fetchtr.cpp

qt-x11-opensource-src-4.1.4.tar.gz源码

/****************************************************************************
**
** Copyright (C) 1992-2006 Trolltech ASA. All rights reserved.
**
** This file is part of the Qt Linguist of the Qt Toolkit.
**
** This file may be used under the terms of the GNU General Public
** License version 2.0 as published by the Free Software Foundation
** and appearing in the file LICENSE.GPL included in the packaging of
** this file.  Please review the following information to ensure GNU
** General Public Licensing requirements will be met:
** http://www.trolltech.com/products/qt/opensource.html
**
** If you are unsure which license is appropriate for your use, please
** review the following information:
** http://www.trolltech.com/products/qt/licensing.html or contact the
** sales department at [e-mail].
**
** This file is provided AS IS with NO WARRANTY OF ANY KIND, INCLUDING THE
** WARRANTY OF DESIGN, MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
**
****************************************************************************/

#include <metatranslator.h>

#include <QFile>
#include <QRegExp>
#include <QString>
#include <QStack>
#include <QtXml>
#include <QTextCodec>

#include <ctype.h>
#include <errno.h>
#include <stdio.h>
#include <string.h>

/* qmake ignore Q_OBJECT */

static const char MagicComment[] = "TRANSLATOR ";

static QMap<QByteArray, int> needs_Q_OBJECT;
static QMap<QByteArray, int> lacks_Q_OBJECT;

/*
  The first part of this source file is the C++ tokenizer.  We skip
  most of C++; the only tokens that interest us are defined here.
  Thus, the code fragment

      int main()
      {
          printf( "Hello, world!\n" );
          return 0;
      }

  is broken down into the following tokens (Tok_ omitted):

      Ident Ident LeftParen RightParen
      LeftBrace
          Ident LeftParen String RightParen Semicolon
          return Semicolon
      RightBrace.

  The 0 doesn't produce any token.
*/

enum { Tok_Eof, Tok_class, Tok_namespace, Tok_return, Tok_tr,
       Tok_trUtf8, Tok_translate, Tok_Q_OBJECT, Tok_Ident,
       Tok_Comment, Tok_String, Tok_Arrow, Tok_Colon,
       Tok_Gulbrandsen, Tok_LeftBrace, Tok_RightBrace, Tok_LeftParen,
       Tok_RightParen, Tok_Comma, Tok_Semicolon, Tok_Integer };

/*
  The tokenizer maintains the following global variables. The names
  should be self-explanatory.
*/
static QByteArray yyFileName;
static int yyCh;
static char yyIdent[128];
static size_t yyIdentLen;
static char yyComment[65536];
static size_t yyCommentLen;
static char yyString[65536];
static size_t yyStringLen;
static qlonglong yyInteger;
static QStack<int> yySavedBraceDepth;
static QStack<int> yySavedParenDepth;
static int yyBraceDepth;
static int yyParenDepth;
static int yyLineNo;
static int yyCurLineNo;
static int yyBraceLineNo;
static int yyParenLineNo;
static QTextCodec *yyCodecForTr = 0;
static QTextCodec *yyCodecForSource = 0;

// the file to read from (if reading from a file)
static FILE *yyInFile;

// the string to read from and current position in the string (otherwise)
static QString yyInStr;
static int yyInPos;

static int (*getChar)();

static bool yyParsingUtf8;

static int getCharFromFile()
{
    int c = getc( yyInFile );
    if ( c == '\n' )
        yyCurLineNo++;
    return c;
}

static int getCharFromString()
{
    if ( yyInPos == (int) yyInStr.length() ) {
        return EOF;
    } else {
        return yyInStr[yyInPos++].toLatin1();
    }
}

static void startTokenizer( const char *fileName, int (*getCharFunc)(), QTextCodec *codecForTr, QTextCodec *codecForSource )
{
    yyInPos = 0;
    getChar = getCharFunc;

    yyFileName = fileName;
    yyCh = getChar();
    yySavedBraceDepth.clear();
    yySavedParenDepth.clear();
    yyBraceDepth = 0;
    yyParenDepth = 0;
    yyCurLineNo = 1;
    yyBraceLineNo = 1;
    yyParenLineNo = 1;
	yyCodecForTr = codecForTr;
	if (!yyCodecForTr)
		yyCodecForTr = QTextCodec::codecForName("ISO-8859-1");
	Q_ASSERT(yyCodecForTr);
	yyCodecForSource = codecForSource;

	yyParsingUtf8 = false;
}

static int getToken()
{
    const char tab[] = "abfnrtv";
    const char backTab[] = "\a\b\f\n\r\t\v";
    uint n;
	bool quiet;

    yyIdentLen = 0;
    yyCommentLen = 0;
    yyStringLen = 0;

    while ( yyCh != EOF ) {
        yyLineNo = yyCurLineNo;

        if ( isalpha(yyCh) || yyCh == '_' ) {
            do {
                if ( yyIdentLen < sizeof(yyIdent) - 1 )
                    yyIdent[yyIdentLen++] = (char) yyCh;
                yyCh = getChar();
            } while ( isalnum(yyCh) || yyCh == '_' );
            yyIdent[yyIdentLen] = '\0';

            switch ( yyIdent[0] ) {
            case 'Q':
                if ( strcmp(yyIdent + 1, "_OBJECT") == 0 ) {
                    return Tok_Q_OBJECT;
                } else if ( strcmp(yyIdent + 1, "T_TR_NOOP") == 0 ) {
					yyParsingUtf8 = false;
                    return Tok_tr;
                } else if ( strcmp(yyIdent + 1, "T_TRANSLATE_NOOP") == 0 ) {
					yyParsingUtf8 = false;
                    return Tok_translate;
                }
                break;
            case 'T':
                // TR() for when all else fails
				if ( qstricmp(yyIdent + 1, "R") == 0 ) {
					yyParsingUtf8 = false;
                    return Tok_tr;
				}
                break;
            case 'c':
                if ( strcmp(yyIdent + 1, "lass") == 0 )
                    return Tok_class;
                break;
            case 'f':
                /*
                  QTranslator::findMessage() has the same parameters as
                  QApplication::translate().
                */
                if ( strcmp(yyIdent + 1, "indMessage") == 0 )
                    return Tok_translate;
                break;
            case 'n':
                if ( strcmp(yyIdent + 1, "amespace") == 0 )
                    return Tok_namespace;
                break;
            case 'r':
                if ( strcmp(yyIdent + 1, "eturn") == 0 )
                    return Tok_return;
                break;
            case 's':
                if ( strcmp(yyIdent + 1, "truct") == 0 )
                    return Tok_class;
                break;
            case 't':
                if ( strcmp(yyIdent + 1, "r") == 0 ) {
					yyParsingUtf8 = false;
                    return Tok_tr;
                } else if ( qstrcmp(yyIdent + 1, "rUtf8") == 0 ) {
					yyParsingUtf8 = true;
                    return Tok_trUtf8;
                } else if ( qstrcmp(yyIdent + 1, "ranslate") == 0 ) {
					yyParsingUtf8 = false;
                    return Tok_translate;
                }
            }
            return Tok_Ident;
        } else {
            switch ( yyCh ) {
            case '#':
                /*
                  Early versions of lupdate complained about
                  unbalanced braces in the following code:

                      #ifdef ALPHA
                          while ( beta ) {
                      #else
                          while ( gamma ) {
                      #endif
                              delta;
                          }

                  The code contains, indeed, two opening braces for
                  one closing brace; yet there's no reason to panic.

                  The solution is to remember yyBraceDepth as it was
                  when #if, #ifdef or #ifndef was met, and to set
                  yyBraceDepth to that value when meeting #elif or
                  #else.
                */
                do {
                    yyCh = getChar();
                } while ( isspace(yyCh) && yyCh != '\n' );

                switch ( yyCh ) {
                case 'i':
                    yyCh = getChar();
                    if ( yyCh == 'f' ) {
                        // if, ifdef, ifndef
                        yySavedBraceDepth.push( yyBraceDepth );
                        yySavedParenDepth.push( yyParenDepth );
                    }
                    break;
                case 'e':
                    yyCh = getChar();
                    if ( yyCh == 'l' ) {
                        // elif, else
                        if ( !yySavedBraceDepth.isEmpty() ) {
                            yyBraceDepth = yySavedBraceDepth.top();
                            yyParenDepth = yySavedParenDepth.top();
                        }
                    } else if ( yyCh == 'n' ) {
                        // endif
                        if ( !yySavedBraceDepth.isEmpty() ) {
                            yySavedBraceDepth.pop();
                            yySavedParenDepth.pop();
                        }
                    }
                }
                while ( isalnum(yyCh) || yyCh == '_' )
                    yyCh = getChar();
                break;
            case '/':
                yyCh = getChar();
                if ( yyCh == '/' ) {
                    do {
                        yyCh = getChar();
                    } while ( yyCh != EOF && yyCh != '\n' );
                } else if ( yyCh == '*' ) {
                    bool metAster = false;
                    bool metAsterSlash = false;

                    while ( !metAsterSlash ) {
                        yyCh = getChar();
                        if ( yyCh == EOF ) {
                            fprintf( stderr,
                                     "%s: Unterminated C++ comment starting at"
                                     " line %d\n",
                                     (const char *) yyFileName, yyLineNo );
                            yyComment[yyCommentLen] = '\0';
                            return Tok_Comment;
                        }
                        if ( yyCommentLen < sizeof(yyComment) - 1 )
                            yyComment[yyCommentLen++] = (char) yyCh;

                        if ( yyCh == '*' )
                            metAster = true;
                        else if ( metAster && yyCh == '/' )
                            metAsterSlash = true;
                        else
                            metAster = false;
                    }
                    yyCh = getChar();
                    yyCommentLen -= 2;
                    yyComment[yyCommentLen] = '\0';
                    return Tok_Comment;
                }
                break;
            case '"':
                yyCh = getChar();