basic/source/comp/basiccharclass.cxx | 139 +++++++++++++++++++++++++++++++++++
basic/source/inc/basiccharclass.hxx | 51 ++++++++++++
2 files changed, 190 insertions(+)
New commits:
commit 25ade2f357f1cedf76919f61ae8d81fda7bdf88d
Author: August Sodora <aug...@gmail.com>
Date: Sun Nov 20 17:15:33 2011 -0500
Refactor BasicCharClass out from scanner and make it a singleton
diff --git a/basic/source/comp/basiccharclass.cxx
b/basic/source/comp/basiccharclass.cxx
new file mode 100644
index 0000000..eb11a39
--- /dev/null
+++ b/basic/source/comp/basiccharclass.cxx
@@ -0,0 +1,139 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*************************************************************************
+ *
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * Copyright 2000, 2010 Oracle and/or its affiliates.
+ *
+ * OpenOffice.org - a multi-platform office productivity suite
+ *
+ * This file is part of OpenOffice.org.
+ *
+ * OpenOffice.org is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Lesser General Public License version 3
+ * only, as published by the Free Software Foundation.
+ *
+ * OpenOffice.org is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU Lesser General Public License version 3 for more details
+ * (a copy is included in the LICENSE file that accompanied this code).
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * version 3 along with OpenOffice.org. If not, see
+ * <http://www.openoffice.org/license.html>
+ * for a copy of the LGPLv3 License.
+ *
+ ************************************************************************/
+
+#include "basiccharclass.hxx"
+
+#include <unotools/charclass.hxx>
+#include <vcl/svapp.hxx>
+
+BasicCharClass::BasicCharClass()
+{
+ for( sal_uInt16 i = 0; i < 256; ++i )
+ IsLetterTab[i] = false;
+
+ IsLetterTab[0xC0] = true; // Ã , CAPITAL LETTER A WITH GRAVE ACCENT
+ IsLetterTab[0xC1] = true; // Ã , CAPITAL LETTER A WITH ACUTE ACCENT
+ IsLetterTab[0xC2] = true; // Ã , CAPITAL LETTER A WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xC3] = true; // Ã , CAPITAL LETTER A WITH TILDE
+ IsLetterTab[0xC4] = true; // Ã , CAPITAL LETTER A WITH DIAERESIS
+ IsLetterTab[0xC5] = true; // Ã
, CAPITAL LETTER A WITH RING ABOVE
+ IsLetterTab[0xC6] = true; // Ã , CAPITAL LIGATURE AE
+ IsLetterTab[0xC7] = true; // Ã , CAPITAL LETTER C WITH CEDILLA
+ IsLetterTab[0xC8] = true; // Ã , CAPITAL LETTER E WITH GRAVE ACCENT
+ IsLetterTab[0xC9] = true; // Ã , CAPITAL LETTER E WITH ACUTE ACCENT
+ IsLetterTab[0xCA] = true; // Ã , CAPITAL LETTER E WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xCB] = true; // Ã , CAPITAL LETTER E WITH DIAERESIS
+ IsLetterTab[0xCC] = true; // Ã , CAPITAL LETTER I WITH GRAVE ACCENT
+ IsLetterTab[0xCD] = true; // Ã , CAPITAL LETTER I WITH ACUTE ACCENT
+ IsLetterTab[0xCE] = true; // Ã , CAPITAL LETTER I WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xCF] = true; // Ã , CAPITAL LETTER I WITH DIAERESIS
+ IsLetterTab[0xD0] = true; // Ã , CAPITAL LETTER ETH
+ IsLetterTab[0xD1] = true; // Ã , CAPITAL LETTER N WITH TILDE
+ IsLetterTab[0xD2] = true; // Ã , CAPITAL LETTER O WITH GRAVE ACCENT
+ IsLetterTab[0xD3] = true; // Ã , CAPITAL LETTER O WITH ACUTE ACCENT
+ IsLetterTab[0xD4] = true; // Ã , CAPITAL LETTER O WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xD5] = true; // Ã , CAPITAL LETTER O WITH TILDE
+ IsLetterTab[0xD6] = true; // Ã , CAPITAL LETTER O WITH DIAERESIS
+ IsLetterTab[0xD8] = true; // Ã , CAPITAL LETTER O WITH STROKE
+ IsLetterTab[0xD9] = true; // Ã , CAPITAL LETTER U WITH GRAVE ACCENT
+ IsLetterTab[0xDA] = true; // Ã , CAPITAL LETTER U WITH ACUTE ACCENT
+ IsLetterTab[0xDB] = true; // Ã , CAPITAL LETTER U WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xDC] = true; // Ã , CAPITAL LETTER U WITH DIAERESIS
+ IsLetterTab[0xDD] = true; // Ã , CAPITAL LETTER Y WITH ACUTE ACCENT
+ IsLetterTab[0xDE] = true; // Ã , CAPITAL LETTER THORN
+ IsLetterTab[0xDF] = true; // Ã , SMALL LETTER SHARP S
+ IsLetterTab[0xE0] = true; // Ã , SMALL LETTER A WITH GRAVE ACCENT
+ IsLetterTab[0xE1] = true; // á , SMALL LETTER A WITH ACUTE ACCENT
+ IsLetterTab[0xE2] = true; // â , SMALL LETTER A WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xE3] = true; // ã , SMALL LETTER A WITH TILDE
+ IsLetterTab[0xE4] = true; // ä , SMALL LETTER A WITH DIAERESIS
+ IsLetterTab[0xE5] = true; // å , SMALL LETTER A WITH RING ABOVE
+ IsLetterTab[0xE6] = true; // æ , SMALL LIGATURE AE
+ IsLetterTab[0xE7] = true; // ç , SMALL LETTER C WITH CEDILLA
+ IsLetterTab[0xE8] = true; // è , SMALL LETTER E WITH GRAVE ACCENT
+ IsLetterTab[0xE9] = true; // é , SMALL LETTER E WITH ACUTE ACCENT
+ IsLetterTab[0xEA] = true; // ê , SMALL LETTER E WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xEB] = true; // ë , SMALL LETTER E WITH DIAERESIS
+ IsLetterTab[0xEC] = true; // ì , SMALL LETTER I WITH GRAVE ACCENT
+ IsLetterTab[0xED] = true; // Ã , SMALL LETTER I WITH ACUTE ACCENT
+ IsLetterTab[0xEE] = true; // î , SMALL LETTER I WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xEF] = true; // ï , SMALL LETTER I WITH DIAERESIS
+ IsLetterTab[0xF0] = true; // ð , SMALL LETTER ETH
+ IsLetterTab[0xF1] = true; // ñ , SMALL LETTER N WITH TILDE
+ IsLetterTab[0xF2] = true; // ò , SMALL LETTER O WITH GRAVE ACCENT
+ IsLetterTab[0xF3] = true; // ó , SMALL LETTER O WITH ACUTE ACCENT
+ IsLetterTab[0xF4] = true; // ô , SMALL LETTER O WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xF5] = true; // õ , SMALL LETTER O WITH TILDE
+ IsLetterTab[0xF6] = true; // ö , SMALL LETTER O WITH DIAERESIS
+ IsLetterTab[0xF8] = true; // ø , SMALL LETTER O WITH OBLIQUE BAR
+ IsLetterTab[0xF9] = true; // ù , SMALL LETTER U WITH GRAVE ACCENT
+ IsLetterTab[0xFA] = true; // ú , SMALL LETTER U WITH ACUTE ACCENT
+ IsLetterTab[0xFB] = true; // û , SMALL LETTER U WITH CIRCUMFLEX ACCENT
+ IsLetterTab[0xFC] = true; // ü , SMALL LETTER U WITH DIAERESIS
+ IsLetterTab[0xFD] = true; // ý , SMALL LETTER Y WITH ACUTE ACCENT
+ IsLetterTab[0xFE] = true; // þ , SMALL LETTER THORN
+ IsLetterTab[0xFF] = true; // ÿ , SMALL LETTER Y WITH DIAERESIS
+}
+
+bool BasicCharClass::isLetter( sal_Unicode c )
+{
+ return (c < 256 ? IsLetterTab[c] : isLetterUnicode( c ));
+}
+
+bool BasicCharClass::isLetterUnicode( sal_Unicode c )
+{
+ static CharClass* pCharClass = NULL;
+ if( pCharClass == NULL )
+ pCharClass = new CharClass( Application::GetSettings().GetLocale() );
+ String aStr( c );
+ // can we get pCharClass to accept a sal_Unicode instead of this waste?
+ return pCharClass->isLetter( aStr, 0 );
+}
+
+bool BasicCharClass::isAlpha( sal_Unicode c, bool bCompatible )
+{
+ return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
+ || (bCompatible && isLetter( c ));
+}
+
+bool BasicCharClass::isDigit( sal_Unicode c )
+{
+ return (c >= '0' && c <= '9');
+}
+
+bool BasicCharClass::isAlphaNumeric( sal_Unicode c, bool bCompatible )
+{
+ return isDigit( c ) || isAlpha( c, bCompatible );
+}
+
+bool BasicCharClass::isWhitespace( sal_Unicode c )
+{
+ return (c == ' ') || (c == '\t') || (c == '\f');
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/basic/source/inc/basiccharclass.hxx
b/basic/source/inc/basiccharclass.hxx
new file mode 100644
index 0000000..0c1e2b4
--- /dev/null
+++ b/basic/source/inc/basiccharclass.hxx
@@ -0,0 +1,51 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*************************************************************************
+ *
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * Copyright 2000, 2010 Oracle and/or its affiliates.
+ *
+ * OpenOffice.org - a multi-platform office productivity suite
+ *
+ * This file is part of OpenOffice.org.
+ *
+ * OpenOffice.org is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Lesser General Public License version 3
+ * only, as published by the Free Software Foundation.
+ *
+ * OpenOffice.org is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU Lesser General Public License version 3 for more details
+ * (a copy is included in the LICENSE file that accompanied this code).
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * version 3 along with OpenOffice.org. If not, see
+ * <http://www.openoffice.org/license.html>
+ * for a copy of the LGPLv3 License.
+ *
+ ************************************************************************/
+
+#include <unotools/charclass.hxx>
+#include <rtl/instance.hxx>
+
+#include <vcl/svapp.hxx>
+
+class BasicCharClass
+{
+ bool IsLetterTab[256];
+
+public:
+ BasicCharClass();
+
+ bool isLetter( sal_Unicode c );
+ bool isLetterUnicode( sal_Unicode c );
+ bool isAlpha( sal_Unicode c, bool bCompatible );
+ bool isDigit( sal_Unicode c );
+ bool isAlphaNumeric( sal_Unicode c, bool bCompatible );
+ bool isWhitespace( sal_Unicode c );
+};
+
+class theBasicCharClass: public rtl::Static<BasicCharClass, theBasicCharClass>
{};
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
_______________________________________________
Libreoffice-commits mailing list
Libreoffice-commits@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
