Dogcows Code - chaz/openbox/blob - otk/ustring.cc

   1 // -*- mode: C++; indent-tabs-mode: nil; c-basic-offset: 2; -*-
   2
   3 #ifdef    HAVE_CONFIG_H
   4 #  include "../config.h"
   5 #endif // HAVE_CONFIG_H
   6
   7 #include "ustring.hh"
   8
   9 extern "C" {
  10 #include <assert.h>
  11 }
  12
  13 namespace otk {
  14
  15 // helper functions
  16
  17 // takes a pointer into a utf8 string and returns a unicode character for the
  18 // first character at the pointer
  19 unichar utf8_get_char (const char *p)
  20 {
  21   unichar result = static_cast<unsigned char>(*p);
  22
  23   // if its not a 7-bit ascii character
  24   if((result & 0x80) != 0) {
  25     // len is the number of bytes this character takes up in the string
  26     unsigned char len = utf8_skip[result];
  27     result &= 0x7F >> len;
  28
  29     while(--len != 0) {
  30       result <<= 6;
  31       result |= static_cast<unsigned char>(*++p) & 0x3F;
  32     }
  33   }
  34
  35   return result;
  36 }
  37
  38 // takes a pointer into a string and finds its offset
  39 static ustring::size_type utf8_ptr_to_offset(const char *str, const char *pos)
  40 {
  41   ustring::size_type offset = 0;
  42
  43   while (str < pos) {
  44     str += utf8_skip[*str];
  45     offset++;
  46   }
  47
  48   return offset;
  49 }
  50
  51 // takes an offset into a string and returns a pointer to it
  52 const char *utf8_offset_to_ptr(const char *str, ustring::size_type offset)
  53 {
  54   while (offset--)
  55     str += utf8_skip[*str];
  56   return str;
  57 }
  58
  59 // First overload: stop on '\0' character.
  60 ustring::size_type utf8_byte_offset(const char* str, ustring::size_type offset)
  61 {
  62   if(offset == ustring::npos)
  63     return ustring::npos;
  64
  65   const char* p = str;
  66
  67   for(; offset != 0; --offset)
  68   {
  69     if(*p == '\0')
  70       return ustring::npos;
  71
  72     p += utf8_skip[*p];
  73   }
  74
  75   return (p - str);
  76 }
  77
  78 // Second overload: stop when reaching maxlen.
  79 ustring::size_type utf8_byte_offset(const char* str, ustring::size_type offset,
  80                                     ustring::size_type maxlen)
  81 {
  82   if(offset == ustring::npos)
  83     return ustring::npos;
  84
  85   const char *const pend = str + maxlen;
  86   const char* p = str;
  87
  88   for(; offset != 0; --offset)
  89   {
  90     if(p >= pend)
  91       return ustring::npos;
  92
  93     p += utf8_skip[*p];
  94   }
  95
  96   return (p - str);
  97 }
  98
  99
 100 // ustring methods
 101
 102 ustring::ustring()
 103 {
 104 }
 105
 106 ustring::~ustring()
 107 {
 108 }
 109
 110 ustring::ustring(const ustring& other)
 111   : _string(other._string), _utf8(other._utf8)
 112 {
 113 }
 114
 115 ustring& ustring::operator=(const ustring& other)
 116 {
 117   _string = other._string;
 118   _utf8 = other._utf8;
 119   return *this;
 120 }
 121
 122 ustring::ustring(const std::string& src)
 123   : _string(src), _utf8(true)
 124 {
 125 }
 126
 127 ustring::ustring(const char* src)
 128   : _string(src), _utf8(true)
 129 {
 130 }
 131
 132 ustring& ustring::operator+=(const ustring& src)
 133 {
 134   assert(_utf8 == src._utf8);
 135   _string += src._string;
 136   return *this;
 137 }
 138
 139 ustring& ustring::operator+=(const char* src)
 140 {
 141   _string += src;
 142   return *this;
 143 }
 144
 145 ustring& ustring::operator+=(char c)
 146 {
 147   _string += c;
 148   return *this;
 149 }
 150
 151 ustring::size_type ustring::size() const
 152 {
 153   if (_utf8) {
 154     const char *const pdata = _string.data();
 155     return utf8_ptr_to_offset(pdata, pdata + _string.size());
 156   } else
 157     return _string.size();
 158 }
 159
 160 ustring::size_type ustring::bytes() const
 161 {
 162   return _string.size();
 163 }
 164
 165 ustring::size_type ustring::capacity() const
 166 {
 167   return _string.capacity();
 168 }
 169
 170 ustring::size_type ustring::max_size() const
 171 {
 172   return _string.max_size();
 173 }
 174
 175 bool ustring::empty() const
 176 {
 177   return _string.empty();
 178 }
 179
 180 void ustring::clear()
 181 {
 182   _string.erase();
 183 }
 184
 185 ustring& ustring::erase(ustring::size_type i, ustring::size_type n)
 186 {
 187   if (_utf8) {
 188     // find a proper offset
 189     size_type utf_i = utf8_byte_offset(_string.c_str(), i);
 190     if (utf_i != npos) {
 191       // if the offset is not npos, find a proper length for 'n'
 192       size_type utf_n = utf8_byte_offset(_string.data() + utf_i, n,
 193                                          _string.size() - utf_i);
 194       _string.erase(utf_i, utf_n);
 195     }
 196   } else
 197     _string.erase(i, n);
 198
 199   return *this;
 200 }
 201
 202 void ustring::resize(ustring::size_type n, char c)
 203 {
 204   if (_utf8) {
 205     const size_type size_now = size();
 206     if(n < size_now)
 207       erase(n, npos);
 208     else if(n > size_now)
 209       _string.append(n - size_now, c);
 210   } else
 211     _string.resize(n, c);
 212 }
 213
 214 ustring::value_type ustring::operator[](ustring::size_type i) const
 215 {
 216   return utf8_get_char(utf8_offset_to_ptr(_string.data(), i));
 217 }
 218
 219 const char* ustring::data() const
 220 {
 221   return _string.data();
 222 }
 223
 224 const char* ustring::c_str() const
 225 {
 226   return _string.c_str();
 227 }
 228
 229 bool ustring::utf8() const
 230 {
 231   return _utf8;
 232 }
 233
 234 void ustring::setUtf8(bool utf8)
 235 {
 236   _utf8 = utf8;
 237 }
 238
 239 }