xlat.cpp

IBM的解析xml的工具Xerces的源代码

/*
 * Copyright 1999-2000,2004 The Apache Software Foundation.
 * 
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 * 
 *      http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */


/*
 * $Log: Xlat.cpp,v $
 * Revision 1.21  2004/09/08 13:57:07  peiyongz
 * Apache License Version 2.0
 *
 * Revision 1.20  2004/05/05 10:53:08  amassari
 * Make it compile again on Windows
 *
 * Revision 1.19  2004/04/20 15:30:49  peiyongz
 * to generate code for private default constructor
 *
 * Revision 1.18  2003/05/14 19:11:34  gareth
 * Added code creation for new method that returns error serverity the DOM way.
 *
 * Revision 1.17  2003/04/14 08:41:00  gareth
 * Xlat now works under linux - Big thanks to Neil Graham (I no longer have to find a windows box). Still slight problems working with glibc before 2.2.4 (If you mess up the parameters it seg faults due to handling of wprintf)
 *
 * Revision 1.16  2002/12/12 23:40:39  peiyongz
 * normlize locale string.
 *
 * Revision 1.15  2002/11/12 17:24:58  tng
 * DOM Message: add new domain for DOM Messages.
 *
 * Revision 1.14  2002/11/04 15:24:50  tng
 * C++ Namespace Support.
 *
 * Revision 1.13  2002/09/30 22:09:58  peiyongz
 * To generate icu resource file (in text) for error message.
 *
 * Revision 1.12  2002/07/04 17:40:07  tng
 * Use new DOM in Xlat.
 *
 * Revision 1.11  2002/03/01 16:18:38  tng
 * Nest entire code in an inner block so that reference counting should recover
 * all document and DOMString storage when this block exits.
 *
 * Revision 1.10  2002/02/07 16:19:40  knoaman
 * sane_include update.
 *
 * Revision 1.9  2002/02/01 23:48:37  peiyongz
 * sane_include
 *
 * Revision 1.8  2001/05/03 19:09:36  knoaman
 * Support Warning/Error/FatalError messaging.
 * Validity constraints errors are treated as errors, with the ability by user to set
 * validity constraints as fatal errors.
 *
 * Revision 1.7  2000/03/18 00:00:58  roddey
 * Improved error reporting
 *
 * Revision 1.6  2000/03/03 01:29:36  roddey
 * Added a scanReset()/parseReset() method to the scanner and
 * parsers, to allow for reset after early exit from a progressive parse.
 * Added calls to new Terminate() call to all of the samples. Improved
 * documentation in SAX and DOM parsers.
 *
 * Revision 1.5  2000/03/02 19:55:52  roddey
 * This checkin includes many changes done while waiting for the
 * 1.1.0 code to be finished. I can't list them all here, but a list is
 * available elsewhere.
 *
 * Revision 1.4  2000/02/14 19:25:44  roddey
 * Fixed some small bugs in how it used strings returned from DOM calls.
 *
 * Revision 1.3  2000/02/06 07:48:41  rahulj
 * Year 2K copyright swat.
 *
 * Revision 1.2  2000/01/05 20:24:58  roddey
 * Some changes to simplify life for the Messge Catalog message loader. The formatter
 * for the message loader now spits out a simple header of ids that allows the loader to
 * be independent of hard coded set numbers.
 *
 * Revision 1.1.1.1  1999/11/09 01:01:12  twl
 * Initial checkin
 *
 * Revision 1.5  1999/11/08 20:42:04  rahul
 * Swat for adding in Product name and CVS comment log variable.
 *
 */


// ---------------------------------------------------------------------------
//  This program is designed to parse an XML file which holds error text
//  data. It will build a DOM tree from that source file and can output it
//  a number of different formats.
//
//  In order to drastically simplify the program, it is designed only to run
//  on platforms/compilers that understand Unicode. It can output the data
//  in whatever format is required, so it can handle outputting for other
//  platforms. This also simplifies bootstrapping new releases up on other
//  platforms. Once the Win32 version is working, it can generate output for
//  the other platforms so that they can have loadable text from day one.
// ---------------------------------------------------------------------------


// ---------------------------------------------------------------------------
//  Includes
// ---------------------------------------------------------------------------
#include "Xlat.hpp"


// ---------------------------------------------------------------------------
//  Static data
//
//  gRelativeInputPath
//      This is the path, relative to the given input source root, to the
//      input file. The given local suffix must also be added to it.
// ---------------------------------------------------------------------------
XMLCh*      gRelativeInputPath = 0;


// ---------------------------------------------------------------------------
//  Global data
// ---------------------------------------------------------------------------
XMLCh* typePrefixes[MsgTypes_Count];

// ---------------------------------------------------------------------------
//  temporary variables/conversion utility functions
//  We need different temps depending on treatment of wide characters
// ---------------------------------------------------------------------------
#ifdef longChars
    char* fTmpStr = 0;
#else 
    wchar_t fTmpWStr[256];
#endif 

// ---------------------------------------------------------------------------
//  Local data
//
//  gLocale
//      This is the locale suffix, e.g. US_EN, that is used to find the
//      correct file and can be used on output files as well. Its set via
//      the /Locale= parameter.
//
//  gOutFormat
//      This is the output format, which is given on the command line as
//      /OutFmt= Its mapped to the internal enum which is stored here.
//
//  gOutPath
//      This is the path to the output path, which is given on the command
//      line as /OutPath=. Its just the path, not a name, since the output
//      might consist of multiple output files. They will all be based on
//      the base part of the input name.
//
//  gSrcRoot
//      This the path to the root of the build tree. The input files needed
//      are found in known places relative to it.
// ---------------------------------------------------------------------------
const XMLCh*    gLocale = 0;
OutFormats      gOutFormat = OutFormat_Unknown;
const XMLCh*    gOutPath = 0;
const XMLCh*    gSrcRoot = 0;


// ---------------------------------------------------------------------------
//  Local utility methods
// ---------------------------------------------------------------------------
  
// Initialize the global "constants" (that really require use of the transcoder)
void init_Globals(void) 
{
    typePrefixes[0] = XMLString::transcode("W_");
    typePrefixes[1] = XMLString::transcode("E_");
    typePrefixes[2] = XMLString::transcode("F_");
    gRelativeInputPath  = XMLString::transcode("src/xercesc/NLS/");
}

// Release the global "constants" (that really require use of the transcoder)
void release_Globals(void) 
{
    for(int i=0; i<3; i++) 
    {
        XMLString::release(&typePrefixes[i]);
    }
    XMLString::release(&gRelativeInputPath);
}

//
//  This method is called to parse the parameters. They must be in this
//  order and format, for simplicity:
//
//  /SrcRoot=xxx /OutPath=xxx /OutFmt=xxx /Locale=xxx
//
//static bool parseParms(const int argC, XMLCh** argV)
bool parseParms(const int argC, XMLCh** argV)
{
    if (argC < 5)
        return false;

    unsigned int curParm = 1;
    XMLCh *tmpXMLStr = XMLString::transcode("/SrcRoot=");
    if (XMLString::startsWith(argV[curParm], tmpXMLStr))
    {
        gSrcRoot = &argV[curParm][9];
    }
     else
    {
        wprintf(L"\nExpected /SrcRoot=xxx. Got: %s\n", argV[curParm]);
        XMLString::release(&tmpXMLStr);
        return false;
    }
    XMLString::release(&tmpXMLStr);

    curParm++;
    tmpXMLStr = XMLString::transcode("/OutPath=");
    if (XMLString::startsWith(argV[curParm], tmpXMLStr ))
    {
        gOutPath = &argV[curParm][9];
    }
     else
    {
        wprintf(L"\nExpected /OutPath=xxx. Got: %s\n", argV[curParm]);
        XMLString::release(&tmpXMLStr);
        return false;
    }
    XMLString::release(&tmpXMLStr);


    curParm++;
    tmpXMLStr = XMLString::transcode("/OutFmt=");
    if (XMLString::startsWith(argV[curParm], tmpXMLStr ))
    {
        XMLString::release(&tmpXMLStr);
        const XMLCh* tmpFmt = &argV[curParm][8];
        tmpXMLStr = XMLString::transcode("ResBundle");
        XMLCh *tmpXMLStr2 = XMLString::transcode("Win32RC");
        XMLCh *tmpXMLStr3 = XMLString::transcode("CppSrc");
        XMLCh *tmpXMLStr4 = XMLString::transcode("MsgCat");
        if (!XMLString::compareIString(tmpFmt, tmpXMLStr ))
            gOutFormat = OutFormat_ResBundle;
        else if (!XMLString::compareIString(tmpFmt, tmpXMLStr2 ))
            gOutFormat = OutFormat_Win32RC;
        else if (!XMLString::compareIString(tmpFmt, tmpXMLStr3 ))
            gOutFormat = OutFormat_CppSrc;
        else if (!XMLString::compareIString(tmpFmt, tmpXMLStr4 ))
            gOutFormat = OutFormat_MsgCatalog;
        else
        {
            wprintf(L"\n'%s' is not a legal output format\n", tmpFmt);
            XMLString::release(&tmpXMLStr);
            XMLString::release(&tmpXMLStr2);
            XMLString::release(&tmpXMLStr3);
            XMLString::release(&tmpXMLStr4);
            return false;
        }
        XMLString::release(&tmpXMLStr);
        XMLString::release(&tmpXMLStr2);
        XMLString::release(&tmpXMLStr3);
        XMLString::release(&tmpXMLStr4);
    }
     else
    {
        wprintf(L"\nExpected /OutFmt=xxx. Got: %s\n", argV[curParm]);
        XMLString::release(&tmpXMLStr);
        return false;
    }

    curParm++;
    tmpXMLStr = XMLString::transcode("/Locale=");
    if (XMLString::startsWith(argV[curParm], tmpXMLStr ))
    {
        gLocale = &argV[curParm][8];
    }
     else
    {
        wprintf(L"\nExpected /Locale=xxx. Got: %s\n", argV[curParm]);
        XMLString::release(&tmpXMLStr);
        return false;
    }
    XMLString::release(&tmpXMLStr);

    return true;
}


//static void parseError(const XMLException& toCatch)
void parseError(const XMLException& toCatch)
{
    wprintf
    (
        L"Exception\n   (Line.File):%d.%s\n   ERROR: %s\n\n"
        , toCatch.getSrcLine()
        , toCatch.getSrcFile()
        , toCatch.getMessage()
    );
    throw ErrReturn_ParseErr;
}


//static void parseError(const SAXParseException& toCatch)
void parseError(const SAXParseException& toCatch)
{
    wprintf
    (
        L"SAX Parse Error:\n   (Line.Col.SysId): %d.%d.%s\n   ERROR: %s\n\n"
        , toCatch.getLineNumber()
        , toCatch.getColumnNumber()
        , toCatch.getSystemId()
        , toCatch.getMessage()
    );
    throw ErrReturn_ParseErr;
}


//static void
void
enumMessages(   const   DOMElement*             srcElem