Attachment 26793 Details for Bug 15914 – 1/4 - Moving WTF Unicode backend to GLib (v2)

[patch] 1/4 - Moving WTF Unicode backend to GLib (v2)

1_4_WTF_Unicode_GLib_v2.diff (text/plain), 30.56 KB, created by Dominik Röttsches (drott) on 2009-01-16 07:46:07 PST

(hide)

Description:

Filename:

MIME Type:

Creator: Dominik Röttsches (drott)

Created: 2009-01-16 07:46:07 PST

Size: 30.56 KB

patch

obsolete

>Index: ChangeLog
>===================================================================
>--- ChangeLog	(revision 39967)
>+++ ChangeLog	(working copy)
>@@ -1,3 +1,23 @@
>+2009-01-16  Dominik RÃ¶ttsches  <dominik.roettsches@access-company.com>
>+
>+        Reviewed by NOBODY (OOPS!).
>+
>+        https://bugs.webkit.org/show_bug.cgi?id=15914
>+        [GTK] Implement Unicode functionality using GLib
>+
>+        Initial version of this patch by JÃ¼rg Billeter.
>+
>+        Adding options for --with-unicode-backend=icu|glib
>+        and checking for pango version >= 1.21.0 if GLib backend
>+        is selected. Temporarily, until remaining parts of
>+        this patch are committed, introduce WTF_USE_GLIB_ICU_UNICODE_HYBRID
>+        macro to allow for a mixed compilation with WTF Unicode
>+        backend based on GLib while text codecs and TextBreakIterator
>+        remain ICU dependent.
>+
>+        * GNUmakefile.am:
>+        * configure.ac:
>+
> 2009-01-11  Xan Lopez  <xan@gnome.org>
> 
>         Reviewed by Holger Freyther.
>Index: GNUmakefile.am
>===================================================================
>--- GNUmakefile.am	(revision 39967)
>+++ GNUmakefile.am	(working copy)
>@@ -50,8 +50,23 @@ IDL_BINDINGS :=
> # Global flags to CPP
> global_cppflags :=
> 
>+if USE_ICU_UNICODE
> global_cppflags += \
> 	-DWTF_USE_ICU_UNICODE=1
>+endif
>+
>+if USE_GLIB_UNICODE
>+# https://bugs.webkit.org/show_bug.cgi?id=15914
>+# In patch 1/4 we are compiling a hybrid version of GLib WTF Unicode
>+# functionality mixed with ICU-based text codecs and TextBreakIterator.
>+# For the transition, introducing an interim HYBRID macro.
>+# This macro is required for compiling WTF with GLib Unicode backend,
>+# but at the same time keeping ICU dependency for codecs and break iterator.
>+# It will be removed with patch 3/4 of the above bug.
>+global_cppflags += \
>+	-DWTF_USE_GLIB_UNICODE=1 \
>+	-DWTF_USE_GLIB_ICU_UNICODE_HYBRID=1
>+endif
> 
> # Default compiler flags
> global_cflags := \
>Index: configure.ac
>===================================================================
>--- configure.ac	(revision 39967)
>+++ configure.ac	(working copy)
>@@ -222,18 +222,22 @@ AC_SUBST([PNG_LIBS])
> # determine the Unicode backend
> AC_MSG_CHECKING([the Unicode backend to use])
> AC_ARG_WITH(unicode_backend,
>-            AC_HELP_STRING([--with-unicode-backend=@<:@icu@:>@],
>+            AC_HELP_STRING([--with-unicode-backend=@<:@icu/glib@:>@],
>                            [Select Unicode backend [default=icu]]),
>-            [],[unicode_backend="icu"])
>+            [],[with_unicode_backend="icu"])
> 
>-case "$unicode_backend" in
>-     icu) ;;
>-     *) AC_MSG_ERROR([Invalid Unicode backend: must be icu.]) ;;
>+case "$with_unicode_backend" in
>+     icu|glib) ;;
>+     *) AC_MSG_ERROR([Invalid Unicode backend: must be icu or glib.]) ;;
> esac
> 
>-AC_MSG_RESULT([$unicode_backend])
>+AC_MSG_RESULT([$with_unicode_backend])
> 
>-if test "$unicode_backend" = "icu"; then
>+# https://bugs.webkit.org/show_bug.cgi?id=15914
>+# Splitting ICU removal patch into smaller portions. We compile a hybrid version
>+# with the WTF Unicode backend being based on GLib while text codecs and TextBreakIterator
>+# keep the ICU dependency. That's why we temporarily add icu headers and libs for glib config case as well.
>+if test "$with_unicode_backend" = "icu" -o "$with_unicode_backend" = "glib"; then
> 	AC_PATH_PROG(icu_config, icu-config, no)
> 	if test "$icu_config" = "no"; then
> 		if test "$os_darwin" = "yes"; then
>@@ -254,6 +258,17 @@ if test "$unicode_backend" = "icu"; then
> 	fi
> fi
> 
>+if test "$with_unicode_backend" = "glib"; then
>+	# https://bugs.webkit.org/show_bug.cgi?id=15914
>+	# Splitting ICU removal patch into smaller portions, that's why we
>+	# temporarily retrieve flags & libs info for glib into UNICODEGLIB
>+	# instead of UNICODE variable, then concatenate.
>+	# Patch 3/4 of the above issue will rename the variable back to UNICODE.
>+	PKG_CHECK_MODULES([UNICODEGLIB], [glib-2.0 pango >= 1.21.0])
>+	UNICODE_CFLAGS="$UNICODE_CFLAGS $UNICODEGLIB_CFLAGS"
>+	UNICODE_LIBS="$UNICODE_LIBS $UNICODEGLIB_LIBS"
>+fi
>+
> AC_SUBST([UNICODE_CFLAGS])
> AC_SUBST([UNICODE_LIBS])
> 
>@@ -712,6 +727,10 @@ AM_CONDITIONAL([TARGET_WIN32], [test "$w
> AM_CONDITIONAL([TARGET_QUARTZ], [test "$with_target" = "quartz"])
> AM_CONDITIONAL([TARGET_DIRECTFB], [test "$with_target" = "directfb"])
> 
>+# Unicode backend conditionals
>+AM_CONDITIONAL([USE_ICU_UNICODE], [test "$with_unicode_backend" = "icu"])
>+AM_CONDITIONAL([USE_GLIB_UNICODE], [test "$with_unicode_backend" = "glib"])
>+
> # HTTP backend conditionals
> AM_CONDITIONAL([USE_CURL], [test "$with_http_backend" = "curl"])
> AM_CONDITIONAL([USE_SOUP], [test "$with_http_backend" = "soup"])
>@@ -761,7 +780,7 @@ Build configuration:
>  Enable debugging (slow)                                  : $enable_debug
>  Enable GCC build optimization                            : $enable_optimizations
>  Code coverage support                                    : $enable_coverage
>- Unicode backend                                          : $unicode_backend
>+ Unicode backend                                          : $with_unicode_backend
>  HTTP backend                                             : $with_http_backend
>  Font backend                                             : $with_font_backend
>  Optimized memory allocator                               : $enable_fast_malloc
>Index: JavaScriptCore/ChangeLog
>===================================================================
>--- JavaScriptCore/ChangeLog	(revision 39967)
>+++ JavaScriptCore/ChangeLog	(working copy)
>@@ -1,3 +1,41 @@
>+2009-01-16  Dominik RÃ¶ttsches  <dominik.roettsches@access-company.com>
>+
>+        Reviewed by NOBODY (OOPS!).
>+
>+        https://bugs.webkit.org/show_bug.cgi?id=15914
>+        [GTK] Implement Unicode functionality using GLib
>+
>+        Original patch by JÃ¼rg Billeter and Naiem Shaik.
>+        Implementing WTF Unicode functionality based on GLib.
>+
>+        * GNUmakefile.am:
>+        * wtf/unicode/Unicode.h:
>+        * wtf/unicode/glib: Added.
>+        * wtf/unicode/glib/UnicodeGLib.cpp: Added.
>+        (WTF::Unicode::foldCase):
>+        (WTF::Unicode::toLower):
>+        (WTF::Unicode::toUpper):
>+        (WTF::Unicode::direction):
>+        (WTF::Unicode::umemcasecmp):
>+        * wtf/unicode/glib/UnicodeGLib.h: Added.
>+        (WTF::Unicode::):
>+        (WTF::Unicode::toLower):
>+        (WTF::Unicode::toUpper):
>+        (WTF::Unicode::toTitleCase):
>+        (WTF::Unicode::isArabicChar):
>+        (WTF::Unicode::isFormatChar):
>+        (WTF::Unicode::isSeparatorSpace):
>+        (WTF::Unicode::isPrintableChar):
>+        (WTF::Unicode::isDigit):
>+        (WTF::Unicode::isPunct):
>+        (WTF::Unicode::mirroredChar):
>+        (WTF::Unicode::category):
>+        (WTF::Unicode::isLower):
>+        (WTF::Unicode::digitValue):
>+        (WTF::Unicode::combiningClass):
>+        (WTF::Unicode::decompositionType):
>+        * wtf/unicode/glib/UnicodeMacrosFromICU.h: Added.
>+
> 2009-01-15  Geoffrey Garen  <ggaren@apple.com>
> 
>         Reviewed by Cameron Zwarich.
>Index: JavaScriptCore/GNUmakefile.am
>===================================================================
>--- JavaScriptCore/GNUmakefile.am	(revision 39967)
>+++ JavaScriptCore/GNUmakefile.am	(working copy)
>@@ -249,9 +249,26 @@ javascriptcore_sources += \
> 	JavaScriptCore/wtf/unicode/CollatorDefault.cpp \
> 	JavaScriptCore/wtf/unicode/UTF8.cpp \
> 	JavaScriptCore/wtf/unicode/UTF8.h \
>-	JavaScriptCore/wtf/unicode/Unicode.h \
>+	JavaScriptCore/wtf/unicode/Unicode.h
>+
>+# ----
>+# icu unicode backend
>+# ----
>+if USE_ICU_UNICODE
>+javascriptcore_sources += \
> 	JavaScriptCore/wtf/unicode/icu/CollatorICU.cpp \
> 	JavaScriptCore/wtf/unicode/icu/UnicodeIcu.h
>+endif # USE_ICU_UNICODE
>+
>+# ----
>+# glib unicode backend
>+# ----
>+if USE_GLIB_UNICODE
>+javascriptcore_sources += \
>+	JavaScriptCore/wtf/unicode/glib/UnicodeGLib.h \
>+	JavaScriptCore/wtf/unicode/glib/UnicodeGLib.cpp \
>+	JavaScriptCore/wtf/unicode/glib/UnicodeMacrosFromICU.h
>+endif
> 
> # Debug build
> if ENABLE_DEBUG
>Index: JavaScriptCore/wtf/unicode/Unicode.h
>===================================================================
>--- JavaScriptCore/wtf/unicode/Unicode.h	(revision 39967)
>+++ JavaScriptCore/wtf/unicode/Unicode.h	(working copy)
>@@ -28,6 +28,8 @@
> #include "qt4/UnicodeQt4.h"
> #elif USE(ICU_UNICODE)
> #include <wtf/unicode/icu/UnicodeIcu.h>
>+#elif USE(GLIB_UNICODE)
>+#include <wtf/unicode/glib/UnicodeGLib.h>
> #else
> #error "Unknown Unicode implementation"
> #endif
>Index: JavaScriptCore/wtf/unicode/glib/UnicodeGLib.cpp
>===================================================================
>--- JavaScriptCore/wtf/unicode/glib/UnicodeGLib.cpp	(revision 0)
>+++ JavaScriptCore/wtf/unicode/glib/UnicodeGLib.cpp	(revision 0)
>@@ -0,0 +1,214 @@
>+/*
>+ *  Copyright (C) 2008 JÃ¼rg Billeter <j@bitron.ch>
>+ *  Copyright (C) 2008 Dominik RÃ¶ttsches <dominik.roettsches@access-company.com>
>+ *
>+ *  This library is free software; you can redistribute it and/or
>+ *  modify it under the terms of the GNU Library General Public
>+ *  License as published by the Free Software Foundation; either
>+ *  version 2 of the License, or (at your option) any later version.
>+ *
>+ *  This library 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
>+ *  Library General Public License for more details.
>+ *
>+ *  You should have received a copy of the GNU Library General Public License
>+ *  along with this library; see the file COPYING.LIB.  If not, write to
>+ *  the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
>+ *  Boston, MA 02110-1301, USA.
>+ *
>+ */
>+
>+#include "UnicodeGLib.h"
>+
>+namespace WTF {
>+namespace Unicode {
>+
>+UChar32 foldCase(UChar32 ch)
>+{
>+    GOwnPtr<GError> gerror;
>+
>+    GOwnPtr<char> utf8char;
>+    utf8char.set(g_ucs4_to_utf8(reinterpret_cast<gunichar*>(&ch), 1, 0, 0, &gerror.outPtr()));
>+    if (gerror)
>+        return ch;
>+
>+    GOwnPtr<char> utf8caseFolded;
>+    utf8caseFolded.set(g_utf8_casefold(utf8char.get(), -1));
>+
>+    GOwnPtr<gunichar> ucs4Result;
>+    ucs4Result.set(g_utf8_to_ucs4_fast(utf8caseFolded.get(), -1, 0));
>+
>+    return *ucs4Result;
>+}
>+
>+int foldCase(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error)
>+{
>+    *error = false;
>+    GOwnPtr<GError> gerror;
>+
>+    GOwnPtr<char> utf8src;
>+    utf8src.set(g_utf16_to_utf8(src, srcLength, 0, 0, &gerror.outPtr()));
>+    if (gerror) {
>+        *error = true;
>+        return -1;
>+    }
>+
>+    GOwnPtr<char> utf8result;
>+    utf8result.set(g_utf8_casefold(utf8src.get(), -1));
>+
>+    long utf16resultLength = -1;
>+    GOwnPtr<UChar> utf16result;
>+    utf16result.set(g_utf8_to_utf16(utf8result.get(), -1, 0, &utf16resultLength, &gerror.outPtr()));
>+    if (gerror) {
>+        *error = true;
>+        return -1;
>+    }
>+
>+    if (utf16resultLength > resultLength) {
>+        *error = true;
>+        return utf16resultLength;
>+    }
>+    memcpy(result, utf16result.get(), utf16resultLength * sizeof(UChar));
>+
>+    return utf16resultLength;
>+}
>+
>+int toLower(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error)
>+{
>+    *error = false;
>+    GOwnPtr<GError> gerror;
>+
>+    GOwnPtr<char> utf8src;
>+    utf8src.set(g_utf16_to_utf8(src, srcLength, 0, 0, &gerror.outPtr()));
>+    if (gerror) {
>+        *error = true;
>+        return -1;
>+    }
>+
>+    GOwnPtr<char> utf8result;
>+    utf8result.set(g_utf8_strdown(utf8src.get(), -1));
>+
>+    long utf16resultLength = -1;
>+    GOwnPtr<UChar> utf16result;
>+    utf16result.set(g_utf8_to_utf16(utf8result.get(), -1, 0, &utf16resultLength, &gerror.outPtr()));
>+    if (gerror) {
>+        *error = true;
>+        return -1;
>+    }
>+
>+    if (utf16resultLength > resultLength) {
>+        *error = true;
>+        return utf16resultLength;
>+    }
>+    memcpy(result, utf16result.get(), utf16resultLength * sizeof(UChar));
>+
>+    return utf16resultLength;
>+}
>+
>+int toUpper(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error)
>+{
>+    *error = false;
>+    GOwnPtr<GError> gerror;
>+
>+    GOwnPtr<char> utf8src;
>+    utf8src.set(g_utf16_to_utf8(src, srcLength, 0, 0, &gerror.outPtr()));
>+    if (gerror) {
>+        *error = true;
>+        return -1;
>+    }
>+
>+    GOwnPtr<char> utf8result;
>+    utf8result.set(g_utf8_strup(utf8src.get(), -1));
>+
>+    long utf16resultLength = -1;
>+    GOwnPtr<UChar> utf16result;
>+    utf16result.set(g_utf8_to_utf16(utf8result.get(), -1, 0, &utf16resultLength, &gerror.outPtr()));
>+    if (gerror) {
>+        *error = true;
>+        return -1;
>+    }
>+
>+    if (utf16resultLength > resultLength) {
>+        *error = true;
>+        return utf16resultLength;
>+    }
>+    memcpy(result, utf16result.get(), utf16resultLength * sizeof(UChar));
>+
>+    return utf16resultLength;
>+}
>+
>+Direction direction(UChar32 c)
>+{
>+    PangoBidiType type = pango_bidi_type_for_unichar(c);
>+    switch (type) {
>+    case PANGO_BIDI_TYPE_L: 
>+        return LeftToRight;
>+    case PANGO_BIDI_TYPE_R: 
>+        return RightToLeft;
>+    case PANGO_BIDI_TYPE_AL:
>+        return RightToLeftArabic;
>+    case PANGO_BIDI_TYPE_LRE:
>+        return LeftToRightEmbedding;
>+    case PANGO_BIDI_TYPE_RLE:
>+        return RightToLeftEmbedding;
>+    case PANGO_BIDI_TYPE_LRO:
>+        return LeftToRightOverride;
>+    case PANGO_BIDI_TYPE_RLO:
>+        return RightToLeftOverride;
>+    case PANGO_BIDI_TYPE_PDF:
>+        return PopDirectionalFormat;
>+    case PANGO_BIDI_TYPE_EN:
>+        return EuropeanNumber;
>+    case PANGO_BIDI_TYPE_AN:
>+        return ArabicNumber;
>+    case PANGO_BIDI_TYPE_ES:
>+        return EuropeanNumberSeparator;
>+    case PANGO_BIDI_TYPE_ET:
>+        return EuropeanNumberTerminator;
>+    case PANGO_BIDI_TYPE_CS:
>+        return CommonNumberSeparator;
>+    case PANGO_BIDI_TYPE_NSM:
>+        return NonSpacingMark;
>+    case PANGO_BIDI_TYPE_BN:
>+        return BoundaryNeutral;
>+    case PANGO_BIDI_TYPE_B:
>+        return BlockSeparator;
>+    case PANGO_BIDI_TYPE_S:
>+        return SegmentSeparator;
>+    case PANGO_BIDI_TYPE_WS:
>+        return WhiteSpaceNeutral;
>+    default:
>+        return OtherNeutral;
>+    } 
>+}
>+
>+int umemcasecmp(const UChar* a, const UChar* b, int len)
>+{
>+    GOwnPtr<char> utf8a;
>+    GOwnPtr<char> utf8b;
>+
>+    utf8a.set(g_utf16_to_utf8(a, len, 0, 0, 0));
>+    utf8b.set(g_utf16_to_utf8(b, len, 0, 0, 0));
>+
>+    GOwnPtr<char> foldedA;
>+    GOwnPtr<char> foldedB;
>+
>+    foldedA.set(g_utf8_casefold(utf8a.get(), -1));
>+    foldedB.set(g_utf8_casefold(utf8b.get(), -1));
>+
>+    // FIXME: umemcasecmp needs to mimic u_memcasecmp of icu
>+    // from the ICU docs:
>+    // "Compare two strings case-insensitively using full case folding.
>+    // his is equivalent to u_strcmp(u_strFoldCase(s1, n, options), u_strFoldCase(s2, n, options))."
>+    //    
>+    // So it looks like we don't need the full g_utf8_collate here, 
>+    // but really a bitwise comparison of casefolded unicode chars (not utf-8 bytes).
>+    // As there is no direct equivalent to this icu function in GLib, for now 
>+    // we'll use g_utf8_collate():
>+
>+    return g_utf8_collate(foldedA.get(), foldedB.get());
>+}
>+
>+
>+} }
>Index: JavaScriptCore/wtf/unicode/glib/UnicodeGLib.h
>===================================================================
>--- JavaScriptCore/wtf/unicode/glib/UnicodeGLib.h	(revision 0)
>+++ JavaScriptCore/wtf/unicode/glib/UnicodeGLib.h	(revision 0)
>@@ -0,0 +1,228 @@
>+/*
>+ *  Copyright (C) 2006 George Staikos <staikos@kde.org>
>+ *  Copyright (C) 2006 Alexey Proskuryakov <ap@nypop.com>
>+ *  Copyright (C) 2007 Apple Computer, Inc. All rights reserved.
>+ *  Copyright (C) 2008 JÃ¼rg Billeter <j@bitron.ch>
>+ *  Copyright (C) 2008 Dominik RÃ¶ttsches <dominik.roettsches@access-company.com>
>+ *
>+ *  This library is free software; you can redistribute it and/or
>+ *  modify it under the terms of the GNU Library General Public
>+ *  License as published by the Free Software Foundation; either
>+ *  version 2 of the License, or (at your option) any later version.
>+ *
>+ *  This library 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
>+ *  Library General Public License for more details.
>+ *
>+ *  You should have received a copy of the GNU Library General Public License
>+ *  along with this library; see the file COPYING.LIB.  If not, write to
>+ *  the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
>+ *  Boston, MA 02110-1301, USA.
>+ *
>+ */
>+
>+#ifndef UnicodeGLib_h
>+#define UnicodeGLib_h
>+
>+#include <stdlib.h>
>+#include <stdint.h>
>+#include <string.h>
>+
>+#include <glib.h>
>+#include <pango/pango.h>
>+
>+#include <wtf/GOwnPtr.h>
>+
>+#include "UnicodeMacrosFromICU.h"
>+
>+typedef uint16_t UChar;
>+typedef int32_t UChar32;
>+
>+namespace WTF {
>+    namespace Unicode {
>+
>+        enum Direction {
>+            LeftToRight,
>+            RightToLeft,
>+            EuropeanNumber,
>+            EuropeanNumberSeparator,
>+            EuropeanNumberTerminator,
>+            ArabicNumber,
>+            CommonNumberSeparator,
>+            BlockSeparator,
>+            SegmentSeparator,
>+            WhiteSpaceNeutral,
>+            OtherNeutral,
>+            LeftToRightEmbedding,
>+            LeftToRightOverride,
>+            RightToLeftArabic,
>+            RightToLeftEmbedding,
>+            RightToLeftOverride,
>+            PopDirectionalFormat,
>+            NonSpacingMark,
>+            BoundaryNeutral
>+        };
>+
>+        enum DecompositionType {
>+            DecompositionNone,
>+            DecompositionCanonical,
>+            DecompositionCompat,
>+            DecompositionCircle,
>+            DecompositionFinal,
>+            DecompositionFont,
>+            DecompositionFraction,
>+            DecompositionInitial,
>+            DecompositionIsolated,
>+            DecompositionMedial,
>+            DecompositionNarrow,
>+            DecompositionNoBreak,
>+            DecompositionSmall,
>+            DecompositionSquare,
>+            DecompositionSub,
>+            DecompositionSuper,
>+            DecompositionVertical,
>+            DecompositionWide,
>+        };
>+
>+        enum CharCategory {
>+            NoCategory =  0,
>+            Other_NotAssigned = U_MASK(G_UNICODE_UNASSIGNED),
>+            Letter_Uppercase = U_MASK(G_UNICODE_UPPERCASE_LETTER),
>+            Letter_Lowercase = U_MASK(G_UNICODE_LOWERCASE_LETTER),
>+            Letter_Titlecase = U_MASK(G_UNICODE_TITLECASE_LETTER),
>+            Letter_Modifier = U_MASK(G_UNICODE_MODIFIER_LETTER),
>+            Letter_Other = U_MASK(G_UNICODE_OTHER_LETTER),
>+
>+            Mark_NonSpacing = U_MASK(G_UNICODE_NON_SPACING_MARK),
>+            Mark_Enclosing = U_MASK(G_UNICODE_ENCLOSING_MARK),
>+            Mark_SpacingCombining = U_MASK(G_UNICODE_COMBINING_MARK),
>+
>+            Number_DecimalDigit = U_MASK(G_UNICODE_DECIMAL_NUMBER),
>+            Number_Letter = U_MASK(G_UNICODE_LETTER_NUMBER),
>+            Number_Other = U_MASK(G_UNICODE_OTHER_NUMBER),
>+
>+            Separator_Space = U_MASK(G_UNICODE_SPACE_SEPARATOR),
>+            Separator_Line = U_MASK(G_UNICODE_LINE_SEPARATOR),
>+            Separator_Paragraph = U_MASK(G_UNICODE_PARAGRAPH_SEPARATOR),
>+
>+            Other_Control = U_MASK(G_UNICODE_CONTROL),
>+            Other_Format = U_MASK(G_UNICODE_FORMAT),
>+            Other_PrivateUse = U_MASK(G_UNICODE_PRIVATE_USE),
>+            Other_Surrogate = U_MASK(G_UNICODE_SURROGATE),
>+
>+            Punctuation_Dash = U_MASK(G_UNICODE_DASH_PUNCTUATION),
>+            Punctuation_Open = U_MASK(G_UNICODE_OPEN_PUNCTUATION),
>+            Punctuation_Close = U_MASK(G_UNICODE_CLOSE_PUNCTUATION),
>+            Punctuation_Connector = U_MASK(G_UNICODE_CONNECT_PUNCTUATION),
>+            Punctuation_Other = U_MASK(G_UNICODE_OTHER_PUNCTUATION),
>+
>+            Symbol_Math = U_MASK(G_UNICODE_MATH_SYMBOL),
>+            Symbol_Currency = U_MASK(G_UNICODE_CURRENCY_SYMBOL),
>+            Symbol_Modifier = U_MASK(G_UNICODE_MODIFIER_SYMBOL),
>+            Symbol_Other = U_MASK(G_UNICODE_OTHER_SYMBOL),
>+
>+            Punctuation_InitialQuote = U_MASK(G_UNICODE_INITIAL_PUNCTUATION),
>+            Punctuation_FinalQuote = U_MASK(G_UNICODE_FINAL_PUNCTUATION)
>+        };
>+
>+        UChar32 foldCase(UChar32);
>+
>+        int foldCase(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error);
>+
>+        int toLower(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error);
>+
>+        inline UChar32 toLower(UChar32 c)
>+        {
>+            return g_unichar_tolower(c);
>+        }
>+
>+        inline UChar32 toUpper(UChar32 c)
>+        {
>+            return g_unichar_toupper(c);
>+        }
>+
>+        int toUpper(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error);
>+
>+        inline UChar32 toTitleCase(UChar32 c)
>+        {
>+            return g_unichar_totitle(c);
>+        }
>+
>+        inline bool isArabicChar(UChar32 c)
>+        {
>+            return c >= 0x0600 && c <= 0x06FF;
>+        }
>+
>+        inline bool isFormatChar(UChar32 c)
>+        {
>+            return g_unichar_type(c) == G_UNICODE_FORMAT;
>+        }
>+
>+        inline bool isSeparatorSpace(UChar32 c)
>+        {
>+            return g_unichar_type(c) == G_UNICODE_SPACE_SEPARATOR;
>+        }
>+
>+        inline bool isPrintableChar(UChar32 c)
>+        {
>+            return g_unichar_isprint(c);
>+        }
>+    
>+        inline bool isDigit(UChar32 c)
>+        {
>+            return g_unichar_isdigit(c);
>+        }
>+
>+        inline bool isPunct(UChar32 c)
>+        {
>+            return g_unichar_ispunct(c);
>+        }
>+
>+        inline UChar32 mirroredChar(UChar32 c)
>+        {
>+            gunichar mirror = 0;
>+            g_unichar_get_mirror_char(c, &mirror);
>+            return mirror;
>+        }
>+
>+        inline CharCategory category(UChar32 c)
>+        {
>+            if (c > 0xffff)
>+                return NoCategory;
>+
>+            return (CharCategory) U_MASK(g_unichar_type(c));
>+        }
>+
>+        Direction direction(UChar32);
>+
>+        inline bool isLower(UChar32 c)
>+        {
>+            return g_unichar_islower(c);
>+        }
>+
>+        inline int digitValue(UChar32 c)
>+        {
>+            return g_unichar_digit_value(c);
>+        }
>+
>+        inline uint8_t combiningClass(UChar32 c)
>+        {
>+            // FIXME
>+            // return g_unichar_combining_class(c);
>+            return 0;
>+        }
>+
>+        inline DecompositionType decompositionType(UChar32 c)
>+        {
>+            // FIXME
>+            return DecompositionNone;
>+        }
>+
>+        int umemcasecmp(const UChar*, const UChar*, int len);
>+
>+    }
>+}
>+
>+#endif
>+
>Index: JavaScriptCore/wtf/unicode/glib/UnicodeMacrosFromICU.h
>===================================================================
>--- JavaScriptCore/wtf/unicode/glib/UnicodeMacrosFromICU.h	(revision 0)
>+++ JavaScriptCore/wtf/unicode/glib/UnicodeMacrosFromICU.h	(revision 0)
>@@ -0,0 +1,58 @@
>+/*
>+ *  Copyright (C) 2006 George Staikos <staikos@kde.org>
>+ *  Copyright (C) 2006 Alexey Proskuryakov <ap@nypop.com>
>+ *  Copyright (C) 2007 Apple Computer, Inc. All rights reserved.
>+ *  Copyright (C) 2008 JÃ¼rg Billeter <j@bitron.ch>
>+ *  Copyright (C) 2008 Dominik RÃ¶ttsches <dominik.roettsches@access-company.com>
>+ *
>+ *  This library is free software; you can redistribute it and/or
>+ *  modify it under the terms of the GNU Library General Public
>+ *  License as published by the Free Software Foundation; either
>+ *  version 2 of the License, or (at your option) any later version.
>+ *
>+ *  This library 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
>+ *  Library General Public License for more details.
>+ *
>+ *  You should have received a copy of the GNU Library General Public License
>+ *  along with this library; see the file COPYING.LIB.  If not, write to
>+ *  the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
>+ *  Boston, MA 02110-1301, USA.
>+ *
>+ */
>+
>+#ifndef UnicodeMacrosFromICU_h
>+#define UnicodeMacrosFromICU_h
>+
>+// some defines from ICU
>+
>+#define U16_IS_LEAD(c) (((c)&0xfffffc00)==0xd800)
>+#define U16_IS_TRAIL(c) (((c)&0xfffffc00)==0xdc00)
>+#define U16_SURROGATE_OFFSET ((0xd800<<10UL)+0xdc00-0x10000)
>+#define U16_GET_SUPPLEMENTARY(lead, trail) \
>+    (((UChar32)(lead)<<10UL)+(UChar32)(trail)-U16_SURROGATE_OFFSET)
>+
>+#define U16_LEAD(supplementary) (UChar)(((supplementary)>>10)+0xd7c0)
>+#define U16_TRAIL(supplementary) (UChar)(((supplementary)&0x3ff)|0xdc00)
>+
>+#define U_IS_SURROGATE(c) (((c)&0xfffff800)==0xd800)
>+#define U16_IS_SINGLE(c) !U_IS_SURROGATE(c)
>+#define U16_IS_SURROGATE(c) U_IS_SURROGATE(c)
>+#define U16_IS_SURROGATE_LEAD(c) (((c)&0x400)==0)
>+
>+#define U16_NEXT(s, i, length, c) { \
>+    (c)=(s)[(i)++]; \
>+    if(U16_IS_LEAD(c)) { \
>+        uint16_t __c2; \
>+        if((i)<(length) && U16_IS_TRAIL(__c2=(s)[(i)])) { \
>+            ++(i); \
>+            (c)=U16_GET_SUPPLEMENTARY((c), __c2); \
>+        } \
>+    } \
>+}
>+
>+#define U_MASK(x) ((uint32_t)1<<(x))
>+
>+#endif
>+
>Index: WebCore/ChangeLog
>===================================================================
>--- WebCore/ChangeLog	(revision 39967)
>+++ WebCore/ChangeLog	(working copy)
>@@ -1,3 +1,30 @@
>+2009-01-16  Dominik RÃ¶ttsches  <dominik.roettsches@access-company.com>
>+
>+        Reviewed by NOBODY (OOPS!).
>+
>+        https://bugs.webkit.org/show_bug.cgi?id=15914
>+        [GTK] Implement Unicode functionality using GLib
>+
>+        Minor changes that allow building a hybrid version of WebKit
>+        that has WTF Unicode backend based on GLib functions while
>+        text codecs and TextBreakIterator remain based on ICU.
>+
>+        Fixed some includes that were assuming a WTF ICU backend and
>+        introduced temporary usage of WTF_USE_GLIB_ICU_UNICODE_HYBRID
>+        macro.
>+
>+        * platform/ThreadGlobalData.cpp:
>+        (WebCore::ThreadGlobalData::ThreadGlobalData):
>+        (WebCore::ThreadGlobalData::~ThreadGlobalData):
>+        * platform/ThreadGlobalData.h:
>+        * platform/text/TextBoundariesICU.cpp:
>+        * platform/text/TextCodecICU.h:
>+        * platform/text/TextEncoding.cpp:
>+        (WebCore::TextEncoding::encode):
>+        * platform/text/TextEncodingRegistry.cpp:
>+        (WebCore::buildBaseTextCodecMaps):
>+        (WebCore::extendTextCodecMaps):
>+
> 2009-01-16  Ariya Hidayat  <ariya.hidayat@trolltech.com>
> 
>         Another attempt at fixing the Qt build.
>Index: WebCore/platform/ThreadGlobalData.cpp
>===================================================================
>--- WebCore/platform/ThreadGlobalData.cpp	(revision 39967)
>+++ WebCore/platform/ThreadGlobalData.cpp	(working copy)
>@@ -31,7 +31,7 @@
> #include "StringImpl.h"
> #include <wtf/UnusedParam.h>
> 
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
> #include "TextCodecICU.h"
> #endif
> 
>@@ -70,7 +70,7 @@ ThreadGlobalData::ThreadGlobalData()
>     : m_emptyString(new StringImpl)
>     , m_atomicStringTable(new HashSet<StringImpl*>)
>     , m_eventNames(new EventNames)
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
>     , m_cachedConverterICU(new ICUConverterWrapper)
> #endif
> #if PLATFORM(MAC)
>@@ -84,7 +84,7 @@ ThreadGlobalData::~ThreadGlobalData()
> #if PLATFORM(MAC)
>     delete m_cachedConverterTEC;
> #endif
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
>     delete m_cachedConverterICU;
> #endif
> 
>Index: WebCore/platform/ThreadGlobalData.h
>===================================================================
>--- WebCore/platform/ThreadGlobalData.h	(revision 39967)
>+++ WebCore/platform/ThreadGlobalData.h	(working copy)
>@@ -46,7 +46,7 @@ namespace WebCore {
>         StringImpl* emptyString() { return m_emptyString; }
>         HashSet<StringImpl*>& atomicStringTable() { return *m_atomicStringTable; }
> 
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
>         ICUConverterWrapper& cachedConverterICU() { return *m_cachedConverterICU; }
> #endif
> 
>@@ -59,7 +59,7 @@ namespace WebCore {
>         HashSet<StringImpl*>* m_atomicStringTable;
>         EventNames* m_eventNames;
> 
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
>         ICUConverterWrapper* m_cachedConverterICU;
> #endif
> 
>Index: WebCore/platform/text/TextBoundariesICU.cpp
>===================================================================
>--- WebCore/platform/text/TextBoundariesICU.cpp	(revision 39967)
>+++ WebCore/platform/text/TextBoundariesICU.cpp	(working copy)
>@@ -27,6 +27,7 @@
> #include "TextBoundaries.h"
> 
> #include <unicode/ubrk.h>
>+#include <unicode/uchar.h>
> 
> #include "StringImpl.h"
> #include "TextBreakIterator.h"
>Index: WebCore/platform/text/TextCodecICU.h
>===================================================================
>--- WebCore/platform/text/TextCodecICU.h	(revision 39967)
>+++ WebCore/platform/text/TextCodecICU.h	(working copy)
>@@ -30,6 +30,8 @@
> #include "TextCodec.h"
> #include "TextEncoding.h"
> 
>+#include <unicode/utypes.h>
>+
> typedef struct UConverter UConverter;
> 
> namespace WebCore {
>Index: WebCore/platform/text/TextEncoding.cpp
>===================================================================
>--- WebCore/platform/text/TextEncoding.cpp	(revision 39967)
>+++ WebCore/platform/text/TextEncoding.cpp	(working copy)
>@@ -32,7 +32,7 @@
> #include "TextCodec.h"
> #include "TextDecoder.h"
> #include "TextEncodingRegistry.h"
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
> #include <unicode/unorm.h>
> #elif USE(QT4_UNICODE)
> #include <QString>
>@@ -82,7 +82,7 @@ CString TextEncoding::encode(const UChar
>     if (!length)
>         return "";
> 
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
>     // FIXME: What's the right place to do normalization?
>     // It's a little strange to do it inside the encode function.
>     // Perhaps normalization should be an explicit step done before calling encode.
>Index: WebCore/platform/text/TextEncodingRegistry.cpp
>===================================================================
>--- WebCore/platform/text/TextEncodingRegistry.cpp	(revision 39967)
>+++ WebCore/platform/text/TextEncodingRegistry.cpp	(working copy)
>@@ -38,7 +38,7 @@
> #include <wtf/StringExtras.h>
> #include <wtf/Threading.h>
> 
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
> #include "TextCodecICU.h"
> #endif
> #if PLATFORM(MAC)
>@@ -185,7 +185,7 @@ static void buildBaseTextCodecMaps()
>     TextCodecUserDefined::registerEncodingNames(addToTextEncodingNameMap);
>     TextCodecUserDefined::registerCodecs(addToTextCodecMap);
> 
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
>     TextCodecICU::registerBaseEncodingNames(addToTextEncodingNameMap);
>     TextCodecICU::registerBaseCodecs(addToTextCodecMap);
> #endif
>@@ -193,7 +193,7 @@ static void buildBaseTextCodecMaps()
> 
> static void extendTextCodecMaps()
> {
>-#if USE(ICU_UNICODE)
>+#if USE(ICU_UNICODE) || USE(GLIB_ICU_UNICODE_HYBRID)
>     TextCodecICU::registerExtendedEncodingNames(addToTextEncodingNameMap);
>     TextCodecICU::registerExtendedCodecs(addToTextCodecMap);
> #endif

Actions: View | Formatted Diff | Diff

Attachments on bug 15914: 18852 | 20824 | 21500 | 23001 | 23031 | 23124 | 23167 | 23263 | 25314 | 25348 | 25627 | 25628 | 25629 | 26711 | 26712 | 26793 | 26794 | 41434 | 41866 | 42210 | 42465