Menu
▾
▴
[xmlwrapp-commits] SF.net SVN: xmlwrapp:[200] trunk/src
|
From: <tbr...@us...> - 2012-03-20 13:35:22
|
Revision: 200
http://xmlwrapp.svn.sourceforge.net/xmlwrapp/?rev=200&view=rev
Author: tbrowder2
Date: 2012-03-20 13:35:13 +0000 (Tue, 20 Mar 2012)
Log Message:
-----------
add ustring source from glibmm project
Modified Paths:
--------------
trunk/src/Makefile.am
Added Paths:
-----------
trunk/src/libustring/
trunk/src/libustring/ustring.cc
trunk/src/libustring/ustring.h
Modified: trunk/src/Makefile.am
===================================================================
--- trunk/src/Makefile.am 2012-03-20 13:21:45 UTC (rev 199)
+++ trunk/src/Makefile.am 2012-03-20 13:35:13 UTC (rev 200)
@@ -2,16 +2,24 @@
AM_CPPFLAGS = -I$(top_srcdir)/include $(CXXFLAGS_VISIBILITY)
if WITH_XSLT
-lib_LTLIBRARIES = libxmlwrapp.la libxsltwrapp.la
+lib_LTLIBRARIES = libxmlwrapp.la libxsltwrapp.la libustring.la
else
-lib_LTLIBRARIES = libxmlwrapp.la
+lib_LTLIBRARIES = libxmlwrapp.la libustring.la
endif
+libustring_la_CPPFLAGS = $(AM_CPPFLAGS) $(LIBXML_CFLAGS)
+libustring_la_LIBADD = $(LIBXML_LIBS)
+libustring_la_LDFLAGS = -version-info 6:0:1 -no-undefined
+
+libustring_la_SOURCES = \
+ libustring/ustring.cc \
+ libustring/ustring.h \
+
libxmlwrapp_la_CPPFLAGS = $(AM_CPPFLAGS) $(LIBXML_CFLAGS)
-libxmlwrapp_la_LIBADD = $(LIBXML_LIBS)
-libxmlwrapp_la_LDFLAGS = -version-info 6:0:1 -no-undefined
+libxmlwrapp_la_LIBADD = $(LIBXML_LIBS)
+libxmlwrapp_la_LDFLAGS = -version-info 6:0:1 -no-undefined
-libxmlwrapp_la_SOURCES = \
+libxmlwrapp_la_SOURCES = \
libxml/ait_impl.cc \
libxml/ait_impl.h \
libxml/attributes.cc \
Added: trunk/src/libustring/ustring.cc
===================================================================
--- trunk/src/libustring/ustring.cc (rev 0)
+++ trunk/src/libustring/ustring.cc 2012-03-20 13:35:13 UTC (rev 200)
@@ -0,0 +1,1418 @@
+// -*- c++ -*-
+/* $Id$ */
+
+/* Copyright (C) 2002 The gtkmm Development Team
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free
+ * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <glibmmconfig.h>
+#include <glibmm/ustring.h>
+#include <glibmm/convert.h>
+#include <glibmm/error.h>
+#include <glibmm/utility.h>
+
+#include <algorithm>
+#include <iostream>
+#include <cstring>
+# include <stdexcept>
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+
+namespace
+{
+
+using Glib::ustring;
+
+// Little helper to make the conversion from gunichar to UTF-8 a one-liner.
+//
+struct UnicharToUtf8
+{
+ char buf[6];
+ ustring::size_type len;
+
+ explicit UnicharToUtf8(gunichar uc)
+ : len (g_unichar_to_utf8(uc, buf)) {}
+};
+
+
+// All utf8_*_offset() functions return npos if offset is out of range.
+// The caller should decide if npos is a valid argument and just marks
+// the whole string, or if it is not allowed (e.g. for start positions).
+// In the latter case std::out_of_range should be thrown, but usually
+// std::string will do that for us.
+
+// First overload: stop on '\0' character.
+static
+ustring::size_type utf8_byte_offset(const char* str, ustring::size_type offset)
+{
+ if(offset == ustring::npos)
+ return ustring::npos;
+
+ const char *const utf8_skip = g_utf8_skip;
+ const char* p = str;
+
+ for(; offset != 0; --offset)
+ {
+ const unsigned int c = static_cast<unsigned char>(*p);
+
+ if(c == 0)
+ return ustring::npos;
+
+ p += utf8_skip[c];
+ }
+
+ return (p - str);
+}
+
+// Second overload: stop when reaching maxlen.
+static
+ustring::size_type utf8_byte_offset(const char* str, ustring::size_type offset,
+ ustring::size_type maxlen)
+{
+ if(offset == ustring::npos)
+ return ustring::npos;
+
+ const char *const utf8_skip = g_utf8_skip;
+ const char *const pend = str + maxlen;
+ const char* p = str;
+
+ for(; offset != 0; --offset)
+ {
+ if(p >= pend)
+ return ustring::npos;
+
+ p += utf8_skip[static_cast<unsigned char>(*p)];
+ }
+
+ return (p - str);
+}
+
+// Third overload: stop when reaching str.size().
+//
+inline
+ustring::size_type utf8_byte_offset(const std::string& str, ustring::size_type offset)
+{
+ return utf8_byte_offset(str.data(), offset, str.size());
+}
+
+// Takes UTF-8 character offset and count in ci and cn.
+// Returns the byte offset and count in i and n.
+//
+struct Utf8SubstrBounds
+{
+ ustring::size_type i;
+ ustring::size_type n;
+
+ Utf8SubstrBounds(const std::string& str, ustring::size_type ci, ustring::size_type cn)
+ :
+ i (utf8_byte_offset(str, ci)),
+ n (ustring::npos)
+ {
+ if(i != ustring::npos)
+ n = utf8_byte_offset(str.data() + i, cn, str.size() - i);
+ }
+};
+
+// Converts byte offset to UTF-8 character offset.
+inline
+ustring::size_type utf8_char_offset(const std::string& str, ustring::size_type offset)
+{
+ if(offset == ustring::npos)
+ return ustring::npos;
+
+ const char *const pdata = str.data();
+ return g_utf8_pointer_to_offset(pdata, pdata + offset);
+}
+
+
+// Helper to implement ustring::find_first_of() and find_first_not_of().
+// Returns the UTF-8 character offset, or ustring::npos if not found.
+static
+ustring::size_type utf8_find_first_of(const std::string& str, ustring::size_type offset,
+ const char* utf8_match, long utf8_match_size,
+ bool find_not_of)
+{
+ const ustring::size_type byte_offset = utf8_byte_offset(str, offset);
+ if(byte_offset == ustring::npos)
+ return ustring::npos;
+
+ long ucs4_match_size = 0;
+ const Glib::ScopedPtr<gunichar> ucs4_match
+ (g_utf8_to_ucs4_fast(utf8_match, utf8_match_size, &ucs4_match_size));
+
+ const gunichar *const match_begin = ucs4_match.get();
+ const gunichar *const match_end = match_begin + ucs4_match_size;
+
+ const char *const str_begin = str.data();
+ const char *const str_end = str_begin + str.size();
+
+ for(const char* pstr = str_begin + byte_offset;
+ pstr < str_end;
+ pstr = g_utf8_next_char(pstr))
+ {
+ const gunichar *const pfound = std::find(match_begin, match_end, g_utf8_get_char(pstr));
+
+ if((pfound != match_end) != find_not_of)
+ return offset;
+
+ ++offset;
+ }
+
+ return ustring::npos;
+}
+
+// Helper to implement ustring::find_last_of() and find_last_not_of().
+// Returns the UTF-8 character offset, or ustring::npos if not found.
+static
+ustring::size_type utf8_find_last_of(const std::string& str, ustring::size_type offset,
+ const char* utf8_match, long utf8_match_size,
+ bool find_not_of)
+{
+ long ucs4_match_size = 0;
+ const Glib::ScopedPtr<gunichar> ucs4_match
+ (g_utf8_to_ucs4_fast(utf8_match, utf8_match_size, &ucs4_match_size));
+
+ const gunichar *const match_begin = ucs4_match.get();
+ const gunichar *const match_end = match_begin + ucs4_match_size;
+
+ const char *const str_begin = str.data();
+ const char* pstr = str_begin;
+
+ // Set pstr one byte beyond the actual start position.
+ const ustring::size_type byte_offset = utf8_byte_offset(str, offset);
+ pstr += (byte_offset < str.size()) ? byte_offset + 1 : str.size();
+
+ while(pstr > str_begin)
+ {
+ // Move to previous character.
+ do
+ --pstr;
+ while((static_cast<unsigned char>(*pstr) & 0xC0u) == 0x80);
+
+ const gunichar *const pfound = std::find(match_begin, match_end, g_utf8_get_char(pstr));
+
+ if((pfound != match_end) != find_not_of)
+ return g_utf8_pointer_to_offset(str_begin, pstr);
+ }
+
+ return ustring::npos;
+}
+
+} // anonymous namespace
+
+
+namespace Glib
+{
+
+#ifndef GLIBMM_HAVE_ALLOWS_STATIC_INLINE_NPOS
+// Initialize static member here,
+// because the compiler did not allow us do it inline.
+const ustring::size_type ustring::npos = std::string::npos;
+#endif
+
+/*
+ * We need our own version of g_utf8_get_char(), because the std::string
+ * iterator is not necessarily a plain pointer (it's in fact not in GCC's
+ * libstdc++-v3). Copying the UTF-8 data into a temporary buffer isn't an
+ * option since this operation is quite time critical. The implementation
+ * is quite different from g_utf8_get_char() -- both more generic and likely
+ * faster.
+ *
+ * By looking at the first byte of a UTF-8 character one can determine the
+ * number of bytes used. GLib offers the g_utf8_skip[] array for this purpose,
+ * but accessing this global variable would, on IA32 at least, introduce
+ * a function call to fetch the Global Offset Table, plus two levels of
+ * indirection in order to read the value. Even worse, fetching the GOT is
+ * always done right at the start of the function instead of the branch that
+ * actually uses the variable.
+ *
+ * Fortunately, there's a better way to get the byte count. As this table
+ * shows, there's a nice regular pattern in the UTF-8 encoding scheme:
+ *
+ * 0x00000000 - 0x0000007F: 0xxxxxxx
+ * 0x00000080 - 0x000007FF: 110xxxxx 10xxxxxx
+ * 0x00000800 - 0x0000FFFF: 1110xxxx 10xxxxxx 10xxxxxx
+ * 0x00010000 - 0x001FFFFF: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
+ * 0x00200000 - 0x03FFFFFF: 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
+ * 0x04000000 - 0x7FFFFFFF: 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
+ *
+ * Except for the single byte case, the number of leading 1-bits equals the
+ * byte count. All that is needed is to shift the first byte to the left
+ * until bit 7 becomes 0. Naturally, doing so requires a loop -- but since
+ * we already have one, no additional cost is introduced. This shifting can
+ * further be combined with the computation of the bitmask needed to eliminate
+ * the leading length bits, thus saving yet another register.
+ *
+ * Note: If you change this code, it is advisable to also review what the
+ * compiler makes of it in the assembler output. Except for some pointless
+ * register moves, the generated code is sufficiently close to the optimum
+ * with GCC 4.1.2 on x86_64.
+ */
+gunichar get_unichar_from_std_iterator(std::string::const_iterator pos)
+{
+ unsigned int result = static_cast<unsigned char>(*pos);
+
+ if((result & 0x80) != 0)
+ {
+ unsigned int mask = 0x40;
+
+ do
+ {
+ result <<= 6;
+ const unsigned int c = static_cast<unsigned char>(*++pos);
+ mask <<= 5;
+ result += c - 0x80;
+ }
+ while((result & mask) != 0);
+
+ result &= mask - 1;
+ }
+
+ return result;
+}
+
+
+/**** Glib::ustring ********************************************************/
+
+ustring::ustring()
+:
+ string_ ()
+{}
+
+ustring::ustring(const ustring& other)
+:
+ string_ (other.string_)
+{}
+
+ustring::ustring(const ustring& src, ustring::size_type i, ustring::size_type n)
+:
+ string_ ()
+{
+ const Utf8SubstrBounds bounds (src.string_, i, n);
+ string_.assign(src.string_, bounds.i, bounds.n);
+}
+
+ustring::ustring(const char* src, ustring::size_type n)
+:
+ string_ (src, utf8_byte_offset(src, n))
+{}
+
+ustring::ustring(const char* src)
+:
+ string_ (src)
+{}
+
+ustring::ustring(ustring::size_type n, gunichar uc)
+:
+ string_ ()
+{
+ if(uc < 0x80)
+ {
+ // Optimize the probably most common case.
+ string_.assign(n, static_cast<char>(uc));
+ }
+ else
+ {
+ const UnicharToUtf8 conv (uc);
+ string_.reserve(n * conv.len);
+
+ for(; n > 0; --n)
+ string_.append(conv.buf, conv.len);
+ }
+}
+
+ustring::ustring(ustring::size_type n, char c)
+:
+ string_ (n, c)
+{}
+
+ustring::ustring(const std::string& src)
+:
+ string_ (src)
+{}
+
+ustring::~ustring()
+{}
+
+void ustring::swap(ustring& other)
+{
+ string_.swap(other.string_);
+}
+
+
+/**** Glib::ustring::operator=() *******************************************/
+
+ustring& ustring::operator=(const ustring& other)
+{
+ string_ = other.string_;
+ return *this;
+}
+
+ustring& ustring::operator=(const std::string& src)
+{
+ string_ = src;
+ return *this;
+}
+
+ustring& ustring::operator=(const char* src)
+{
+ string_ = src;
+ return *this;
+}
+
+ustring& ustring::operator=(gunichar uc)
+{
+ const UnicharToUtf8 conv (uc);
+ string_.assign(conv.buf, conv.len);
+ return *this;
+}
+
+ustring& ustring::operator=(char c)
+{
+ string_ = c;
+ return *this;
+}
+
+
+/**** Glib::ustring::assign() **********************************************/
+
+ustring& ustring::assign(const ustring& src)
+{
+ string_ = src.string_;
+ return *this;
+}
+
+ustring& ustring::assign(const ustring& src, ustring::size_type i, ustring::size_type n)
+{
+ const Utf8SubstrBounds bounds (src.string_, i, n);
+ string_.assign(src.string_, bounds.i, bounds.n);
+ return *this;
+}
+
+ustring& ustring::assign(const char* src, ustring::size_type n)
+{
+ string_.assign(src, utf8_byte_offset(src, n));
+ return *this;
+}
+
+ustring& ustring::assign(const char* src)
+{
+ string_ = src;
+ return *this;
+}
+
+ustring& ustring::assign(ustring::size_type n, gunichar uc)
+{
+ ustring temp (n, uc);
+ string_.swap(temp.string_);
+ return *this;
+}
+
+ustring& ustring::assign(ustring::size_type n, char c)
+{
+ string_.assign(n, c);
+ return *this;
+}
+
+
+/**** Glib::ustring::operator+=() ******************************************/
+
+ustring& ustring::operator+=(const ustring& src)
+{
+ string_ += src.string_;
+ return *this;
+}
+
+ustring& ustring::operator+=(const char* src)
+{
+ string_ += src;
+ return *this;
+}
+
+ustring& ustring::operator+=(gunichar uc)
+{
+ const UnicharToUtf8 conv (uc);
+ string_.append(conv.buf, conv.len);
+ return *this;
+}
+
+ustring& ustring::operator+=(char c)
+{
+ string_ += c;
+ return *this;
+}
+
+
+/**** Glib::ustring::push_back() *******************************************/
+
+void ustring::push_back(gunichar uc)
+{
+ const UnicharToUtf8 conv (uc);
+ string_.append(conv.buf, conv.len);
+}
+
+void ustring::push_back(char c)
+{
+ string_ += c;
+}
+
+
+/**** Glib::ustring::append() **********************************************/
+
+ustring& ustring::append(const ustring& src)
+{
+ string_ += src.string_;
+ return *this;
+}
+
+ustring& ustring::append(const ustring& src, ustring::size_type i, ustring::size_type n)
+{
+ const Utf8SubstrBounds bounds (src.string_, i, n);
+ string_.append(src.string_, bounds.i, bounds.n);
+ return *this;
+}
+
+ustring& ustring::append(const char* src, ustring::size_type n)
+{
+ string_.append(src, utf8_byte_offset(src, n));
+ return *this;
+}
+
+ustring& ustring::append(const char* src)
+{
+ string_ += src;
+ return *this;
+}
+
+ustring& ustring::append(ustring::size_type n, gunichar uc)
+{
+ string_.append(ustring(n, uc).string_);
+ return *this;
+}
+
+ustring& ustring::append(ustring::size_type n, char c)
+{
+ string_.append(n, c);
+ return *this;
+}
+
+
+/**** Glib::ustring::insert() **********************************************/
+
+ustring& ustring::insert(ustring::size_type i, const ustring& src)
+{
+ string_.insert(utf8_byte_offset(string_, i), src.string_);
+ return *this;
+}
+
+ustring& ustring::insert(ustring::size_type i, const ustring& src,
+ ustring::size_type i2, ustring::size_type n)
+{
+ const Utf8SubstrBounds bounds2 (src.string_, i2, n);
+ string_.insert(utf8_byte_offset(string_, i), src.string_, bounds2.i, bounds2.n);
+ return *this;
+}
+
+ustring& ustring::insert(ustring::size_type i, const char* src, ustring::size_type n)
+{
+ string_.insert(utf8_byte_offset(string_, i), src, utf8_byte_offset(src, n));
+ return *this;
+}
+
+ustring& ustring::insert(ustring::size_type i, const char* src)
+{
+ string_.insert(utf8_byte_offset(string_, i), src);
+ return *this;
+}
+
+ustring& ustring::insert(ustring::size_type i, ustring::size_type n, gunichar uc)
+{
+ string_.insert(utf8_byte_offset(string_, i), ustring(n, uc).string_);
+ return *this;
+}
+
+ustring& ustring::insert(ustring::size_type i, ustring::size_type n, char c)
+{
+ string_.insert(utf8_byte_offset(string_, i), n, c);
+ return *this;
+}
+
+ustring::iterator ustring::insert(ustring::iterator p, gunichar uc)
+{
+ const size_type offset = p.base() - string_.begin();
+ const UnicharToUtf8 conv (uc);
+ string_.insert(offset, conv.buf, conv.len);
+ return iterator(string_.begin() + offset);
+}
+
+ustring::iterator ustring::insert(ustring::iterator p, char c)
+{
+ return iterator(string_.insert(p.base(), c));
+}
+
+void ustring::insert(ustring::iterator p, ustring::size_type n, gunichar uc)
+{
+ string_.insert(p.base() - string_.begin(), ustring(n, uc).string_);
+}
+
+void ustring::insert(ustring::iterator p, ustring::size_type n, char c)
+{
+ string_.insert(p.base(), n, c);
+}
+
+
+/**** Glib::ustring::replace() *********************************************/
+
+ustring& ustring::replace(ustring::size_type i, ustring::size_type n, const ustring& src)
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ string_.replace(bounds.i, bounds.n, src.string_);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::size_type i, ustring::size_type n,
+ const ustring& src, ustring::size_type i2, ustring::size_type n2)
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ const Utf8SubstrBounds bounds2 (src.string_, i2, n2);
+ string_.replace(bounds.i, bounds.n, src.string_, bounds2.i, bounds2.n);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::size_type i, ustring::size_type n,
+ const char* src, ustring::size_type n2)
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ string_.replace(bounds.i, bounds.n, src, utf8_byte_offset(src, n2));
+ return *this;
+}
+
+ustring& ustring::replace(ustring::size_type i, ustring::size_type n, const char* src)
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ string_.replace(bounds.i, bounds.n, src);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::size_type i, ustring::size_type n,
+ ustring::size_type n2, gunichar uc)
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ string_.replace(bounds.i, bounds.n, ustring(n2, uc).string_);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::size_type i, ustring::size_type n,
+ ustring::size_type n2, char c)
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ string_.replace(bounds.i, bounds.n, n2, c);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::iterator pbegin, ustring::iterator pend, const ustring& src)
+{
+ string_.replace(pbegin.base(), pend.base(), src.string_);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::iterator pbegin, ustring::iterator pend,
+ const char* src, ustring::size_type n)
+{
+ string_.replace(pbegin.base(), pend.base(), src, utf8_byte_offset(src, n));
+ return *this;
+}
+
+ustring& ustring::replace(ustring::iterator pbegin, ustring::iterator pend, const char* src)
+{
+ string_.replace(pbegin.base(), pend.base(), src);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::iterator pbegin, ustring::iterator pend,
+ ustring::size_type n, gunichar uc)
+{
+ string_.replace(pbegin.base(), pend.base(), ustring(n, uc).string_);
+ return *this;
+}
+
+ustring& ustring::replace(ustring::iterator pbegin, ustring::iterator pend,
+ ustring::size_type n, char c)
+{
+ string_.replace(pbegin.base(), pend.base(), n, c);
+ return *this;
+}
+
+
+/**** Glib::ustring::erase() ***********************************************/
+
+void ustring::clear()
+{
+ string_.erase();
+}
+
+ustring& ustring::erase(ustring::size_type i, ustring::size_type n)
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ string_.erase(bounds.i, bounds.n);
+ return *this;
+}
+
+ustring& ustring::erase()
+{
+ string_.erase();
+ return *this;
+}
+
+ustring::iterator ustring::erase(ustring::iterator p)
+{
+ ustring::iterator iter_end = p;
+ ++iter_end;
+
+ return iterator(string_.erase(p.base(), iter_end.base()));
+}
+
+ustring::iterator ustring::erase(ustring::iterator pbegin, ustring::iterator pend)
+{
+ return iterator(string_.erase(pbegin.base(), pend.base()));
+}
+
+
+/**** Glib::ustring::compare() *********************************************/
+
+int ustring::compare(const ustring& rhs) const
+{
+ return g_utf8_collate(string_.c_str(), rhs.string_.c_str());
+}
+
+int ustring::compare(const char* rhs) const
+{
+ return g_utf8_collate(string_.c_str(), rhs);
+}
+
+int ustring::compare(ustring::size_type i, ustring::size_type n, const ustring& rhs) const
+{
+ return ustring(*this, i, n).compare(rhs);
+}
+
+int ustring::compare(ustring::size_type i, ustring::size_type n,
+ const ustring& rhs, ustring::size_type i2, ustring::size_type n2) const
+{
+ return ustring(*this, i, n).compare(ustring(rhs, i2, n2));
+}
+
+int ustring::compare(ustring::size_type i, ustring::size_type n,
+ const char* rhs, ustring::size_type n2) const
+{
+ return ustring(*this, i, n).compare(ustring(rhs, n2));
+}
+
+int ustring::compare(ustring::size_type i, ustring::size_type n, const char* rhs) const
+{
+ return ustring(*this, i, n).compare(rhs);
+}
+
+
+/**** Glib::ustring -- index access ****************************************/
+
+ustring::value_type ustring::operator[](ustring::size_type i) const
+{
+ return g_utf8_get_char(g_utf8_offset_to_pointer(string_.data(), i));
+}
+
+ustring::value_type ustring::at(ustring::size_type i) const
+{
+ const size_type byte_offset = utf8_byte_offset(string_, i);
+
+ // Throws std::out_of_range if the index is invalid.
+ return g_utf8_get_char(&string_.at(byte_offset));
+}
+
+
+/**** Glib::ustring -- iterator access *************************************/
+
+ustring::iterator ustring::begin()
+{
+ return iterator(string_.begin());
+}
+
+ustring::iterator ustring::end()
+{
+ return iterator(string_.end());
+}
+
+ustring::const_iterator ustring::begin() const
+{
+ return const_iterator(string_.begin());
+}
+
+ustring::const_iterator ustring::end() const
+{
+ return const_iterator(string_.end());
+}
+
+ustring::reverse_iterator ustring::rbegin()
+{
+ return reverse_iterator(iterator(string_.end()));
+}
+
+ustring::reverse_iterator ustring::rend()
+{
+ return reverse_iterator(iterator(string_.begin()));
+}
+
+ustring::const_reverse_iterator ustring::rbegin() const
+{
+ return const_reverse_iterator(const_iterator(string_.end()));
+}
+
+ustring::const_reverse_iterator ustring::rend() const
+{
+ return const_reverse_iterator(const_iterator(string_.begin()));
+}
+
+
+/**** Glib::ustring::find() ************************************************/
+
+ustring::size_type ustring::find(const ustring& str, ustring::size_type i) const
+{
+ return utf8_char_offset(string_, string_.find(str.string_, utf8_byte_offset(string_, i)));
+}
+
+ustring::size_type ustring::find(const char* str, ustring::size_type i, ustring::size_type n) const
+{
+ return utf8_char_offset(string_, string_.find(str, utf8_byte_offset(string_, i),
+ utf8_byte_offset(str, n)));
+}
+
+ustring::size_type ustring::find(const char* str, ustring::size_type i) const
+{
+ return utf8_char_offset(string_, string_.find(str, utf8_byte_offset(string_, i)));
+}
+
+ustring::size_type ustring::find(gunichar uc, ustring::size_type i) const
+{
+ const UnicharToUtf8 conv (uc);
+ return utf8_char_offset(string_, string_.find(conv.buf, utf8_byte_offset(string_, i), conv.len));
+}
+
+ustring::size_type ustring::find(char c, ustring::size_type i) const
+{
+ return utf8_char_offset(string_, string_.find(c, utf8_byte_offset(string_, i)));
+}
+
+
+/**** Glib::ustring::rfind() ***********************************************/
+
+ustring::size_type ustring::rfind(const ustring& str, ustring::size_type i) const
+{
+ return utf8_char_offset(string_, string_.rfind(str.string_, utf8_byte_offset(string_, i)));
+}
+
+ustring::size_type ustring::rfind(const char* str, ustring::size_type i,
+ ustring::size_type n) const
+{
+ return utf8_char_offset(string_, string_.rfind(str, utf8_byte_offset(string_, i),
+ utf8_byte_offset(str, n)));
+}
+
+ustring::size_type ustring::rfind(const char* str, ustring::size_type i) const
+{
+ return utf8_char_offset(string_, string_.rfind(str, utf8_byte_offset(string_, i)));
+}
+
+ustring::size_type ustring::rfind(gunichar uc, ustring::size_type i) const
+{
+ const UnicharToUtf8 conv (uc);
+ return utf8_char_offset(string_, string_.rfind(conv.buf, utf8_byte_offset(string_, i), conv.len));
+}
+
+ustring::size_type ustring::rfind(char c, ustring::size_type i) const
+{
+ return utf8_char_offset(string_, string_.rfind(c, utf8_byte_offset(string_, i)));
+}
+
+
+/**** Glib::ustring::find_first_of() ***************************************/
+
+ustring::size_type ustring::find_first_of(const ustring& match, ustring::size_type i) const
+{
+ return utf8_find_first_of(string_, i, match.string_.data(), match.string_.size(), false);
+}
+
+ustring::size_type ustring::find_first_of(const char* match,
+ ustring::size_type i, ustring::size_type n) const
+{
+ return utf8_find_first_of(string_, i, match, n, false);
+}
+
+ustring::size_type ustring::find_first_of(const char* match, ustring::size_type i) const
+{
+ return utf8_find_first_of(string_, i, match, -1, false);
+}
+
+ustring::size_type ustring::find_first_of(gunichar uc, ustring::size_type i) const
+{
+ return find(uc, i);
+}
+
+ustring::size_type ustring::find_first_of(char c, ustring::size_type i) const
+{
+ return find(c, i);
+}
+
+
+/**** Glib::ustring::find_last_of() ****************************************/
+
+ustring::size_type ustring::find_last_of(const ustring& match, ustring::size_type i) const
+{
+ return utf8_find_last_of(string_, i, match.string_.data(), match.string_.size(), false);
+}
+
+ustring::size_type ustring::find_last_of(const char* match,
+ ustring::size_type i, ustring::size_type n) const
+{
+ return utf8_find_last_of(string_, i, match, n, false);
+}
+
+ustring::size_type ustring::find_last_of(const char* match, ustring::size_type i) const
+{
+ return utf8_find_last_of(string_, i, match, -1, false);
+}
+
+ustring::size_type ustring::find_last_of(gunichar uc, ustring::size_type i) const
+{
+ return rfind(uc, i);
+}
+
+ustring::size_type ustring::find_last_of(char c, ustring::size_type i) const
+{
+ return rfind(c, i);
+}
+
+
+/**** Glib::ustring::find_first_not_of() ***********************************/
+
+ustring::size_type ustring::find_first_not_of(const ustring& match, ustring::size_type i) const
+{
+ return utf8_find_first_of(string_, i, match.string_.data(), match.string_.size(), true);
+}
+
+ustring::size_type ustring::find_first_not_of(const char* match,
+ ustring::size_type i, ustring::size_type n) const
+{
+ return utf8_find_first_of(string_, i, match, n, true);
+}
+
+ustring::size_type ustring::find_first_not_of(const char* match, ustring::size_type i) const
+{
+ return utf8_find_first_of(string_, i, match, -1, true);
+}
+
+// Unfortunately, all of the find_*_not_of() methods for single
+// characters need their own special implementation.
+//
+ustring::size_type ustring::find_first_not_of(gunichar uc, ustring::size_type i) const
+{
+ const size_type bi = utf8_byte_offset(string_, i);
+ if(bi != npos)
+ {
+ const char *const pbegin = string_.data();
+ const char *const pend = pbegin + string_.size();
+
+ for(const char* p = pbegin + bi;
+ p < pend;
+ p = g_utf8_next_char(p), ++i)
+ {
+ if(g_utf8_get_char(p) != uc)
+ return i;
+ }
+ }
+ return npos;
+}
+
+ustring::size_type ustring::find_first_not_of(char c, ustring::size_type i) const
+{
+ const size_type bi = utf8_byte_offset(string_, i);
+ if(bi != npos)
+ {
+ const char *const pbegin = string_.data();
+ const char *const pend = pbegin + string_.size();
+
+ for(const char* p = pbegin + bi;
+ p < pend;
+ p = g_utf8_next_char(p), ++i)
+ {
+ if(*p != c)
+ return i;
+ }
+ }
+ return npos;
+}
+
+
+/**** Glib::ustring::find_last_not_of() ************************************/
+
+ustring::size_type ustring::find_last_not_of(const ustring& match, ustring::size_type i) const
+{
+ return utf8_find_last_of(string_, i, match.string_.data(), match.string_.size(), true);
+}
+
+ustring::size_type ustring::find_last_not_of(const char* match,
+ ustring::size_type i, ustring::size_type n) const
+{
+ return utf8_find_last_of(string_, i, match, n, true);
+}
+
+ustring::size_type ustring::find_last_not_of(const char* match, ustring::size_type i) const
+{
+ return utf8_find_last_of(string_, i, match, -1, true);
+}
+
+// Unfortunately, all of the find_*_not_of() methods for single
+// characters need their own special implementation.
+//
+ustring::size_type ustring::find_last_not_of(gunichar uc, ustring::size_type i) const
+{
+ const char *const pbegin = string_.data();
+ const char *const pend = pbegin + string_.size();
+ size_type i_cur = 0;
+ size_type i_found = npos;
+
+ for(const char* p = pbegin;
+ p < pend && i_cur <= i;
+ p = g_utf8_next_char(p), ++i_cur)
+ {
+ if(g_utf8_get_char(p) != uc)
+ i_found = i_cur;
+ }
+ return i_found;
+}
+
+ustring::size_type ustring::find_last_not_of(char c, ustring::size_type i) const
+{
+ const char *const pbegin = string_.data();
+ const char *const pend = pbegin + string_.size();
+ size_type i_cur = 0;
+ size_type i_found = npos;
+
+ for(const char* p = pbegin;
+ p < pend && i_cur <= i;
+ p = g_utf8_next_char(p), ++i_cur)
+ {
+ if(*p != c)
+ i_found = i_cur;
+ }
+ return i_found;
+}
+
+
+/**** Glib::ustring -- get size and resize *********************************/
+
+bool ustring::empty() const
+{
+ return string_.empty();
+}
+
+ustring::size_type ustring::size() const
+{
+ const char *const pdata = string_.data();
+ return g_utf8_pointer_to_offset(pdata, pdata + string_.size());
+}
+
+ustring::size_type ustring::length() const
+{
+ const char *const pdata = string_.data();
+ return g_utf8_pointer_to_offset(pdata, pdata + string_.size());
+}
+
+ustring::size_type ustring::bytes() const
+{
+ return string_.size();
+}
+
+ustring::size_type ustring::capacity() const
+{
+ return string_.capacity();
+}
+
+ustring::size_type ustring::max_size() const
+{
+ return string_.max_size();
+}
+
+void ustring::resize(ustring::size_type n, gunichar uc)
+{
+ const size_type size_now = size();
+ if(n < size_now)
+ erase(n, npos);
+ else if(n > size_now)
+ append(n - size_now, uc);
+}
+
+void ustring::resize(ustring::size_type n, char c)
+{
+ const size_type size_now = size();
+ if(n < size_now)
+ erase(n, npos);
+ else if(n > size_now)
+ string_.append(n - size_now, c);
+}
+
+void ustring::reserve(ustring::size_type n)
+{
+ string_.reserve(n);
+}
+
+
+/**** Glib::ustring -- C string access *************************************/
+
+const char* ustring::data() const
+{
+ return string_.data();
+}
+
+const char* ustring::c_str() const
+{
+ return string_.c_str();
+}
+
+// Note that copy() requests UTF-8 character offsets as
+// parameters, but returns the number of copied bytes.
+//
+ustring::size_type ustring::copy(char* dest, ustring::size_type n, ustring::size_type i) const
+{
+ const Utf8SubstrBounds bounds (string_, i, n);
+ return string_.copy(dest, bounds.n, bounds.i);
+}
+
+
+/**** Glib::ustring -- UTF-8 utilities *************************************/
+
+bool ustring::validate() const
+{
+ return (g_utf8_validate(string_.data(), string_.size(), 0) != 0);
+}
+
+bool ustring::validate(ustring::iterator& first_invalid)
+{
+ const char *const pdata = string_.data();
+ const char* valid_end = pdata;
+ const int is_valid = g_utf8_validate(pdata, string_.size(), &valid_end);
+
+ first_invalid = iterator(string_.begin() + (valid_end - pdata));
+ return (is_valid != 0);
+}
+
+bool ustring::validate(ustring::const_iterator& first_invalid) const
+{
+ const char *const pdata = string_.data();
+ const char* valid_end = pdata;
+ const int is_valid = g_utf8_validate(pdata, string_.size(), &valid_end);
+
+ first_invalid = const_iterator(string_.begin() + (valid_end - pdata));
+ return (is_valid != 0);
+}
+
+bool ustring::is_ascii() const
+{
+ const char* p = string_.data();
+ const char *const pend = p + string_.size();
+
+ for(; p != pend; ++p)
+ {
+ if((static_cast<unsigned char>(*p) & 0x80u) != 0)
+ return false;
+ }
+
+ return true;
+}
+
+ustring ustring::normalize(NormalizeMode mode) const
+{
+ const ScopedPtr<char> buf (g_utf8_normalize(string_.data(), string_.size(),
+ static_cast<GNormalizeMode>(int(mode))));
+ return ustring(buf.get());
+}
+
+ustring ustring::uppercase() const
+{
+ const ScopedPtr<char> buf (g_utf8_strup(string_.data(), string_.size()));
+ return ustring(buf.get());
+}
+
+ustring ustring::lowercase() const
+{
+ const ScopedPtr<char> buf (g_utf8_strdown(string_.data(), string_.size()));
+ return ustring(buf.get());
+}
+
+ustring ustring::casefold() const
+{
+ const ScopedPtr<char> buf (g_utf8_casefold(string_.data(), string_.size()));
+ return ustring(buf.get());
+}
+
+std::string ustring::collate_key() const
+{
+ const ScopedPtr<char> buf (g_utf8_collate_key(string_.data(), string_.size()));
+ return std::string(buf.get());
+}
+
+std::string ustring::casefold_collate_key() const
+{
+ char *const casefold_buf = g_utf8_casefold(string_.data(), string_.size());
+ char *const key_buf = g_utf8_collate_key(casefold_buf, -1);
+ g_free(casefold_buf);
+ return std::string(ScopedPtr<char>(key_buf).get());
+}
+
+/**** Glib::ustring -- Message formatting **********************************/
+
+// static
+ustring ustring::compose_argv(const Glib::ustring& fmt, int argc, const ustring* const* argv)
+{
+ std::string::size_type result_size = fmt.raw().size();
+
+ // Guesstimate the final string size.
+ for (int i = 0; i < argc; ++i)
+ result_size += argv[i]->raw().size();
+
+ std::string result;
+ result.reserve(result_size);
+
+ const char* const pfmt = fmt.raw().c_str();
+ const char* start = pfmt;
+
+ while (const char* const stop = std::strchr(start, '%'))
+ {
+ if (stop[1] == '%')
+ {
+ result.append(start, stop - start + 1);
+ start = stop + 2;
+ }
+ else
+ {
+ const int index = Ascii::digit_value(stop[1]) - 1;
+
+ if (index >= 0 && index < argc)
+ {
+ result.append(start, stop - start);
+ result += argv[index]->raw();
+ start = stop + 2;
+ }
+ else
+ {
+ const char* const next = (stop[1] != '\0') ? g_utf8_next_char(stop + 1) : (stop + 1);
+
+ // Copy invalid substitutions literally to the output.
+ result.append(start, next - start);
+
+ g_warning("invalid substitution \"%s\" in fmt string \"%s\"",
+ result.c_str() + result.size() - (next - stop), pfmt);
+ start = next;
+ }
+ }
+ }
+
+ result.append(start, pfmt + fmt.raw().size() - start);
+
+ return result;
+}
+
+/**** Glib::ustring::SequenceToString **************************************/
+
+ustring::SequenceToString<Glib::ustring::iterator,gunichar>
+ ::SequenceToString(Glib::ustring::iterator pbegin, Glib::ustring::iterator pend)
+:
+ std::string(pbegin.base(), pend.base())
+{}
+
+ustring::SequenceToString<Glib::ustring::const_iterator,gunichar>
+ ::SequenceToString(Glib::ustring::const_iterator pbegin, Glib::ustring::const_iterator pend)
+:
+ std::string(pbegin.base(), pend.base())
+{}
+
+/**** Glib::ustring::FormatStream ******************************************/
+
+ustring::FormatStream::FormatStream()
+:
+ stream_ ()
+{}
+
+ustring::FormatStream::~FormatStream()
+{}
+
+ustring ustring::FormatStream::to_string() const
+{
+ GError* error = 0;
+
+#ifdef GLIBMM_HAVE_WIDE_STREAM
+ const std::wstring str = stream_.str();
+
+# if defined(__STDC_ISO_10646__) && SIZEOF_WCHAR_T == 4
+ // Avoid going through iconv if wchar_t always contains UCS-4.
+ glong n_bytes = 0;
+ const ScopedPtr<char> buf (g_ucs4_to_utf8(reinterpret_cast<const gunichar*>(str.data()),
+ str.size(), 0, &n_bytes, &error));
+# elif defined(G_OS_WIN32) && SIZEOF_WCHAR_T == 2
+ // Avoid going through iconv if wchar_t always contains UTF-16.
+ glong n_bytes = 0;
+ const ScopedPtr<char> buf (g_utf16_to_utf8(reinterpret_cast<const gunichar2*>(str.data()),
+ str.size(), 0, &n_bytes, &error));
+# else
+ gsize n_bytes = 0;
+ const ScopedPtr<char> buf (g_convert(reinterpret_cast<const char*>(str.data()),
+ str.size() * sizeof(std::wstring::value_type),
+ "UTF-8", "WCHAR_T", 0, &n_bytes, &error));
+# endif /* !(__STDC_ISO_10646__ || G_OS_WIN32) */
+
+#else /* !GLIBMM_HAVE_WIDE_STREAM */
+ const std::string str = stream_.str();
+
+ gsize n_bytes = 0;
+ const ScopedPtr<char> buf (g_locale_to_utf8(str.data(), str.size(), 0, &n_bytes, &error));
+#endif /* !GLIBMM_HAVE_WIDE_STREAM */
+
+ if (error)
+ {
+ Glib::Error::throw_exception(error);
+ }
+
+ return ustring(buf.get(), buf.get() + n_bytes);
+}
+
+/**** Glib::ustring -- stream I/O operators ********************************/
+
+std::istream& operator>>(std::istream& is, Glib::ustring& utf8_string)
+{
+ std::string str;
+ is >> str;
+
+ GError* error = 0;
+ gsize n_bytes = 0;
+ const ScopedPtr<char> buf (g_locale_to_utf8(str.data(), str.size(), 0, &n_bytes, &error));
+
+ if (error)
+ {
+ Glib::Error::throw_exception(error);
+ }
+
+ utf8_string.assign(buf.get(), buf.get() + n_bytes);
+
+ return is;
+}
+
+std::ostream& operator<<(std::ostream& os, const Glib::ustring& utf8_string)
+{
+ GError* error = 0;
+ const ScopedPtr<char> buf (g_locale_from_utf8(utf8_string.raw().data(),
+ utf8_string.raw().size(), 0, 0, &error));
+ if (error)
+ {
+ Glib::Error::throw_exception(error);
+ }
+
+ // This won't work if the string contains NUL characters. Unfortunately,
+ // std::ostream::write() ignores format flags, so we cannot use that.
+ // The only option would be to create a temporary std::string. However,
+ // even then GCC's libstdc++-v3 prints only the characters up to the first
+ // NUL. Given this, there doesn't seem much of a point in allowing NUL in
+ // formatted output. The semantics would be unclear anyway: what's the
+ // screen width of a NUL?
+ os << buf.get();
+
+ return os;
+}
+
+#ifdef GLIBMM_HAVE_WIDE_STREAM
+
+std::wistream& operator>>(std::wistream& is, ustring& utf8_string)
+{
+ GError* error = 0;
+
+ std::wstring wstr;
+ is >> wstr;
+
+#if defined(__STDC_ISO_10646__) && SIZEOF_WCHAR_T == 4
+ // Avoid going through iconv if wchar_t always contains UCS-4.
+ glong n_bytes = 0;
+ const ScopedPtr<char> buf (g_ucs4_to_utf8(reinterpret_cast<const gunichar*>(wstr.data()),
+ wstr.size(), 0, &n_bytes, &error));
+#elif defined(G_OS_WIN32) && SIZEOF_WCHAR_T == 2
+ // Avoid going through iconv if wchar_t always contains UTF-16.
+ glong n_bytes = 0;
+ const ScopedPtr<char> buf (g_utf16_to_utf8(reinterpret_cast<const gunichar2*>(wstr.data()),
+ wstr.size(), 0, &n_bytes, &error));
+#else
+ gsize n_bytes = 0;
+ const ScopedPtr<char> buf (g_convert(reinterpret_cast<const char*>(wstr.data()),
+ wstr.size() * sizeof(std::wstring::value_type),
+ "UTF-8", "WCHAR_T", 0, &n_bytes, &error));
+#endif // !(__STDC_ISO_10646__ || G_OS_WIN32)
+
+ if (error)
+ {
+ Glib::Error::throw_exception(error);
+ }
+
+ utf8_string.assign(buf.get(), buf.get() + n_bytes);
+
+ return is;
+}
+
+std::wostream& operator<<(std::wostream& os, const ustring& utf8_string)
+{
+ GError* error = 0;
+
+#if defined(__STDC_ISO_10646__) && SIZEOF_WCHAR_T == 4
+ // Avoid going through iconv if wchar_t always contains UCS-4.
+ const ScopedPtr<gunichar> buf (g_utf8_to_ucs4(utf8_string.raw().data(),
+ utf8_string.raw().size(), 0, 0, &error));
+#elif defined(G_OS_WIN32) && SIZEOF_WCHAR_T == 2
+ // Avoid going through iconv if wchar_t always contains UTF-16.
+ const ScopedPtr<gunichar2> buf (g_utf8_to_utf16(utf8_string.raw().data(),
+ utf8_string.raw().size(), 0, 0, &error));
+#else
+ // TODO: For some reason the conversion from UTF-8 to WCHAR_T doesn't work
+ // with g_convert(), while iconv on the command line handles it just fine.
+ // Maybe a bug in GLib?
+ const ScopedPtr<char> buf (g_convert(utf8_string.raw().data(), utf8_string.raw().size(),
+ "WCHAR_T", "UTF-8", 0, 0, &error));
+#endif // !(__STDC_ISO_10646__ || G_OS_WIN32)
+
+ if (error)
+ {
+ Glib::Error::throw_exception(error);
+ }
+
+ // This won't work if the string contains NUL characters. Unfortunately,
+ // std::wostream::write() ignores format flags, so we cannot use that.
+ // The only option would be to create a temporary std::wstring. However,
+ // even then GCC's libstdc++-v3 prints only the characters up to the first
+ // NUL. Given this, there doesn't seem much of a point in allowing NUL in
+ // formatted output. The semantics would be unclear anyway: what's the
+ // screen width of a NUL?
+ os << reinterpret_cast<wchar_t*>(buf.get());
+
+ return os;
+}
+
+#endif /* GLIBMM_HAVE_WIDE_STREAM */
+
+} // namespace Glib
Property changes on: trunk/src/libustring/ustring.cc
___________________________________________________________________
Added: svn:mime-type
+ text/plain
Added: svn:eol-style
+ native
Added: trunk/src/libustring/ustring.h
===================================================================
--- trunk/src/libustring/ustring.h (rev 0)
+++ trunk/src/libustring/ustring.h 2012-03-20 13:35:13 UTC (rev 200)
@@ -0,0 +1,1615 @@
+// -*- c++ -*-
+#ifndef _GLIBMM_USTRING_H
+#define _GLIBMM_USTRING_H
+
+/* $Id$ */
+
+/* Copyright (C) 2002 The gtkmm Development Team
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free
+ * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <glibmmconfig.h>
+#include <glibmm/unicode.h>
+#include <glib.h>
+
+#include <iosfwd>
+#include <iterator>
+#include <sstream>
+#include <string>
+#ifndef GLIBMM_HAVE_STD_ITERATOR_TRAITS
+#include <cstddef> /* for ptrdiff_t */
+#endif
+
+namespace Glib
+{
+
+#ifndef DOXYGEN_SHOULD_SKIP_THIS
+#ifndef GLIBMM_HAVE_STD_ITERATOR_TRAITS
+
+template <class T>
+struct IteratorTraits
+{
+ typedef typename T::iterator_category iterator_category;
+ typedef typename T::value_type value_type;
+ typedef typename T::difference_type difference_type;
+ typedef typename T::pointer pointer;
+ typedef typename T::reference reference;
+};
+
+template <class T>
+struct IteratorTraits<T*>
+{
+ typedef std::random_access_iterator_tag iterator_category;
+ typedef T value_type;
+ typedef ptrdiff_t difference_type;
+ typedef T* pointer;
+ typedef T& reference;
+};
+
+template <class T>
+struct IteratorTraits<const T*>
+{
+ typedef std::random_access_iterator_tag iterator_category;
+ typedef T value_type;
+ typedef ptrdiff_t difference_type;
+ typedef const T* pointer;
+ typedef const T& reference;
+};
+
+#endif /* GLIBMM_HAVE_STD_ITERATOR_TRAITS */
+#endif /* DOXYGEN_SHOULD_SKIP_THIS */
+
+
+/** The iterator type of Glib::ustring.
+ * Note this is not a random access iterator but a bidirectional one,
+ * since all index operations need to iterate over the UTF-8 data. Use
+ * std::advance() to move to a certain position. However, all of the
+ * relational operators are available:
+ * <tt>== != < > <= >=</tt>
+ *
+ * A writeable iterator isn't provided because: The number of bytes of
+ * the old UTF-8 character and the new one to write could be different.
+ * Therefore, any write operation would invalidate all other iterators
+ * pointing into the same string.
+ */
+template <class T>
+class ustring_Iterator
+{
+public:
+ typedef std::bidirectional_iterator_tag iterator_category;
+ typedef gunichar value_type;
+ typedef std::string::difference_type difference_type;
+ typedef value_type reference;
+ typedef void pointer;
+
+ inline ustring_Iterator();
+ inline ustring_Iterator(const ustring_Iterator<std::string::iterator>& other);
+
+ inline value_type operator*() const;
+
+ inline ustring_Iterator<T> & operator++();
+ inline const ustring_Iterator<T> operator++(int);
+ inline ustring_Iterator<T> & operator--();
+ inline const ustring_Iterator<T> operator--(int);
+
+ explicit inline ustring_Iterator(T pos);
+ inline T base() const;
+
+private:
+ T pos_;
+};
+
+
+/** Extract a UCS-4 character from UTF-8 data.
+ * Convert a single UTF-8 (multibyte) character starting at @p pos to
+ * a UCS-4 wide character. This may read up to 6 bytes after the start
+ * position, depending on the UTF-8 character width. You have to make
+ * sure the source contains at least one valid UTF-8 character.
+ *
+ * This is mainly used by the implementation of Glib::ustring::iterator,
+ * but it might be useful as utility function if you prefer using
+ * std::string even for UTF-8 encoding.
+ */
+gunichar get_unichar_from_std_iterator(std::string::const_iterator pos) G_GNUC_PURE;
+
+
+/** Glib::ustring has much the same interface as std::string, but contains
+ * %Unicode characters encoded as UTF-8.
+ *
+ * @par About UTF-8 and ASCII
+ * @par
+ * The standard character set ANSI_X3.4-1968 -- more commonly known as
+ * ASCII -- is a subset of UTF-8. So, if you want to, you can use
+ * Glib::ustring without even thinking about UTF-8.
+ * @par
+ * Whenever ASCII is mentioned in this manual, we mean the @em real ASCII
+ * (i.e. as defined in ANSI_X3.4-1968), which contains only 7-bit characters.
+ * Glib::ustring can @em not be used with ASCII-compatible extended 8-bit
+ * charsets like ISO-8859-1. It's a good idea to avoid string literals
+ * containing non-ASCII characters (e.g. German umlauts) in source code,
+ * or at least you should use UTF-8 literals.
+ * @par
+ * You can find a detailed UTF-8 and %Unicode FAQ here:
+ * http://www.cl.cam.ac.uk/~mgk25/unicode.html
+ *
+ * @par Glib::ustring vs. std::string
+ * @par
+ * Glib::ustring has implicit type conversions to and from std::string.
+ * These conversions do @em not convert to/from the current locale (see
+ * Glib::locale_from_utf8() and Glib::locale_to_utf8() if you need that). You
+ * can always use std::string instead of Glib::ustring -- however, using
+ * std::string with multi-byte characters is quite hard. For instance,
+ * <tt>std::string::operator[]</tt> might return a byte in the middle of a
+ * character, and <tt>std::string::length()</tt> returns the number of bytes
+ * rather than characters. So don't do that without a good reason.
+ * @par
+ * In a perfect world the C++ Standard Library would contain a UTF-8 string
+ * class. Unfortunately, the C++ standard doesn't mention UTF-8 at all. Note
+ * that std::wstring is not a UTF-8 string class because it contains only
+ * fixed-width characters (where width could be 32, 16, or even 8 bits).
+ *
+ * @par Glib::ustring and stream input/output
+ * @par
+ * The stream I/O operators, that is operator<<() and operator>>(), perform
+ * implicit charset conversion to/from the current locale. If that's not
+ * what you intented (e.g. when writing to a configuration file that should
+ * always be UTF-8 encoded) use ustring::raw() to override this behaviour.
+ * @par
+ * If you're using std::ostringstream to build strings for display in the
+ * user interface, you must convert the result back to UTF-8 as shown below:
+ * @code
+ * std::ostringstream output;
+ * output.imbue(std::locale("")); // use the user's locale for this stream
+ * output << percentage << " % done";
+ * label->set_text(Glib::locale_to_utf8(output.str()));
+ * @endcode
+ *
+ * @par Formatted output and internationalization
+ * @par
+ * The methods ustring::compose() and ustring::format() provide a convenient
+ * and powerful alternative to string streams, as shown in the example below.
+ * Refer to the method documentation of compose() and format() for details.
+ * @code
+ * using Glib::ustring;
+ *
+ * ustring message = ustring::compose("%1 is lower than 0x%2.",
+ * 12, ustring::format(std::hex, 16));
+ * @endcode
+ *
+ * @par Implementation notes
+ * @par
+ * Glib::ustring does not inherit from std::string, because std::string was
+ * intended to be a final class. For instance, it does not have a virtual
+ * destructor. Also, a HAS-A relationship is more appropriate because
+ * ustring can't just enhance the std::string interface. Rather, it has to
+ * reimplement the interface so that all operations are based on characters
+ * instead of bytes.
+ */
+class ustring
+{
+public:
+ typedef std::string::size_type size_type;
+ typedef std::string::difference_type difference_type;
+
+ typedef gunichar value_type;
+ typedef gunichar & reference;
+ typedef const gunichar & const_reference;
+
+ typedef ustring_Iterator<std::string::iterator> iterator;
+ typedef ustring_Iterator<std::string::const_iterator> const_iterator;
+
+#ifndef GLIBMM_HAVE_SUN_REVERSE_ITERATOR
+
+ typedef std::reverse_iterator<iterator> reverse_iterator;
+ typedef std::reverse_iterator<const_iterator> const_reverse_iterator;
+
+#else
+
+ typedef std::reverse_iterator<iterator,
+ iterator::iterator_category,
+ iterator::value_type,
+ iterator::reference,
+ iterator::pointer,
+ iterator::difference_type> reverse_iterator;
+ typedef std::reverse_iterator<const_iterator,
+ const_iterator::iterator_category,
+ const_iterator::value_type,
+ const_iterator::reference,
+ const_iterator::pointer,
+ const_iterator::difference_type> const_reverse_iterator;
+
+#endif /* GLIBMM_HAVE_SUN_REVERSE_ITERATOR */
+
+#ifdef GLIBMM_HAVE_ALLOWS_STATIC_INLINE_NPOS
+ static GLIBMM_API const size_type npos = std::string::npos;
+#else
+ //The IRIX MipsPro compiler says "The indicated constant value is not known",
+ //so we need to initalize the static member data elsewhere.
+ static GLIBMM_API const size_type npos;
+#endif
+
+ /*! Default constructor, which creates an empty string.
+ */
+ ustring();
+
+ ~ustring();
+
+ /*! Construct a ustring as a copy of another ustring.
+ * @param other A source string.
+ */
+ ustring(const ustring& other);
+
+ /*! Assign the value of another string to this string.
+ * @param other A source string.
+ */
+ ustring& operator=(const ustring& other);
+
+ /*! Swap contents with another string.
+ * @param other String to swap with.
+ */
+ void swap(ustring& other);
+
+ /*! Construct a ustring as a copy of another std::string.
+ * @param src A source <tt>std::string</tt> containing text encoded as UTF-8.
+ */
+ ustring(const std::string& src);
+
+ /*! Construct a ustring as a copy of a substring.
+ * @param src %Source ustring.
+ * @param i Index of first character to copy from.
+ * @param n Number of UTF-8 characters to copy (defaults to copying the remainder).
+ */
+ ustring(const ustring& src, size_type i, size_type n=npos);
+
+ /*! Construct a ustring as a partial copy of a C string.
+ * @param src %Source C string encoded as UTF-8.
+ * @param n Number of UTF-8 characters to copy.
+ */
+ ustring(const char* src, size_type n);
+
+ /*! Construct a ustring as a copy of a C string.
+ * @param src %Source C string encoded as UTF-8.
+ */
+ ustring(const char* src);
+
+ /*! Construct a ustring as multiple characters.
+ * @param n Number of characters.
+ * @param uc UCS-4 code point to use.
+ */
+ ustring(size_type n, gunichar uc);
+
+ /*! Construct a ustring as multiple characters.
+ * @param n Number of characters.
+ * @param c ASCII character to use.
+ */
+ ustring(size_type n, char c);
+
+ /*! Construct a ustring as a copy of a range.
+ * @param pbegin Start of range.
+ * @param pend End of range.
+ */
+ template <class In> ustring(In pbegin, In pend);
+
+
+//! @name Assign new contents.
+//! @{
+
+ ustring& operator=(const std::string& src);
+ ustring& operator=(const char* src);
+ ustring& operator=(gunichar uc);
+ ustring& operator=(char c);
+
+ ustring& assign(const ustring& src);
+ ustring& assign(const ustring& src, size_type i, size_type n);
+ ustring& assign(const char* src, size_type n);
+ ustring& assign(const char* src);
+ ustring& assign(size_type n, gunichar uc);
+ ustring& assign(size_type n, char c);
+ template <class In> ustring& assign(In pbegin, In pend);
+
+//! @}
+//! @name Append to the string.
+//! @{
+
+ ustring& operator+=(const ustring& src);
+ ustring& operator+=(const char* src);
+ ustring& operator+=(gunichar uc);
+ ustring& operator+=(char c);
+ void push_back(gunichar uc);
+ void push_back(char c);
+
+ ustring& append(const ustring& src);
+ ustring& append(const ustring& src, size_type i, size_type n);
+ ustring& append(const char* src, size_type n);
+ ustring& append(const char* src);
+ ustring& append(size_type n, gunichar uc);
+ ustring& append(size_type n, char c);
+ template <class In> ustring& append(In pbegin, In pend);
+
+//! @}
+//! @name Insert into the string.
+//! @{
+
+ ustring& insert(size_type i, const ustring& src);
+ ustring& insert(size_type i, const ustring& src, size_type i2, size_type n);
+ ustring& insert(size_type i, const char* src, size_type n);
+ ustring& insert(size_type i, const char* src);
+ ustring& insert(size_type i, size_type n, gunichar uc);
+ ustring& insert(size_type i, size_type n, char c);
+
+ iterator insert(iterator p, gunichar uc);
+ iterator insert(iterator p, char c);
+ void insert(iterator p, size_type n, gunichar uc);
+ void insert(iterator p, size_type n, char c);
+ template <class In> void insert(iterator p, In pbegin, In pend);
+
+//! @}
+//! @name Replace sub-strings.
+//! @{
+
+ ustring& replace(size_type i, size_type n, const ustring& src);
+ ustring& replace(size_type i, size_type n, const ustring& src, size_type i2, size_type n2);
+ ustring& replace(size_type i, size_type n, const char* src, size_type n2);
+ ustring& replace(size_type i, size_type n, const char* src);
+ ustring& replace(size_type i, size_type n, size_type n2, gunichar uc);
+ ustring& replace(size_type i, size_type n, size_type n2, char c);
+
+ ustring& replace(iterator pbegin, iterator pend, const ustring& src);
+ ustring& replace(iterator pbegin, iterator pend, const char* src, size_type n);
+ ustring& replace(iterator pbegin, iterator pend, const char* src);
+ ustring& replace(iterator pbegin, iterator pend, size_type n, gunichar uc);
+ ustring& replace(iterator pbegin, iterator pend, size_type n, char c);
+ template <class In> ustring& replace(iterator pbegin, iterator pend, In pbegin2, In pend2);
+
+//! @}
+//! @name Erase sub-strings.
+//! @{
+
+ void clear();
+ ustring& erase(size_type i, size_type n=npos);
+ ustring& erase();
+ iterator erase(iterator p);
+ iterator erase(iterator pbegin, iterator pend);
+
+//! @}
+//! @name Compare and collate.
+//! @{
+
+ int compare(const ustring& rhs) const;
+ int compare(const char* rhs) const;
+ int compare(size_type i, size_type n, const ustring& rhs) const;
+ int compare(size_type i, size_type n, const ustring& rhs, size_type i2, size_type n2) const;
+ int compare(size_type i, size_type n, const char* rhs, size_type n2) const;
+ int compare(size_type i, size_type n, const char* rhs) const;
+
+ /*! Create a unique sorting key for the UTF-8 string. If you need to
+ * compare UTF-8 strings regularly, e.g. for sorted containers such as
+ * <tt>std::set<></tt>, you should consider creating a collate key first
+ * and compare this key instead of the actual string.
+ *
+ * The ustring::compare() methods as well as the relational operators
+ * <tt>== != < > <= >=</tt> are quite costly
+ * because they have to deal with %Unicode and the collation rules defined by
+ * the current locale. Converting both operands to UCS-4 is just the first
+ * of several costly steps involved when comparing ustrings. So be careful.
+ */
+ std::string collate_key() const;
+
+ /*! Create a unique key for the UTF-8 string that can be used for caseless
+ * sorting. <tt>ustr.casefold_collate_key()</tt> results in the same string
+ * as <tt>ustr.casefold().collate_key()</tt>, but the former is likely more
+ * efficient.
+ */
+ std::string casefold_collate_key() const;
+
+//! @}
+//! @name Extract characters and sub-strings.
+//! @{
+
+ /*! No reference return; use replace() to write characters. */
+ value_type operator[](size_type i) const;
+
+ /*! No reference return; use replace() to write characters. @throw std::out_of_range */
+ value_type at(size_type i) const;
+
+ inline ustring substr(size_type i=0, size_type n=npos) const;
+
+//! @}
+//! @name Access a sequence of characters.
+//! @{
+
+ iterator begin();
+ iterator end();
+ const_iterator begin() const;
+ const_iterator end() const;
+ reverse_iterator rbegin();
+ reverse_iterator rend();
+ const_reverse_iterator rbegin() const;
+ const_reverse_iterator rend() const;
+
+//! @}
+//! @name Find sub-strings.
+//! @{
+
+ size_type find(const ustring& str, size_type i=0) const;
+ size_type find(const char* str, size_type i, size_type n) const;
+ size_type find(const char* str, size_type i=0) const;
+ size_type find(gunichar uc, size_type i=0) const;
+ size_type find(char c, size_type i=0) const;
+
+ size_type rfind(const ustring& str, size_type i=npos) const;
+ size_type rfind(const char* str, size_type i, size_type n) const;
+ size_type rfind(const char* str, size_type i=npos) const;
+ size_type rfind(gunichar uc, size_type i=npos) const;
+ size_type rfind(char c, size_type i=npos) const;
+
+//! @}
+//! @name Match against a set of characters.
+//! @{
+
+ size_type find_first_of(const ustring& match, size_type i=0) const;
+ size_type find_first_of(const char* match, size_type i, size_type n) const;
+ size_type find_first_of(const char* match, size_type i=0) const;
+ size_type find_first_of(gunichar uc, size_type i=0) const;
+ size_type find_first_of(char c, size_type i=0) const;
+
+ size_type find_last_of(const ustri...
[truncated message content] |
