for later selection.</li>
    </ul>

    <p>It is also possible to build each library individually, using the
    workspaces in each respective directory. They have to be built in the
    following order:<br>
    </p>

    <ol start="1" type="1">
      <li>common</li>

      <li>i18n</li>

      <li>makedata (which invokes makeconv, genrb, gencol, genccode
      etc.)</li>

      <li>ctestfw</li>

      <li>intltest and cintltst, if you want to run the test suite.</li>
    </ol>
    Regarding the test suite, please read the directions in <a href=
    "docs/intltest.html">docs/intltest.html</a> and <a href=
    "docs/cintltst.html">docs/cintltst.html</a><br>
     <br>
     

    <h3><a name="HowToInstallUnix">How To Build And Install On Unix</a></h3>

    <p>There is a set of Makefiles for Unix that supports Linux w/gcc,
    Solaris w/gcc and Workshop CC, AIX w/xlc and OS/390 with C++.</p>

    <p>Building International Components for Unicode on Unix requires:</p>

    <p>A UNIX C++ compiler, (gcc, cc, xlc_r, etc...) installed on the target
    machine. A recent version of GNU make (3.7+). OS/390 gnu utilities for
    both make (gmake) and zip (gzip/gunzip) can be found at the MKS web site
    at <a href="http://www.mks.com">http://www.mks.com</a>. Please do a
    search on "os/390".</p>

    <p>The steps are:</p>

    <ol start="1" type="1">
      <li>Decompress the icuXXXX.tar (or icuXXXX.tgz) file.</li>

      <li>Before running the test programs or samples, please set the
      environment variable <strong>ICU_DATA</strong>, the full pathname of
      the data directory, to indicate where the locale data files and
      conversion mapping tables are. If this variable is not set, the default
      user data directory will be used. The trailing "/" is required after
      the directory name (e.g. "$Root\data\" will work, but the value
      "$Root\data" is not acceptable). The <strong>TZ</strong> environment
      variable does not need to be set.</li>

      <li>Change directory to the "icu/source".</li>

      <li>If it is not already set, please set the executable flag for the
      following files (by executing 'chmod +x' command): runConfigureICU,
      configure, install.sh and config.*,</li>

      <li>You also need to set other environment variables for different
      build systems. Use this <a href="docs/build_env.htm">table</a> or the
      provided <a href="source/runConfigureICU">script</a>.</li>

      <li>Type "./configure" or type "./configure --help" to print the
      available options.</li>

      <li>Type "make" to compile the libraries and all the data files. On
      OS/390, both IEEE binary floating point and native S/390 hexadecimal
      floating point calculations are supported. The default is to build with
      native floating-point support. Please set the environment variable
      IEEE390=1 if you would like to make the ICU DLLs with IEEE floating
      point support.</li>

      <li>Optionally, type "make check" to verify the test suite.</li>

      <li>Type "Make install" to install.</li>
    </ol>

    <p>Regarding the test suite, please read the directions in <a href=
    "docs/intltest.html">docs/intltest.html</a> and <a href=
    "docs/cintltst.html">docs/cintltst.html</a>.</p>

    <p>It is also possible to build each library individually, using the
    Makefiles in each respective directory. They have to be built in the
    following order:</p>

    <ol start="1" type="1">
      <li>common</li>

      <li>i18n</li>

      <li>makeconv</li>

      <li>genrb</li>

      <li>gencol</li>

      <li>gentz</li>

      <li>genccode</li>

      <li>ctestfw</li>

      <li>intltest and cintltst, if you want to run the test suite.</li>
    </ol>

    <h3><a name="sharedLibNote">Using Shared Data Libraries</a></h3>

    <p style='margin-left:.5in'>HP/UX has a documented characteristic where
    the shl_unload() function always unloads a library, regardless of how
    many times the library has been loaded. Most operating systems
    reference-count libraries as they are opened. In the future (Jitterbug
    414) this may be corrected in the ICU, but at present we work around this
    problem by simply NOT ever unloading shared libraries. This means that
    once a data library is loaded (ex: libicudata.sl) by a process, it cannot
    be unloaded and replaced without stopping and restarting the process.</p>

    <h2><a name="dataHandling">How ICU handles data</a></h2>

    <h3><a name="addDataHandling">How to add a locale data file</a></h3>

    <p>To add locale data files to International Components for Unicode do
    the following:</p>

    <ol start="1" type="1">
      <li style='margin-top:.25in'>Create a file containing the key-value
      pairs which value you are overriding from the parent locale data file.
      Make sure the filename is the locale ID with the extension ".txt". We
      recommend you copy a parent file and change the values that need to be
      changed, remove all other key-pairs. Be sure to update the locale ID
      key (the outmost brace) with the name of the locale id your a
      creating.</li>

      <li style='margin-top:.25in'>Name the file with locale ID you are
      creating with a ".txt" at the end (e.g. the file "fr_BF.txt" would
      create a locale that inherits all the key-value pairs from
      "fr.txt".).</li>

      <li style='margin-top:.25in'>Add the name of that file (without the
      ".txt" extension) as a single line in "index.txt" file in the default
      locale directory (icu/data/).</li>

      <li style='margin-top:.25in'>Regenerate the data DLL file. Please see
      "<a href="#HowToInstall">How to Install</a>" section for more details
      on how to verify the ICU release.</li>
    </ol>

    <h3><a name="addRBDataToApp">How to add resource bundle data to your
    application</a></h3>

    <p>Adding resource bundle data to your application is simple. Just create
    the resource bundle files with the right format and names in your
    application directory tree. For more information on the resource bundle
    file format see the <a href=
    "../icuhtml/ResourceBundle.html#DOC.DOCU">resource bundle
    documentation</a> or the <a href=
    "http://oss.software.ibm.com/icu/userguide/Fallbackmechanism.html">User's
    Guide</a>).</p>

    <p><strong>Note:</strong> resource bundle tag names should contain only
    invariant 7-bit ASCII characters (e.g. ones from the following set:
    <code>A-Z, a-z, 0-9, &lt;SP&gt;, ", %, &amp;, `, (, ), *, +, ,, -, ., /,
    :, ;, &lt;, =, &gt;, ?, _)</code>. Use that same directory name (absolute
    path) when instantiating a resource bundle at run time.</p>

    <h3><a name="WhereCollation">Where Collation Data is stored</a></h3>

    <p>Collation data is stored in a single directory on a local disk. Each
    locale's data is stored in a corresponding ASCII text file indicated by a
    "CollationElements" tag . For instance, the data for de_CH is stored with
    a tag "CollationElements" in a file named "de_CH.txt". Reading the
    collation data from these files can be time-consuming, especially for
    large pieces of data that occur in languages such as Japanese. For this
    reason, the Collation Framework implements a second file format, a
    performance-optimized, non-portable, binary format. These binary files
    are generated automatically by the framework the first time a collation
    table is parsed. They have names of the form "de_CH.col". Once the files
    are generated by the framework, future loading of those collations occur
    from the binary file, rather than the text file, at much higher
    speed.</p>

    <p>In general, you don't have to do anything special with these files.
    They can be generated directly by using the "gencol" tool. In addition,
    they can also be generated and used automatically by the framework,
    without intervention on your part. However, there are situations in which
    you will have to regenerate them. To do so, you must manually delete the
    ".col" files from your collation data directory and re-run the gencol
    tool.</p>

    <p>You will need to regenerate your ".col" files in the following
    circumstances:</p>

    <ol start="1" type="1">
      <li>You are moving your data to another platform. Since the ".col"
      files are non-portable, you must make sure they are regenerated.</li>

      <li><strong>DO NOT</strong> copy them from one platform to
      another.</li>

      <li>You have changed the "CollationElements" data in the locale's
      ".txt" file. <strong>Note:</strong> if you change the default rules for
      some reason, which underlie all collations, then you will have to
      rebuild ALL your ".col" files, since they all are merged with the
      default rule set.</li>
    </ol>

    <h2><a name="CharsetConvert">Character Set Conversion
    Information</a></h2>

    <p>The charset conversion library provides ways to convert simple text
    strings (e.g., char*) such as ISO 8859-1 to and from Unicode. The
    objective is to provide clean, simple, reliable, portable and adaptable
    data structures and algorithms to support the International Components
    for Unicode's character codeset Conversion APIs. The conversion data in
    the library originated from the NLTC lab in IBM. The IBM character set
    conversion tables are publicly available in the published IBM document
    called "CHARACTER DATA REPRESENTATION ARCHITECTURE - REFERENCE AND
    REGISTRY". The character set conversion library includes single-byte,
    double-byte and some UCS encodings to and from Unicode. This document can
    be ordered through Mechanicsberg and it comes with 2 CD ROMs which have
    machine-readable conversion tables on them. The license agreement is
    included in International Components for Unicode agreement.</p>

    <p>Click <a href="data/convrtrs.txt">here</a> to view converters
    implemented in ICU. To see converters in action, please visit <a href=
    "http://oss.software.ibm.com/developerworks/opensource/icu/localeexplorer/?converter&amp;">
    http://oss.software.ibm.com/developerworks/opensource/icu/localeexplorer/?converter&amp;</a></p>

    <p>To order the document in the US you can call 1-800-879-2755 and
    request document number SC09-2190-00. The cost of this publication is
    $75.00 US not including tax.</p>

    <h2><a name="VersionNumbers">Version Numbers In ICU</a></h2>

    <p>ICU supports extensive versioning of its code and data. Versioning
    allows clients to determine when parts of ICU change, and what the effect
    of the change is.</p>

    <p>ICU as a whole has a version number. ICU components such as Collator
    have their own distinct version numbers. Each resource bundle, including
    all the locale data resource bundles, has its own version number.
    Individual tagged items within a resource bundle have their own version
    numbers.</p>

    <p>All version numbers are in the form of a UVersionInfo structure, which
    is an array of four unsigned bytes. These bytes are:</p>

    <ul>
      <li>0: Major version number</li>

      <li>1: Minor version number</li>

      <li>2: Milli version number</li>

      <li>3: Patch version number</li>
    </ul>

    <p>UVersionNumber structures can be converted to and from string
    representations as dotted integers, such as "1.4.5.0", using the
    u_versionToString() and u_stringToVersion() functions.</p>

    <p>Version numbers monotonically increase as changes are made. Two
    UVersionInfo structure may be compared using binary comparison (memcmp)
    to see which is larger (newer). It only makes sense to compare the same
    flavor of version number; you cannot compare the ICU version number to
    the Collator version number, for instance.</p>

    <p>The interpretation of version numbers depends on what is being
    described.</p>

    <h3><a name="VersionNumbersRelease">ICU Release Version Number</a></h3>

    <p>0 (Major): Reference release with major feature addition or
    change.</p>

    <p>1 (Minor): Reference release without major feature addition.</p>

    <p>2 (Milli): Maintenance update to the reference releases.</p>

    <p>3 (Patch): Enhancement/patch update.</p>

    <h3><a name="VersionNumbersCode">Code Component Version Numbers</a></h3>

    <p>0 (Major): Breaking change. Results and data generated by the new
    version are incompatible with those generated by the preceding version.
    <em>Example</em>: In ICU 1.5, the implementation of ResourceBundle
    changed drastically. The data structure, algorithm for parsing data, and
    so on are completely different in 1.5. This required an increment of the
    major version number.</p>

    <p>1 (Minor): Backward-compatible change. The new version of the code can
    read or use data generated by the old version, but the old version cannot
    read or use data generated by the new version. <em>Example</em>: The
    delimiter in the CollationKey gets changed from 0x0000 to 0xFFFF. The
    algorithm keeps track of the differences and recognize these two
    different formats before and after a particular release.</p>

    <p>2 (Milli): Compatible change. Results and data generated by the new
    version are compatible with those generated by the preceding version.
    <em>Example</em>: A byte in the reserved bytes in the data structure is
    now used as a flag/bitmask or whatever, e.g. UDataInfo. The size of the
    data structure is changed and new code is added to check for this flag.
    No other changes are made.</p>

    <p>3 (Patch): Enhancement. A minor change. <em>Example</em>: Performance
    enhancements applied to the code but no changes other than that.</p>

    <h3><a name="VersionNumbersData">Data Component Version Numbers</a></h3>

    <p>0 (Major): Incompatible format change. The layout or format of the
    data has changed. For example, an additional array element has been
    added, or an additional tag. <em>Example</em>: ICU 1.6 changes the
    element layout in "CollationElements". We changed this from a tag with
    plain string value to a tagged array with 3 new subtags, "Version",
    "Override" and "Sequence". This change is incompatible with pre-1.6 code
    and data.</p>

    <p>1 (Minor): Backward-compatible format change. A change that can be
    read and used by previous versions of ICU, but that adds data used by
    newer versions. <em>Example</em>: We added a new tag called "Author" to
    the data file. The only difference between the previous version of the
    data files and the current version is this tag.</p>

    <p>2 (Milli): Compatible change. A change to the data without
    modification of the format. <em>Example</em>: We updated the value of a
    tag "LocaleID" from "041C" to "3801". No other changes were made.</p>

    <p>3 (Patch): Enhancement. A minor change. <em>Example</em>: We changed
    the comments in the data file, perhaps the copyright notices.</p>

    <h3><a name="VersionNumbersRB">Resource Bundles and Elements</a></h3>

    <p>The data stored in resource bundles is tagged with version numbers. A
    resource bundle can contain a tagged string named "Version" that declares
    the version number in dotted-integer format. <em>Example</em>:</p>
<pre>
en {
  Version { "1.0.3.5" }
  ...
}
</pre>

    <p>A resource bundle may omit the "Version" element, in which case it
    will inherit one along the usual chain. <em>Example</em>: If the resource
    bundle en_US contained no "Version" element, it would inherit "1.0.3.5"
    from en.</p>

    <p>If inheritance passes all the way to the root resource bundle and it
    contains no "Version" resource, then the default version number 1.0.0.0
    is returned.</p>

    <p>Elements within a resource bundle may also contain version numbers,
    for example:</p>
<pre>
be {
  CollationElements { 
    Version { "1.0.0.0" }
    ...
  }
}
</pre>

    <p>Here the CollationElements data is version 1.0.0.0. This version may
    differ from the version of the enclosing bundle.</p>

    <p>If a resource element lacks a "Version" element, then it inherits the
    "Version" element of its enclosing resource bundle. (This is a special
    case; in general, resource bundle elements do not inherit data from
    enclosing structures.) <em>Example</em>:</p>
<pre>
en {
  Version { "1.0.3.5" }
  ...
}

en_US {
  CollationElements { 
    ...(contains no "Version" element)
  }
}
</pre>

    <p>Here, the version of the CollationElements in en_US is 1.0.3.5. It
    inherits the en_US version, which is inherited from en.</p>

    <p><strong>Note:</strong> The API and code to fully support the mechanism
    described above is not in place yet as of ICU 1.6. See <a href=
    "#VersionNumbersFuture">Future Enhancements</a> below.</p>

    <h3><a name="VersionNumbersWhatComponents">What Components are
    Versioned</a></h3>

    <p>Currently, the following components are versioned.</p>

    <ul>
      <li>The version of ICU as a whole is returned by
      <code>u_getVersion()</code>.</li>

      <li>The version of a ResourceBundle is returned by
      <code>ures_getVersion()</code> and
      <code>ResourceBundle::getVersion()</code>. This is a data version
      number for the bundle as a whole.</li>

      <li>The version of the Unicode character data underlying ICU is
      returned by <code>u_getUnicodeVersion()</code> and
      <code>Unicode::getUnicodeVersion()</code>. This version reflects the
      numbering of the Unicode releases; see <a href=
      "http://www.unicode.org">http://www.unicode.org</a>.</li>

      <li>The version of the Collator is returned by
      <code>Collator::getVersion()</code>. This is a code version number for
      the collation code and algorithm.</li>
    </ul>

    <h3><a name="VersionNumbersFuture">Future Enhancements</a></h3>

    <ul>
      <li>The ResourceBundle version number inheritance mechanism is not
      fully implemented and tested.</li>

      <li>The resource element version number is not implemented at all. API
      for this does not yet exist.</li>

      <li>Versioning of a RuleBasedCollator's data is only possible through
      the ResourceBundle API. There should probably be API on
      RuleBasedCollator (or Collator) to obtain the data version number.</li>

      <li>Versioning of the Normalizer data is not implemented.</li>

      <li>Versioning of the Normalizer algorithm is not implemented.</li>

      <li>Versioning of the Transliterators is not implemented.</li>

      <li>Versioning of formatters, break iterators, and so on is not
      implemented.</li>
    </ul>

    <h2><a name="ProgrammingNotes">Programming Notes</a></h2>

    <h3><a name="ReportingErrors">Reporting Errors</a></h3>

    <p>In order for the code to be portable, only a subset of the C++
    language that will compile correctly on even the oldest of C++ compilers
    (and also to provide a usable C interface) can be used in the
    implementation, which means that there's no use the C++ exception
    mechanism in the code.</p>

    <p>After considering many alternatives, the decision was that every
    function that can fail takes an error-code parameter by reference. This
    is always the last parameter in the function&rsquo;s parameter list. The
    ErrorCode type is defined as a enumerated type. Zero represents no error,
    positive values represent errors, and negative values represent non-error
    status codes. Macros were provided, SUCCESS and FAILURE, to check the
    error code.</p>

    <p>The ErrorCode parameter is an input-output parameter. Every function
    tests the error code before doing anything else, and immediately exits if
    it&rsquo;s a FAILURE error code. If the function fails later on, it sets
    the error code appropriately and exits without doing any other work
    (except, of course, any cleanup it has to do). If the function encounters
    a non-error condition it wants to signal (such as "encountered an
    unmapped character" in transcoding), it sets the error code appropriately
    and continues. Otherwise, the function leaves the error code
    unchanged.</p>

    <p>Generally, only functions that don&rsquo;t take an ErrorCode
    parameter, but call functions that do, have to declare one. Almost all
    functions that take an ErrorCode parameter and also call other functions
    that do merely have to propagate the error code they were passed down to
    the functions they call. Functions that declare a new ErrorCode parameter
    must initialize it to ZERO_ERROR before calling any other functions.</p>

    <p>The rationale here is to allow a function to call several functions
    (that take error codes) in a row without having to check the error code
    after each one. [A function usually will have to check the error code
    before doing any other processing, however, since it is supposed to stop
    immediately after receiving an error code.] Propagating the error-code
    parameter down the call chain saves the programmer from having to declare
    one everywhere, and also allows us to more closely mimic the C++
    exception protocol.</p>

    <h3><a name="FuncDataNaming">C Function and Data Type Naming</a></h3>

    <p><strong>Function names.</strong> If a function is identical (or almost
    identical) to an ANSI or POSIX function, we give it the same name and (as
    much as possible) the same parameter list. A "u" is prepended onto the
    beginning of the name.</p>

    <p>For functions that exist prior to version 1.2.1, that the function
    name should begin with a lower-case "u". After the "u" is a short code
    identifying the subsystem it belongs to (e.g., "loc", "rb", "cnv",
    "coll", etc.). This code is separated from the actual function name by an
    underscore, and the actual function name can be anything. For
    example,</p>
<pre>
UChar* uloc_getLanguage(...);
void uloc_setDefaultLocale(...);
UChar* ures_getString(...);
</pre>

    <p><strong>Struct and enum type names.</strong> For structs and enum
    types, the rule is that their names begin with a capital "U." There is no
    underscore for struct names.</p>
<pre>
UResourceBundle;
UCollator;
UCollationResult;
</pre>

    <p><strong>Enum value names.</strong> Enumeration values have names that
    begin with "UXXX" where XXX stands for the name of the functional
    category.</p>
<pre>
UNUM_DECIMAL;
UCOL_GREATER;
</pre>

    <p><strong>Macro names.</strong> Macro names are in all caps, but there
    are currently no other requirements.</p>

    <p><strong>Constant names.</strong> Many constant names (constants
    defined with "const", not macros defined with "#define" that are used as
    constants) begin with a lowercase k, but this isn&rsquo;t universally
    enforced.</p>

    <h3><a name="OverflowHandling">Preflighting and Overflow
    Handling</a></h3>

    <p>In ICU's C APIs, the user needs to adhere to the following principles
    for consistency across all functional categories:</p>

    <ol start="1" type="1">
      <li>All the Unicode string processing should be expressed in terms of a
      UChar* buffer that is always null terminated.</li>

      <li>The APIs assume that the input string parameters are statically
      allocated fix-sized character buffers.</li>

      <li>When the value a function is going to return is already stored as a
      constant value in static space (e.g., it&rsquo;s coming from a fixed
      table, or is stored in a cache), the function will just return the
      const UChar* pointer.</li>

      <li>When the function can&rsquo;t return a UChar* to storage the user
      doesn&rsquo;t have to delete, the caller needs to pass in a pointer to
      a character buffer that the function can fill with the result. This
      pointer needs to be accompanied by a <code>int32_t</code> parameter
      that gives the size of the buffer.</li>
    </ol>

    <p>To find out how large the result buffer should be, ICU provides a
    <strong>preflighting</strong> C interface. The interface works like
    this:</p>

    <ol start="1" type="1">
      <li>When using the "<strong>preflighting</strong>" option: you need to
      pass the function a <code>NULL</code> pointer for the buffer pointer,
      and the function returns the actual size of the result. You can then
      choose to allocate a buffer of the correct size and re-run the
      operation if you would like to.</li>

      <li>After allocating a buffer of some reasonable size on the stack and
      passes that to the function, if the result can fit in that buffer,
      everything works fine. If the result doesn&rsquo;t fit, the function
      will return the actual size needed. You can then allocate a buffer of
      the correct size on the heap and try calling the same function
      again.</li>

      <li>Now you have created a buffer of some reasonable size on the stack
      and passes it to the function. If you don't care about the completeness
      of the result and the allocated buffer is too small, you can continue
      on using the truncated result.</li>
    </ol>

    <p>The following three options demonstrates how to use the preflighting
    interface,</p>
    <hr>
<pre>
/**
 * @param result is a pointer to where the actual result will be.
 * @param maxResultSize is the number of characters the
 *                      buffer pointed to be result has room for.
 * @return The actual length of the result including the
 *         terminating &lt;code&gt;NULL&lt;/code&gt;.
 */
int32_t doSomething( /* input parameters */,
                    UChar* result,
                    int32_t maxResultSize,
                    UErrorCode* err);
</pre>
    <hr>

    <p>In this sample, if the actual result doesn&rsquo;t fit in the space
    available in <code>maxResultSize</code>, this function returns the amount
    of space necessary to hold the result, and result holds as many
    characters of the actual result as possible. If you don&rsquo;t care
    about this, no further action is necessary. If you <i>do</i> care about
    the truncated characters, you can then allocate a buffer on the heap of
    the size specified by the return value and call the function again,
    passing <i>that</i> buffer&rsquo;s address for result.</p>

    <p>All preflighting functions have a fill-in <code>ErrorCode</code>
    parameter (and follow the normal <code>ErrorCode</code> rules), even if
    they are not currently doing so. Buffer overflow would be treated as a
    FAILURE error condition, but would <i>not</i> be reported when the caller
    passes in <code>NULL</code> for <code>actualResultSize</code>
    (presumably, a <code>NULL</code> for this parameter means the client
    doesn&rsquo;t care if he got a buffer overflow). All other failing error
    conditions will overwrite the "buffer overflow" error, e.g.
    <code>MISSING_RESOURCE_ERROR</code> etc..</p>

    <h3><a name="ArrayReturn">Arrays as return types</a></h3>

    <p>Returning an array of strings is fairly easy in C++, but very hard in
    C. Instead of returning the array pointer directly, we opted for an
    iterative interface instead: split the function into two functions. One
    returns the number of elements in the array, and the other one returns a
    single specified element from the array.</p>
    <hr>
<pre>
int32_t countArrayItems(/* parameters */);
int32_t getArrayElement(int32_t elementIndex,

                        /* other parameters */,

                        UChar* result,
                        int32_t maxResultSize,
                        UErrorCode* err);
</pre>
    <hr>

    <p>In this case, iterating across all the elements in the array would
    amount to a call to the count() function followed by multiple calls to
    the getElement() function.</p>
    <hr>
<pre>
UChar element[50];

for (i = 0; i &lt; countArrayItems(...); i++) {
    getArrayItem(i, ..., element, 50, &amp;err);
    /* do something with element */
}
</pre>
    <hr>

    <p>In the case of the resource bundle <code>ures_XXXX</code> functions
    returning 2-dimensional arrays, the getElement() function takes both x
    and y coordinates for the desired element, and the count() function
    returns the number of arrays (x axis). Since the size of each array
    element in the resource 2-D arrays should always be the same, this
    provides an easy-to-use C interface.</p>
    <hr>
<pre>
void countArrayItems(int32_t* rows,
                     int32_t* columns,
                     /* other parameters */);

int32_t get2dArrayElement(int32_t rowIndex,
                          int32_t colIndex,

                          /* other parameters */,

                          UChar* result,
                          int32_t maxResultSize,
                          UErrorCode* err);
</pre>
    <hr>

    <h3><a name="ErrcodeChanges">Important Change Of Error Codes From
    Streaming Conversion Functions</a></h3>

    <p>We have decided to make a semantic change to the conversion API which
    affects applications using ICU that are migrated to use ICU version 1.6
    compared to earlier ICU versions:<br>
     The error code that is set from streaming conversion like</p>
<pre>
ucnv_fromUnicode() - ucnv_toUnicode()
ucnv_fromUChars()  - ucnv_toUChars()
scsu_compress()    - scsu_decompress()
</pre>
    when the target buffer is full but the source not empty is changed from
    <code>U_INDEX_OUTOFBOUNDS_ERROR</code> to
    <code>U_BUFFER_OVERFLOW_ERROR</code>. This change makes the error codes
    more consistent with their names and with their use in other icu
    APIs.<br>
     <br>
     

    <p>You need to test for this new error code if your code uses ICU for
    conversion and used the old error code. ucnv.h and scsu.h are updated
    with this information. Please search in your source code for
    <code>U_INDEX_OUTOFBOUNDS_ERROR</code>. If it is used with the above
    functions (<em>not</em> with <code>ucnv_getNextUChar()</code>), then you
    need to change it to <code>U_BUFFER_OVERFLOW_ERROR</code> in order to get
    your code to work with icu 1.6.</p>

    <p>See the updated sample code in <code>icu/source/samples</code>. All
    samples are updated. See <a href=
    "http://oss.software.ibm.com/developerworks/opensource/icu/bugs?findid=516">
    jitterbug 516</a> for details. This was discussed in july 2000 on the icu
    mailing list. Please see the list archive for the <a href=
    "http://oss.software.ibm.com/icu/archives/icu/icu.0007/msg00142.html">discussion</a>.</p>

    <h2><a name="WhereToFindMore">Where To Find More Information</a></h2>

    <p><a href=
    "http://oss.software.ibm.com/icu/">http://oss.software.ibm.com/icu/</a>
    is a pointer to general information about the International Components
    for Unicode.</p>

    <p><a href="docs/udata.html">docs/udata.html</a> is a raw draft of ICU
    data handling.</p>

    <p><a href="../icuhtml/aindex.html">icuhtml/aindex.html</a> is an
    alphabetical index to detailed API documentation.<br>
     <a href="../icuhtml/HIER.html">icuhtml/HIER.html</a> is a hierarchical
    index to detailed API documentation.</p>

    <p><a href="docs/collate.html">docs/collate.html</a> is an overview to
    Collation.</p>

    <p><a href="docs/BreakIterator.html">docs/BreakIterator.html</a> is a
    diagram showing how BreakIterator processes text elements.</p>

    <p><a href=
    "http://www.ibm.com/developer/unicode/">http://www.ibm.com/developer/unicode/</a>
    is a pointer to information on how to make applications global.<br>
    </p>

    <h2><a name="SubmittingComments">Submitting Comments, Requesting Features
    and Reporting Bugs</a></h2>

    <p>To submit comments, request features and report bugs, please contact
    us. While we are not able to respond individually to each comment, we do
    review all comments. Send Internet email to <a href=
    "mailto:icu@oss.software.ibm.com">icu@oss.software.ibm.com</a>.</p>
    <hr>

    <p>Copyright &copy; 1997-2000 International Business Machines Corporation
    and others. All Rights Reserved.<br>
     IBM Center for Emerging Technologies Silicon Valley,<br>
     10275 N De Anza Blvd., Cupertino, CA 95014<br>
     All rights reserved.</p>
  </body>
</html>