GNU CommonC++
tokenizer.h
Go to the documentation of this file.
1// Copyright (C) 1999-2005 Open Source Telecom Corporation.
2// Copyright (C) 2006-2010 David Sugar, Tycho Softworks.
3//
4// This program is free software; you can redistribute it and/or modify
5// it under the terms of the GNU General Public License as published by
6// the Free Software Foundation; either version 2 of the License, or
7// (at your option) any later version.
8//
9// This program is distributed in the hope that it will be useful,
10// but WITHOUT ANY WARRANTY; without even the implied warranty of
11// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12// GNU General Public License for more details.
13//
14// You should have received a copy of the GNU General Public License
15// along with this program; if not, write to the Free Software
16// Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
17//
18// As a special exception, you may use this file as part of a free software
19// library without restriction. Specifically, if other files instantiate
20// templates or use macros or inline functions from this file, or you compile
21// this file and link it with other files to produce an executable, this
22// file does not by itself cause the resulting executable to be covered by
23// the GNU General Public License. This exception does not however
24// invalidate any other reasons why the executable file might be covered by
25// the GNU General Public License.
26//
27// This exception applies only to the code released under the name GNU
28// Common C++. If you copy code from other releases into a copy of GNU
29// Common C++, as the General Public License permits, the exception does
30// not apply to the code that you add in this way. To avoid misleading
31// anyone as to the status of such modified files, you must delete
32// this exception notice from them.
33//
34// If you write modifications of your own for GNU Common C++, it is your choice
35// whether to permit this exception to apply to your modifications.
36// If you do not wish that, delete this exception notice.
37//
38
44#ifndef CCXX_TOKENIZER_H_
45#define CCXX_TOKENIZER_H_
46
47#ifndef CCXX_MISSING_H_
48#include <cc++/missing.h>
49#endif
50
51#ifndef CCXX_THREAD_H_
52#include <cc++/thread.h>
53#endif
54
55#ifdef CCXX_NAMESPACES
56namespace ost {
57#endif
58
103public:
109 static const char * const SPACE;
110
120 // maybe move more global ?
122
128 friend class StringTokenizer; // access our private constructors
129 private:
130 const StringTokenizer *myTok; // my StringTokenizer
131 const char *start; // start of current token
132 const char *tokEnd; // end of current token (->nxDelimiter)
133 const char *endp; // one before next token
134 char *token; // allocated token, if requested
135
136 // for initialization of the itEnd iterator
137 iterator(const StringTokenizer &tok, const char *end)
138 : myTok(&tok),tokEnd(0),endp(end),token(0) {}
139
140 iterator(const StringTokenizer &tok)
141 : myTok(&tok),tokEnd(0),endp(myTok->str-1),token(0) {
142 ++(*this); // init first token.
143 }
144
145 public:
146 iterator() : myTok(0),start(0),tokEnd(0),endp(0),token(0) {}
147
148 // see also: comment in implementation of operator++
149 virtual ~iterator()
150 { if (token) *token='\0'; delete [] token; }
151
155 // everything, but not responsible for the allocated token.
156 iterator(const iterator& i) :
157 myTok(i.myTok),start(i.start),tokEnd(i.tokEnd),
158 endp(i.endp),token(0) {}
159
163 // everything, but not responsible for the allocated token.
164 iterator &operator = (const iterator &i)
165 {
166 myTok = i.myTok;
167 start = i.start; endp = i.endp; tokEnd = i.tokEnd;
168 if ( token )
169 delete [] token;
170 token = 0;
171 return *this;
172 }
173
178
187 const char* operator * () THROWS (NoSuchElementException);
188
195 inline char nextDelimiter() const
196 {return (tokEnd) ? *tokEnd : '\0';}
197
202 // only compare the end-position. speed.
203 inline bool operator == (const iterator &other) const
204 {return (endp == other.endp);}
205
210 // only compare the end position. speed.
211 inline bool operator != (const iterator &other) const
212 {return (endp != other.endp);}
213 };
214private:
216 const char *str;
217 const char *delim;
218 bool skipAll, trim;
219 iterator itEnd;
220
221public:
260 StringTokenizer (const char *str,
261 const char *delim,
262 bool skipAllDelim = false,
263 bool trim = false);
264
274 StringTokenizer (const char *s);
275
280 {return iterator(*this);}
281
286 void setDelimiters (const char *d)
287 {delim = d;}
288
293 iterator begin(const char *d)
294 {
295 delim = d;
296 return iterator(*this);
297 }
298
302 const iterator& end() const
303 {return itEnd;}
304};
305
306#ifdef CCXX_NAMESPACES
307}
308#endif
309
310#endif
311
Exception thrown, if someone tried to read beyond the end of the tokens.
Definition tokenizer.h:121
The input forward iterator for tokens.
Definition tokenizer.h:127
iterator(const iterator &i)
copy constructor.
Definition tokenizer.h:156
virtual ~iterator()
Definition tokenizer.h:149
iterator()
Definition tokenizer.h:146
Splits delimited string into tokens.
Definition tokenizer.h:102
void setDelimiters(const char *d)
changes the set of delimiters used in subsequent iterations.
Definition tokenizer.h:286
static const char *const SPACE
a delimiter string containing all usual whitespace delimiters.
Definition tokenizer.h:109
iterator begin() const
returns the begin iterator
Definition tokenizer.h:279
StringTokenizer(const char *str, const char *delim, bool skipAllDelim=false, bool trim=false)
creates a new StringTokenizer for a string and a given set of delimiters.
const iterator & end() const
the iterator marking the end.
Definition tokenizer.h:302
StringTokenizer(const char *s)
create a new StringTokenizer which splits the input string at whitespaces.
iterator begin(const char *d)
returns a begin iterator with an alternate set of delimiters.
Definition tokenizer.h:293
#define THROWS(x)
Definition config.h:1010
#define __EXPORT
Definition config.h:1045
substitute functions which may be missing in target platform libc.
Definition address.h:64
Synchronization and threading services.