[100] | 1 | /**********************************************************************
|
---|
[94] | 2 | *
|
---|
| 3 | * text_t.h -- a simple 16-bit charater string class
|
---|
[100] | 4 | * Copyright (C) 1999 The New Zealand Digital Library Project
|
---|
[94] | 5 | *
|
---|
[100] | 6 | * PUT COPYRIGHT NOTICE HERE
|
---|
[94] | 7 | *
|
---|
| 8 | * $Id: text_t.h 114 1999-01-19 01:38:20Z rjmcnab $
|
---|
| 9 | *
|
---|
[100] | 10 | *********************************************************************/
|
---|
[94] | 11 |
|
---|
| 12 |
|
---|
| 13 | #ifndef TEXT_T_H
|
---|
| 14 | #define TEXT_T_H
|
---|
| 15 |
|
---|
[114] | 16 | #include "gsdlconf.h"
|
---|
| 17 |
|
---|
| 18 | #if defined(GSDL_USE_OBJECTSPACE)
|
---|
| 19 | # include <ospace\std\vector>
|
---|
| 20 | # include <ospace\std\list>
|
---|
| 21 | # include <ospace\std\set>
|
---|
| 22 | #elif defined(GSDL_USE_STL_H)
|
---|
| 23 | # include <vector.h>
|
---|
| 24 | # include <list.h>
|
---|
| 25 | # include <set.h>
|
---|
| 26 | #else
|
---|
[94] | 27 | # include <vector>
|
---|
| 28 | # include <list>
|
---|
| 29 | # include <set>
|
---|
| 30 | #endif
|
---|
| 31 |
|
---|
| 32 | // use the standard namespace
|
---|
[114] | 33 | #if defined(GSDL_USE_OBJECTSPACE)
|
---|
| 34 | using namespace ospace::std;
|
---|
| 35 | #elif !defined(GSDL_NAMESPACE_BROKEN)
|
---|
[94] | 36 | using namespace std;
|
---|
| 37 | #endif
|
---|
| 38 |
|
---|
| 39 |
|
---|
| 40 | // class prototypes
|
---|
| 41 | class inconvertclass;
|
---|
| 42 | class outconvertclass;
|
---|
| 43 |
|
---|
| 44 |
|
---|
[114] | 45 | // just in case the compiler requires it :-)
|
---|
| 46 | inline void destroy(unsigned short *) {};
|
---|
| 47 |
|
---|
[94] | 48 | typedef vector<unsigned short> usvector;
|
---|
| 49 |
|
---|
| 50 | // The class text_t can handle long strings which may contain
|
---|
| 51 | // null characters. It uses unsigned shorts to represent up to
|
---|
| 52 | // 64K character values.
|
---|
| 53 | class text_t {
|
---|
| 54 | public:
|
---|
| 55 | //type support for ucvector
|
---|
| 56 | typedef usvector::iterator iterator;
|
---|
| 57 | typedef usvector::const_iterator const_iterator;
|
---|
| 58 | typedef usvector::reference reference;
|
---|
| 59 | typedef usvector::const_reference const_reference;
|
---|
| 60 | typedef usvector::size_type size_type;
|
---|
| 61 | typedef usvector::difference_type difference_type;
|
---|
| 62 | typedef usvector::const_reverse_iterator const_reverse_iterator;
|
---|
| 63 | typedef usvector::reverse_iterator reverse_iterator;
|
---|
| 64 |
|
---|
| 65 | protected:
|
---|
| 66 | usvector text;
|
---|
| 67 | unsigned short encoding; // 0 = unicode, 1 = other
|
---|
| 68 |
|
---|
| 69 | public:
|
---|
| 70 | // constructors
|
---|
| 71 | text_t ();
|
---|
| 72 | text_t (int i);
|
---|
| 73 | text_t (char *s); // assumed to be a normal c string
|
---|
| 74 |
|
---|
| 75 | void setencoding (unsigned short theencoding) {encoding=theencoding;};
|
---|
| 76 | unsigned short getencoding () {return encoding;};
|
---|
| 77 |
|
---|
| 78 | // basic container support
|
---|
| 79 | iterator begin () {return text.begin();}
|
---|
| 80 | const_iterator begin () const {return text.begin();}
|
---|
| 81 | iterator end () {return text.end();}
|
---|
| 82 | const_iterator end () const {return text.end();}
|
---|
| 83 |
|
---|
| 84 | void erase(iterator pos) {text.erase(pos);}
|
---|
| 85 | void erase(iterator first, iterator last) {text.erase(first, last);}
|
---|
| 86 | void push_back(unsigned short c) {text.push_back(c);}
|
---|
| 87 | void pop_back() {text.pop_back();}
|
---|
| 88 | text_t &operator=(const text_t &x) {text=x.text; encoding=x.encoding; return *this;}
|
---|
| 89 | reference operator[](size_type n) {return text[n];};
|
---|
| 90 | const_reference operator[](size_type n) const {return text[n];};
|
---|
| 91 |
|
---|
| 92 | bool empty () const {return text.empty();}
|
---|
| 93 | size_type size() const {return text.size();}
|
---|
| 94 | friend bool operator!=(const text_t& x, const text_t& y);
|
---|
| 95 | friend bool operator==(const text_t& x, const text_t& y);
|
---|
| 96 | friend bool operator<(const text_t& x, const text_t& y);
|
---|
| 97 | friend bool operator>(const text_t& x, const text_t& y);
|
---|
| 98 |
|
---|
| 99 | // added functionality
|
---|
| 100 | void clear () {text.erase(text.begin(),text.end());}
|
---|
| 101 | void append (const text_t &t);
|
---|
| 102 | void appendrange (iterator first, iterator last);
|
---|
| 103 | void appendrange (const_iterator first, const_iterator last);
|
---|
| 104 | text_t &operator+= (const text_t &t) {append(t);return *this;}
|
---|
| 105 |
|
---|
| 106 | // support for integers
|
---|
| 107 | void appendint (int i);
|
---|
| 108 | void setint (int i) {clear();appendint(i);}
|
---|
| 109 | text_t &operator=(int i) {setint (i);return *this;}
|
---|
| 110 | text_t &operator+= (int i) {appendint(i);return *this;}
|
---|
| 111 | int getint ();
|
---|
| 112 |
|
---|
| 113 | // support for arrays of chars
|
---|
| 114 | void appendcarr (char *s, size_type len);
|
---|
| 115 | void setcarr (char *s, size_type len) {clear();appendcarr(s,len);}
|
---|
| 116 |
|
---|
| 117 | // support for null-terminated C strings
|
---|
| 118 | void appendcstr (char *s);
|
---|
| 119 | void setcstr (char *s) {clear();appendcstr(s);}
|
---|
| 120 | text_t &operator= (char *s) {setcstr(s);return *this;} // c string
|
---|
| 121 | text_t &operator+= (char *s) {appendcstr(s);return *this;} // c string
|
---|
| 122 |
|
---|
| 123 | // strings returned from getcarr and getcstr become the callers
|
---|
| 124 | // responsibility and should be deallocated with "delete"
|
---|
| 125 | char *getcarr(size_type &len) const;
|
---|
| 126 | char *getcstr() const;
|
---|
| 127 | };
|
---|
| 128 |
|
---|
| 129 |
|
---|
| 130 | inline text_t operator+(const text_t &t1, const text_t &t2)
|
---|
| 131 | {
|
---|
| 132 | text_t tnew = t1;
|
---|
| 133 | tnew.append(t2);
|
---|
| 134 | return tnew;
|
---|
| 135 | }
|
---|
| 136 |
|
---|
| 137 | inline text_t operator+(const text_t &t1, int i1)
|
---|
| 138 | {
|
---|
| 139 | text_t tnew = t1;
|
---|
| 140 | tnew.appendint(i1);
|
---|
| 141 | return tnew;
|
---|
| 142 | }
|
---|
| 143 |
|
---|
| 144 | inline text_t operator+(const text_t &t1, char *s1)
|
---|
| 145 | {
|
---|
| 146 | text_t tnew = t1;
|
---|
| 147 | tnew.appendcstr(s1);
|
---|
| 148 | return tnew;
|
---|
| 149 | }
|
---|
| 150 |
|
---|
| 151 |
|
---|
| 152 | inline bool operator!=(const text_t& x, const text_t& y)
|
---|
| 153 | {
|
---|
| 154 | return (x.text != y.text);
|
---|
| 155 | }
|
---|
| 156 |
|
---|
| 157 | inline bool operator==(const text_t& x, const text_t& y)
|
---|
| 158 | {
|
---|
| 159 | return (x.text == y.text);
|
---|
| 160 | }
|
---|
| 161 |
|
---|
| 162 | inline bool operator<(const text_t& x, const text_t& y)
|
---|
| 163 | {
|
---|
| 164 | return (x.text < y.text);
|
---|
| 165 | }
|
---|
| 166 |
|
---|
| 167 | inline bool operator>(const text_t& x, const text_t& y)
|
---|
| 168 | {
|
---|
| 169 | return (x.text > y.text);
|
---|
| 170 | }
|
---|
| 171 |
|
---|
| 172 |
|
---|
| 173 | struct eqtext_t
|
---|
| 174 | {
|
---|
| 175 | bool operator()(const text_t t1, const text_t t2) const
|
---|
| 176 | { return t1 == t2; }
|
---|
| 177 | };
|
---|
| 178 |
|
---|
| 179 | struct lttext_t
|
---|
| 180 | {
|
---|
| 181 | bool operator()(const text_t t1, const text_t t2) const
|
---|
| 182 | { return t1 < t2; }
|
---|
| 183 | };
|
---|
| 184 |
|
---|
| 185 |
|
---|
| 186 | // frequently used derived types
|
---|
| 187 | typedef set<text_t,lttext_t> text_tset;
|
---|
| 188 | typedef list<text_t> text_tlist; // more efficient for insertions/deletions
|
---|
| 189 | typedef vector<text_t> text_tarray; // more space efficient than text_tlist
|
---|
| 190 |
|
---|
| 191 |
|
---|
| 192 | // general functions which work on text_ts
|
---|
| 193 |
|
---|
| 194 | // find a character within a range
|
---|
| 195 | text_t::const_iterator findchar (text_t::const_iterator first, text_t::const_iterator last,
|
---|
| 196 | unsigned short c);
|
---|
| 197 | text_t::iterator findchar (text_t::iterator first, text_t::iterator last,
|
---|
| 198 | unsigned short c);
|
---|
| 199 |
|
---|
| 200 | // get a string up to the next delimiter (which is skipped)
|
---|
| 201 | text_t::const_iterator getdelimitstr (text_t::const_iterator first,
|
---|
| 202 | text_t::const_iterator last,
|
---|
| 203 | unsigned short c, text_t &outstr);
|
---|
| 204 | text_t::iterator getdelimitstr (text_t::iterator first, text_t::iterator last,
|
---|
| 205 | unsigned short c, text_t &outstr);
|
---|
| 206 |
|
---|
| 207 | // split a string with a character
|
---|
| 208 | void splitchar (text_t::const_iterator first, text_t::const_iterator last,
|
---|
| 209 | unsigned short c, text_tlist &outlist);
|
---|
| 210 | void splitchar (text_t::const_iterator first, text_t::const_iterator last,
|
---|
| 211 | unsigned short c, text_tarray &outlist);
|
---|
| 212 |
|
---|
| 213 | // join a string using a character
|
---|
| 214 | void joinchar (const text_tlist &inlist, unsigned short c, text_t &outtext);
|
---|
| 215 | void joinchar (const text_tarray &inlist, unsigned short c, text_t &outtext);
|
---|
| 216 |
|
---|
| 217 | // count the occurances of a character within a range
|
---|
| 218 | int countchar (text_t::const_iterator first, text_t::const_iterator last,
|
---|
| 219 | unsigned short c);
|
---|
| 220 |
|
---|
| 221 |
|
---|
| 222 |
|
---|
| 223 |
|
---|
| 224 |
|
---|
| 225 |
|
---|
| 226 |
|
---|
| 227 |
|
---|
| 228 | // conversion classes used for getting information in to and out of
|
---|
| 229 | // the text_t class.
|
---|
| 230 |
|
---|
| 231 | class convertclass
|
---|
| 232 | {
|
---|
| 233 | public:
|
---|
| 234 | enum status_t {finished, stopped, unfinished};
|
---|
| 235 |
|
---|
| 236 | convertclass ();
|
---|
| 237 | virtual void reset ();
|
---|
| 238 | };
|
---|
| 239 |
|
---|
| 240 |
|
---|
| 241 |
|
---|
| 242 | // convert from a char stream to the text_t class
|
---|
| 243 | // the default version assumes the input is a ascii
|
---|
| 244 | // character array
|
---|
| 245 | class inconvertclass : public convertclass
|
---|
| 246 | {
|
---|
| 247 | public:
|
---|
| 248 | inconvertclass ();
|
---|
| 249 | void reset ();
|
---|
| 250 | void setinput (char *thestart, size_t thelen);
|
---|
| 251 |
|
---|
| 252 | // output will be cleared before the conversion
|
---|
| 253 | virtual void convert (text_t &output, status_t &status);
|
---|
| 254 |
|
---|
| 255 | // will treat the text_t as a 8-bit string and convert
|
---|
| 256 | // it to a 16-bit string using the about convert method.
|
---|
| 257 | text_t convert (const text_t &t);
|
---|
| 258 |
|
---|
| 259 | protected:
|
---|
| 260 | char *start;
|
---|
| 261 | size_t len;
|
---|
| 262 | };
|
---|
| 263 |
|
---|
| 264 | // to get something which will do the conversion
|
---|
| 265 | // to ascii declare a (non global!) instance like
|
---|
| 266 | // this
|
---|
| 267 | // inconvertclass ascii2text_t;
|
---|
| 268 |
|
---|
| 269 |
|
---|
| 270 | // Convert from a text_t class to a char stream
|
---|
| 271 | // This default version assumes the output is a ascii
|
---|
| 272 | // character array. If you set the output stream you
|
---|
| 273 | // can use this class to output to a stream using the
|
---|
| 274 | // << operator. The << operator can also be conveniently
|
---|
| 275 | // used to set the output stream by doing something like
|
---|
| 276 | //
|
---|
| 277 | // cout << text_t2ascii << textstr << anothertextstr;
|
---|
| 278 | //
|
---|
| 279 | // this class assumes that the input text doesn't change
|
---|
| 280 | // while the conversion takes place
|
---|
| 281 | class outconvertclass : public convertclass
|
---|
| 282 | {
|
---|
| 283 | public:
|
---|
| 284 | outconvertclass ();
|
---|
| 285 | void reset ();
|
---|
| 286 | void setinput (text_t *theinput);
|
---|
| 287 | // note that convert does not null-terminate the
|
---|
| 288 | // output array of characters
|
---|
| 289 | virtual void convert (char *output, size_t maxlen,
|
---|
| 290 | size_t &len, status_t &status);
|
---|
| 291 |
|
---|
| 292 | // will convert the 16-bit string to a 8-bit stream
|
---|
| 293 | // and place the result in a text_t. This method uses
|
---|
| 294 | // the above convert function.
|
---|
| 295 | text_t convert (const text_t &t);
|
---|
| 296 |
|
---|
| 297 | void setostream (ostream *theouts);
|
---|
| 298 | ostream *getostream ();
|
---|
| 299 |
|
---|
| 300 | protected:
|
---|
| 301 | text_t *input;
|
---|
| 302 | text_t::iterator texthere; // only valid if input is valid
|
---|
| 303 |
|
---|
| 304 | ostream *outs;
|
---|
| 305 | };
|
---|
| 306 |
|
---|
| 307 | // to get something which will do the conversion
|
---|
| 308 | // to text_t declare a (non global!) instance like
|
---|
| 309 | // this
|
---|
| 310 | // outconvertclass text_t2ascii;
|
---|
| 311 |
|
---|
| 312 |
|
---|
| 313 | // stream operators for the output class
|
---|
| 314 | outconvertclass &operator<< (ostream &theouts, outconvertclass &outconverter);
|
---|
| 315 | outconvertclass &operator<< (outconvertclass &outconverter, const text_t &t);
|
---|
| 316 |
|
---|
| 317 | #endif
|
---|