Refactor BasicCharClass out from scanner and make it a singleton

4 files changed, 13 insertions, 142 deletions

diff --git a/basic/Library_sb.mk b/basic/Library_sb.mk
index 8b2e152a874c..50608d4e0585 100644
--- a/basic/Library_sb.mk
+++ b/basic/Library_sb.mk
@@ -80,6 +80,7 @@ $(eval $(call gb_Library_add_exception_objects,sb,\
 	basic/source/classes/sbintern \
 	basic/source/classes/sbunoobj \
 	basic/source/classes/sbxmod \
+	basic/source/comp/basiccharclass \
 	basic/source/comp/buffer \
 	basic/source/comp/codegen \
 	basic/source/comp/dim \
diff --git a/basic/source/comp/scanner.cxx b/basic/source/comp/scanner.cxx
index 3336f0d4ce5d..8c0d10c2b6d7 100644
--- a/basic/source/comp/scanner.cxx
+++ b/basic/source/comp/scanner.cxx
@@ -29,6 +29,7 @@
 // MARKER(update_precomp.py): autogen include statement, do not remove
 #include "precompiled_basic.hxx"
 
+#include "basiccharclass.hxx"
 #include "sbcomp.hxx"
 
 #include <vcl/svapp.hxx>
@@ -151,20 +152,10 @@ static SbxDataType GetSuffixType( sal_Unicode c )
 // return value is sal_False at EOF or errors
 #define BUF_SIZE 80
 
-namespace {
-
-/** Returns true, if the passed character is a white space character. */
-inline bool lclIsWhitespace( sal_Unicode cChar )
-{
-    return (cChar == ' ') || (cChar == '\t') || (cChar == '\f');
-}
-
-} // namespace
-
 void SbiScanner::scanGoto()
 {
     short nTestCol = nCol;
-    while(nTestCol < aLine.getLength() && lclIsWhitespace(aLine[nTestCol]))
+    while(nTestCol < aLine.getLength() && theBasicCharClass::get().isWhitespace(aLine[nTestCol]))
         nTestCol++;
 
     if(nTestCol + 1 < aLine.getLength())
@@ -206,7 +197,7 @@ bool SbiScanner::NextSym()
             p2++, n++;
         // #163944# ignore trailing whitespace
         sal_Int32 nCopyEndPos = n;
-        while( (nBufPos < nCopyEndPos) && lclIsWhitespace( aBuf[ nCopyEndPos - 1 ] ) )
+        while( (nBufPos < nCopyEndPos) && theBasicCharClass::get().isWhitespace( aBuf[ nCopyEndPos - 1 ] ) )
             --nCopyEndPos;
         aLine = aBuf.copy( nBufPos, nCopyEndPos - nBufPos );
         if( n < nLen )
@@ -224,7 +215,7 @@ bool SbiScanner::NextSym()
     }
 
 
-    while( lclIsWhitespace( *pLine ) )
+    while( theBasicCharClass::get().isWhitespace( *pLine ) )
         pLine++, nCol++, bSpaces = true;
 
     nCol1 = nCol;
@@ -244,7 +235,7 @@ bool SbiScanner::NextSym()
     }
 
     // copy character if symbol
-    if( BasicSimpleCharClass::isAlpha( *pLine, bCompatible ) || *pLine == '_' )
+    if( theBasicCharClass::get().isAlpha( *pLine, bCompatible ) || *pLine == '_' )
     {
         // if there's nothing behind '_' , it's the end of a line!
         if( *pLine == '_' && !*(pLine+1) )
@@ -252,7 +243,7 @@ bool SbiScanner::NextSym()
             goto eoln;  }
         bSymbol = true;
         short n = nCol;
-        for ( ; (BasicSimpleCharClass::isAlphaNumeric( *pLine, bCompatible ) || ( *pLine == '_' ) ); pLine++ )
+        for ( ; (theBasicCharClass::get().isAlphaNumeric( *pLine, bCompatible ) || ( *pLine == '_' ) ); pLine++ )
             nCol++;
         aSym = aLine.copy( n, nCol - n );
 
@@ -275,7 +266,7 @@ bool SbiScanner::NextSym()
         // type recognition?
         // don't test the exclamation mark
         // if there's a symbol behind it
-        else if( *pLine != '!' || !BasicSimpleCharClass::isAlpha( pLine[ 1 ], bCompatible ) )
+        else if( *pLine != '!' || !theBasicCharClass::get().isAlpha( pLine[ 1 ], bCompatible ) )
         {
             SbxDataType t = GetSuffixType( *pLine );
             if( t != SbxVARIANT )
@@ -288,8 +279,8 @@ bool SbiScanner::NextSym()
     }
 
     // read in and convert if number
-    else if( BasicSimpleCharClass::isDigit( *pLine & 0xFF )
-        || ( *pLine == '.' && BasicSimpleCharClass::isDigit( *(pLine+1) & 0xFF ) ) )
+    else if( theBasicCharClass::get().isDigit( *pLine & 0xFF )
+             || ( *pLine == '.' && theBasicCharClass::get().isDigit( *(pLine+1) & 0xFF ) ) )
     {
         short exp = 0;
         short comma = 0;
@@ -394,7 +385,7 @@ bool SbiScanner::NextSym()
         long l = 0;
         int i;
         bool bBufOverflow = false;
-        while( BasicSimpleCharClass::isAlphaNumeric( *pLine & 0xFF, bCompatible ) )
+        while( theBasicCharClass::get().isAlphaNumeric( *pLine & 0xFF, bCompatible ) )
         {
             sal_Unicode ch = sal::static_int_cast< sal_Unicode >(
                 toupper( *pLine & 0xFF ) );
@@ -527,85 +518,4 @@ eoln:
     }
 }
 
-LetterTable BasicSimpleCharClass::aLetterTable;
-
-LetterTable::LetterTable( void )
-{
-    for( int i = 0 ; i < 256 ; ++i )
-        IsLetterTab[i] = false;
-
-    IsLetterTab[0xC0] = true;   // � , CAPITAL LETTER A WITH GRAVE ACCENT
-    IsLetterTab[0xC1] = true;   // � , CAPITAL LETTER A WITH ACUTE ACCENT
-    IsLetterTab[0xC2] = true;   // � , CAPITAL LETTER A WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xC3] = true;   // � , CAPITAL LETTER A WITH TILDE
-    IsLetterTab[0xC4] = true;   // � , CAPITAL LETTER A WITH DIAERESIS
-    IsLetterTab[0xC5] = true;   // � , CAPITAL LETTER A WITH RING ABOVE
-    IsLetterTab[0xC6] = true;   // � , CAPITAL LIGATURE AE
-    IsLetterTab[0xC7] = true;   // � , CAPITAL LETTER C WITH CEDILLA
-    IsLetterTab[0xC8] = true;   // � , CAPITAL LETTER E WITH GRAVE ACCENT
-    IsLetterTab[0xC9] = true;   // � , CAPITAL LETTER E WITH ACUTE ACCENT
-    IsLetterTab[0xCA] = true;   // � , CAPITAL LETTER E WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xCB] = true;   // � , CAPITAL LETTER E WITH DIAERESIS
-    IsLetterTab[0xCC] = true;   // � , CAPITAL LETTER I WITH GRAVE ACCENT
-    IsLetterTab[0xCD] = true;   // � , CAPITAL LETTER I WITH ACUTE ACCENT
-    IsLetterTab[0xCE] = true;   // � , CAPITAL LETTER I WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xCF] = true;   // � , CAPITAL LETTER I WITH DIAERESIS
-    IsLetterTab[0xD0] = true;   // � , CAPITAL LETTER ETH
-    IsLetterTab[0xD1] = true;   // � , CAPITAL LETTER N WITH TILDE
-    IsLetterTab[0xD2] = true;   // � , CAPITAL LETTER O WITH GRAVE ACCENT
-    IsLetterTab[0xD3] = true;   // � , CAPITAL LETTER O WITH ACUTE ACCENT
-    IsLetterTab[0xD4] = true;   // � , CAPITAL LETTER O WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xD5] = true;   // � , CAPITAL LETTER O WITH TILDE
-    IsLetterTab[0xD6] = true;   // � , CAPITAL LETTER O WITH DIAERESIS
-    IsLetterTab[0xD8] = true;   // � , CAPITAL LETTER O WITH STROKE
-    IsLetterTab[0xD9] = true;   // � , CAPITAL LETTER U WITH GRAVE ACCENT
-    IsLetterTab[0xDA] = true;   // � , CAPITAL LETTER U WITH ACUTE ACCENT
-    IsLetterTab[0xDB] = true;   // � , CAPITAL LETTER U WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xDC] = true;   // � , CAPITAL LETTER U WITH DIAERESIS
-    IsLetterTab[0xDD] = true;   // � , CAPITAL LETTER Y WITH ACUTE ACCENT
-    IsLetterTab[0xDE] = true;   // � , CAPITAL LETTER THORN
-    IsLetterTab[0xDF] = true;   // � , SMALL LETTER SHARP S
-    IsLetterTab[0xE0] = true;   // � , SMALL LETTER A WITH GRAVE ACCENT
-    IsLetterTab[0xE1] = true;   // � , SMALL LETTER A WITH ACUTE ACCENT
-    IsLetterTab[0xE2] = true;   // � , SMALL LETTER A WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xE3] = true;   // � , SMALL LETTER A WITH TILDE
-    IsLetterTab[0xE4] = true;   // � , SMALL LETTER A WITH DIAERESIS
-    IsLetterTab[0xE5] = true;   // � , SMALL LETTER A WITH RING ABOVE
-    IsLetterTab[0xE6] = true;   // � , SMALL LIGATURE AE
-    IsLetterTab[0xE7] = true;   // � , SMALL LETTER C WITH CEDILLA
-    IsLetterTab[0xE8] = true;   // � , SMALL LETTER E WITH GRAVE ACCENT
-    IsLetterTab[0xE9] = true;   // � , SMALL LETTER E WITH ACUTE ACCENT
-    IsLetterTab[0xEA] = true;   // � , SMALL LETTER E WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xEB] = true;   // � , SMALL LETTER E WITH DIAERESIS
-    IsLetterTab[0xEC] = true;   // � , SMALL LETTER I WITH GRAVE ACCENT
-    IsLetterTab[0xED] = true;   // � , SMALL LETTER I WITH ACUTE ACCENT
-    IsLetterTab[0xEE] = true;   // � , SMALL LETTER I WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xEF] = true;   // � , SMALL LETTER I WITH DIAERESIS
-    IsLetterTab[0xF0] = true;   // � , SMALL LETTER ETH
-    IsLetterTab[0xF1] = true;   // � , SMALL LETTER N WITH TILDE
-    IsLetterTab[0xF2] = true;   // � , SMALL LETTER O WITH GRAVE ACCENT
-    IsLetterTab[0xF3] = true;   // � , SMALL LETTER O WITH ACUTE ACCENT
-    IsLetterTab[0xF4] = true;   // � , SMALL LETTER O WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xF5] = true;   // � , SMALL LETTER O WITH TILDE
-    IsLetterTab[0xF6] = true;   // � , SMALL LETTER O WITH DIAERESIS
-    IsLetterTab[0xF8] = true;   // � , SMALL LETTER O WITH OBLIQUE BAR
-    IsLetterTab[0xF9] = true;   // � , SMALL LETTER U WITH GRAVE ACCENT
-    IsLetterTab[0xFA] = true;   // � , SMALL LETTER U WITH ACUTE ACCENT
-    IsLetterTab[0xFB] = true;   // � , SMALL LETTER U WITH CIRCUMFLEX ACCENT
-    IsLetterTab[0xFC] = true;   // � , SMALL LETTER U WITH DIAERESIS
-    IsLetterTab[0xFD] = true;   // � , SMALL LETTER Y WITH ACUTE ACCENT
-    IsLetterTab[0xFE] = true;   // � , SMALL LETTER THORN
-    IsLetterTab[0xFF] = true;   // � , SMALL LETTER Y WITH DIAERESIS
-}
-
-bool LetterTable::isLetterUnicode( sal_Unicode c )
-{
-    static CharClass* pCharClass = NULL;
-    if( pCharClass == NULL )
-        pCharClass = new CharClass( Application::GetSettings().GetLocale() );
-    String aStr( c );
-    bool bRet = pCharClass->isLetter( aStr, 0 );
-    return bRet;
-}
-
 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/basic/source/comp/token.cxx b/basic/source/comp/token.cxx
index dc8287fe39e4..e3d7b1c0ea13 100644
--- a/basic/source/comp/token.cxx
+++ b/basic/source/comp/token.cxx
@@ -29,7 +29,7 @@
 // MARKER(update_precomp.py): autogen include statement, do not remove
 #include "precompiled_basic.hxx"
 
-#include <ctype.h>
+#include "basiccharclass.hxx"
 #include "sbcomp.hxx"
 
 struct TokenTable { SbiToken t; const char *s; };
@@ -392,7 +392,7 @@ SbiToken SbiTokenizer::Next()
         } while( delta );
         // Symbol? if not >= token
         sal_Unicode ch = aSym[0];
-        if( !BasicSimpleCharClass::isAlpha( ch, bCompatible ) && !bSymbol )
+        if( !theBasicCharClass::get().isAlpha( ch, bCompatible ) && !bSymbol )
             return eCurTok = (SbiToken) (ch & 0x00FF);
         return eCurTok = SYMBOL;
     }
diff --git a/basic/source/inc/scanner.hxx b/basic/source/inc/scanner.hxx
index e98ebc32c9f3..c254cf2977b3 100644
--- a/basic/source/inc/scanner.hxx
+++ b/basic/source/inc/scanner.hxx
@@ -103,46 +103,6 @@ public:
     double    GetDbl()              { return nVal;  }
 };
 
-class LetterTable
-{
-    bool        IsLetterTab[256];
-
-public:
-    LetterTable( void );
-
-    inline bool isLetter( sal_Unicode c )
-    {
-        bool bRet = (c < 256) ? IsLetterTab[c] : isLetterUnicode( c );
-        return bRet;
-    }
-    bool isLetterUnicode( sal_Unicode c );
-};
-
-class BasicSimpleCharClass
-{
-    static LetterTable aLetterTable;
-
-public:
-    static sal_Bool isAlpha( sal_Unicode c, bool bCompatible )
-    {
-        sal_Bool bRet = (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
-                    || (bCompatible && aLetterTable.isLetter( c ));
-        return bRet;
-    }
-
-    static sal_Bool isDigit( sal_Unicode c )
-    {
-        sal_Bool bRet = (c >= '0' && c <= '9');
-        return bRet;
-    }
-
-    static sal_Bool isAlphaNumeric( sal_Unicode c, bool bCompatible )
-    {
-        sal_Bool bRet = isDigit( c ) || isAlpha( c, bCompatible );
-        return bRet;
-    }
-};
-
 #endif
 
 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */