1968fe35ad
the need for BUILD_OUT_OF_SOURCE to be defined at configure time. X-SVN-Rev: 2383 |
||
---|---|---|
.. | ||
.cvsignore | ||
genmbcs.c | ||
genmbcs.h | ||
makeconv.c | ||
makeconv.dsp | ||
makeconv.dsw | ||
Makefile.in | ||
mkcnvfle.bat | ||
readme.htm | ||
ucmfiles.mk | ||
ucnvstat.c |
<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN"> <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> <meta name="GENERATOR" content="Microsoft FrontPage 2.0"> <title>ReadMe: Character Set Conversion Tool for Unicode</title> </head> <body bgcolor="#FFFFFF"> <p><font size="5"><b>ReadMe: The Character Set Conversion Tool for Unicode</b></font> </p> <p>Version: <!--webbot bot="Timestamp" startspan s-type="EDITED" s-format="%m/%d/%y" -->10/01/98<!--webbot bot="Timestamp" i-checksum="13821" endspan --> <br> </p> <hr> <p>COPYRIGHT: <br> (C) Copyright International Business Machines Corporation, 1998 <br> Licensed Material - Program-Property of IBM - All Rights Reserved. <br> US Government Users Restricted Rights - Use, duplication, or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. <br> </p> <hr> <p><font size="4"><b>Introduction</b></font> </p> <p><b>makeconv</b> is a tool that converts character set conversion data into binary files for use by the UnicodeConverter C code. The UnicodeConverter code will not function with text format data files. The user of charset conversion library will need to convert all the data files in text format into binary format. </p> <p>For a IBM registered code page to/from Unicode conversion data file in the specified data path, it generates .cnv files, which it puts in the same directory. The tool takes a list of file names as argument, with their absolute or relative path, and will generate a corresponding ".cnv" file with the same path. On Unix shells one can easily convert all the ".cnv" files from the text files in the data directory by using wildcards (e.g. <b>makeconv ../../data/locales/*.ucm</b>). On Win32 we have provided a batch file "<b>mkcnvfle.bat</b>" in the same directory you will find <b>makeconv.c</b>. <b>mkcnvfle</b> converts all the ".ucm" files in intlwork/data/locales to ".cnv" files. It requires one argument : Debug or Release so it knows where to look for the makeconv.exe. <br> (Win32 usage:<b> \intlwork\tools\makeconv\mkcnvfle.bat Debug</b> for debug build, <b>\intlwork\tools\makeconv\mkcnvfle.bat Release</b>, otherwise<b>.</b>) </p> <ul> <li><b>Create a binary file for codepage IBM-949 in the data path "\intlwork\data\locales\"</b>: type in at the command prompt, "<b>makeconv</b><font face="Courier"> \intlwork\data\locales\ibm-949.ucm". </font>The generated binary file will be located in the same directory.</li> <li><b>Look for the codepage IBM-949 data file in the current working directory</b>: type in at the command prompt, "<b>makeconv</b><font face="Courier"> ibm-949.ucm</font>". The tool will try to locate the file "<font face="Courier">ibm-949.ucm</font>" in the current directory. The generated binary file, "<font face="Courier">ibm-949.cnv</font>" will be located in current working directory.</li> </ul> <p><font size="4"><b>Building the Tool</b></font> </p> <p><font size="3">To build the tool on Win32 environment is very easy, simply open the "makeconv</font><font size="3" face="Courier">.dsw</font><font size="3">" workspace project file in MSVC++ 5.0 compiler.</font> <br> On Unix the process of building the tool and the ".cnv" files is included in the build process of the library. The provided makefiles will <br> attempt to build <b>makeconv</b> as it builds other parts of the library. It will also convert all the initial ".ucm" files in data/locales to ".cnv" files. </p> <p><font size="4"><b>Limitations</b></font> </p> <p>The tools uses stdio.h functions to communicate success or failure to the user. Platforms that don't support stdio will have to adapt <br> that portion of the code. <br> </p> </body> </html>