]> git.saurik.com Git - wxWidgets.git/blame - src/common/tokenzr.cpp
Disable handling of wxEVT_MOUSEWHEEL in wxVarScrollHelperEvtHandler in wxGTK.
[wxWidgets.git] / src / common / tokenzr.cpp
CommitLineData
f4ada568 1/////////////////////////////////////////////////////////////////////////////
aaa6d89a 2// Name: src/common/tokenzr.cpp
f4ada568
GL
3// Purpose: String tokenizer
4// Author: Guilhem Lavaux
1e6feb95 5// Modified by: Vadim Zeitlin (almost full rewrite)
f4ada568 6// Created: 04/22/98
f4ada568 7// Copyright: (c) Guilhem Lavaux
65571936 8// Licence: wxWindows licence
f4ada568
GL
9/////////////////////////////////////////////////////////////////////////////
10
bbf8fc53
VZ
11// ============================================================================
12// declarations
13// ============================================================================
14
15// ----------------------------------------------------------------------------
16// headers
17// ----------------------------------------------------------------------------
18
fcc6dddd
JS
19// For compilers that support precompilation, includes "wx.h".
20#include "wx/wxprec.h"
21
22#ifdef __BORLANDC__
85833f5c 23 #pragma hdrstop
fcc6dddd
JS
24#endif
25
f4ada568 26#include "wx/tokenzr.h"
aaa6d89a
WS
27
28#ifndef WX_PRECOMP
29 #include "wx/arrstr.h"
0bf751e7 30 #include "wx/crt.h"
aaa6d89a 31#endif
f4ada568 32
3f8e5072
JS
33// Required for wxIs... functions
34#include <ctype.h>
35
bbf8fc53
VZ
36// ============================================================================
37// implementation
38// ============================================================================
39
f0dfc29c
VS
40// ----------------------------------------------------------------------------
41// helpers
42// ----------------------------------------------------------------------------
43
44static wxString::const_iterator
45find_first_of(const wxChar *delims, size_t len,
46 const wxString::const_iterator& from,
47 const wxString::const_iterator& end)
48{
9a83f860 49 wxASSERT_MSG( from <= end, wxT("invalid index") );
f0dfc29c
VS
50
51 for ( wxString::const_iterator i = from; i != end; ++i )
52 {
53 if ( wxTmemchr(delims, *i, len) )
54 return i;
55 }
56
57 return end;
58}
59
60static wxString::const_iterator
61find_first_not_of(const wxChar *delims, size_t len,
62 const wxString::const_iterator& from,
63 const wxString::const_iterator& end)
64{
9a83f860 65 wxASSERT_MSG( from <= end, wxT("invalid index") );
f0dfc29c
VS
66
67 for ( wxString::const_iterator i = from; i != end; ++i )
68 {
69 if ( !wxTmemchr(delims, *i, len) )
70 return i;
71 }
72
73 return end;
74}
75
bbf8fc53
VZ
76// ----------------------------------------------------------------------------
77// wxStringTokenizer construction
78// ----------------------------------------------------------------------------
79
7c968cee 80wxStringTokenizer::wxStringTokenizer(const wxString& str,
f4ada568 81 const wxString& delims,
7c968cee 82 wxStringTokenizerMode mode)
bbf8fc53 83{
7c968cee 84 SetString(str, delims, mode);
bbf8fc53
VZ
85}
86
7c968cee 87void wxStringTokenizer::SetString(const wxString& str,
bbf8fc53 88 const wxString& delims,
7c968cee 89 wxStringTokenizerMode mode)
f4ada568 90{
7c968cee
VZ
91 if ( mode == wxTOKEN_DEFAULT )
92 {
93 // by default, we behave like strtok() if the delimiters are only
94 // whitespace characters and as wxTOKEN_RET_EMPTY otherwise (for
95 // whitespace delimiters, strtok() behaviour is better because we want
96 // to count consecutive spaces as one delimiter)
66c71d8a
VS
97 wxString::const_iterator p;
98 for ( p = delims.begin(); p != delims.end(); ++p )
7c968cee
VZ
99 {
100 if ( !wxIsspace(*p) )
101 break;
102 }
103
66c71d8a 104 if ( p != delims.end() )
7c968cee
VZ
105 {
106 // not whitespace char in delims
107 mode = wxTOKEN_RET_EMPTY;
108 }
109 else
110 {
111 // only whitespaces
112 mode = wxTOKEN_STRTOK;
113 }
114 }
115
f0dfc29c
VS
116#if wxUSE_UNICODE // FIXME-UTF8: only wc_str()
117 m_delims = delims.wc_str();
118#else
119 m_delims = delims.mb_str();
120#endif
121 m_delimsLen = delims.length();
122
7c968cee 123 m_mode = mode;
bbf8fc53 124
7c968cee 125 Reinit(str);
f4ada568
GL
126}
127
7c968cee 128void wxStringTokenizer::Reinit(const wxString& str)
f4ada568 129{
9a83f860 130 wxASSERT_MSG( IsOk(), wxT("you should call SetString() first") );
7c968cee
VZ
131
132 m_string = str;
f0dfc29c
VS
133 m_stringEnd = m_string.end();
134 m_pos = m_string.begin();
9a83f860 135 m_lastDelim = wxT('\0');
f0dfc29c 136 m_hasMoreTokens = MoreTokens_Unknown;
f4ada568
GL
137}
138
bbf8fc53 139// ----------------------------------------------------------------------------
7c968cee 140// access to the tokens
bbf8fc53
VZ
141// ----------------------------------------------------------------------------
142
7c968cee
VZ
143// do we have more of them?
144bool wxStringTokenizer::HasMoreTokens() const
f0dfc29c
VS
145{
146 // GetNextToken() calls HasMoreTokens() and so HasMoreTokens() is called
147 // twice in every interation in the following common usage patten:
148 // while ( HasMoreTokens() )
149 // GetNextToken();
150 // We optimize this case by caching HasMoreTokens() return value here:
151 if ( m_hasMoreTokens == MoreTokens_Unknown )
152 {
153 bool r = DoHasMoreTokens();
154 wxConstCast(this, wxStringTokenizer)->m_hasMoreTokens =
155 r ? MoreTokens_Yes : MoreTokens_No;
156 return r;
157 }
158 else
159 return m_hasMoreTokens == MoreTokens_Yes;
160}
161
162bool wxStringTokenizer::DoHasMoreTokens() const
f4ada568 163{
9a83f860 164 wxCHECK_MSG( IsOk(), false, wxT("you should call SetString() first") );
7c968cee 165
f0dfc29c
VS
166 if ( find_first_not_of(m_delims, m_delimsLen, m_pos, m_stringEnd)
167 != m_stringEnd )
bbf8fc53 168 {
4626c57c
VZ
169 // there are non delimiter characters left, so we do have more tokens
170 return true;
7c968cee 171 }
4626c57c
VZ
172
173 switch ( m_mode )
7c968cee 174 {
4626c57c
VZ
175 case wxTOKEN_RET_EMPTY:
176 case wxTOKEN_RET_DELIMS:
177 // special hack for wxTOKEN_RET_EMPTY: we should return the initial
178 // empty token even if there are only delimiters after it
f0dfc29c 179 return !m_string.empty() && m_pos == m_string.begin();
4626c57c
VZ
180
181 case wxTOKEN_RET_EMPTY_ALL:
182 // special hack for wxTOKEN_RET_EMPTY_ALL: we can know if we had
183 // already returned the trailing empty token after the last
184 // delimiter by examining m_lastDelim: it is set to NUL if we run
185 // up to the end of the string in GetNextToken(), but if it is not
186 // NUL yet we still have this last token to return even if m_pos is
187 // already at m_string.length()
9a83f860 188 return m_pos < m_stringEnd || m_lastDelim != wxT('\0');
4626c57c
VZ
189
190 case wxTOKEN_INVALID:
191 case wxTOKEN_DEFAULT:
9a83f860 192 wxFAIL_MSG( wxT("unexpected tokenizer mode") );
4626c57c
VZ
193 // fall through
194
195 case wxTOKEN_STRTOK:
196 // never return empty delimiters
197 break;
7c968cee 198 }
4626c57c
VZ
199
200 return false;
7c968cee 201}
bbf8fc53 202
4626c57c 203// count the number of (remaining) tokens in the string
7c968cee
VZ
204size_t wxStringTokenizer::CountTokens() const
205{
9a83f860 206 wxCHECK_MSG( IsOk(), 0, wxT("you should call SetString() first") );
bbf8fc53 207
7c968cee 208 // VZ: this function is IMHO not very useful, so it's probably not very
4626c57c
VZ
209 // important if its implementation here is not as efficient as it
210 // could be -- but OTOH like this we're sure to get the correct answer
7c968cee 211 // in all modes
f0dfc29c 212 wxStringTokenizer tkz(wxString(m_pos, m_stringEnd), m_delims, m_mode);
bbf8fc53 213
7c968cee 214 size_t count = 0;
4626c57c 215 while ( tkz.HasMoreTokens() )
bbf8fc53
VZ
216 {
217 count++;
7c968cee 218
4626c57c 219 (void)tkz.GetNextToken();
bbf8fc53
VZ
220 }
221
222 return count;
223}
224
225// ----------------------------------------------------------------------------
226// token extraction
227// ----------------------------------------------------------------------------
228
229wxString wxStringTokenizer::GetNextToken()
230{
231 wxString token;
7c968cee 232 do
bbf8fc53 233 {
7c968cee 234 if ( !HasMoreTokens() )
85833f5c 235 {
7c968cee 236 break;
85833f5c 237 }
4626c57c 238
f0dfc29c
VS
239 m_hasMoreTokens = MoreTokens_Unknown;
240
7c968cee 241 // find the end of this token
f0dfc29c
VS
242 wxString::const_iterator pos =
243 find_first_of(m_delims, m_delimsLen, m_pos, m_stringEnd);
7c968cee
VZ
244
245 // and the start of the next one
f0dfc29c 246 if ( pos == m_stringEnd )
85833f5c 247 {
7c968cee
VZ
248 // no more delimiters, the token is everything till the end of
249 // string
f0dfc29c 250 token.assign(m_pos, m_stringEnd);
7c968cee 251
4626c57c 252 // skip the token
f0dfc29c 253 m_pos = m_stringEnd;
bbf8fc53 254
4626c57c 255 // it wasn't terminated
9a83f860 256 m_lastDelim = wxT('\0');
85833f5c 257 }
4626c57c 258 else // we found a delimiter at pos
7c968cee 259 {
7c968cee 260 // in wxTOKEN_RET_DELIMS mode we return the delimiter character
4626c57c 261 // with token, otherwise leave it out
f06a1f33 262 wxString::const_iterator tokenEnd(pos);
4626c57c 263 if ( m_mode == wxTOKEN_RET_DELIMS )
f06a1f33 264 ++tokenEnd;
4626c57c 265
f06a1f33 266 token.assign(m_pos, tokenEnd);
dab58492 267
4626c57c
VZ
268 // skip the token and the trailing delimiter
269 m_pos = pos + 1;
bbf8fc53 270
9a83f860 271 m_lastDelim = (pos == m_stringEnd) ? wxT('\0') : (wxChar)*pos;
7c968cee 272 }
85833f5c 273 }
4626c57c 274 while ( !AllowEmpty() && token.empty() );
bbf8fc53
VZ
275
276 return token;
f4ada568 277}
1e6feb95
VZ
278
279// ----------------------------------------------------------------------------
280// public functions
281// ----------------------------------------------------------------------------
282
283wxArrayString wxStringTokenize(const wxString& str,
284 const wxString& delims,
285 wxStringTokenizerMode mode)
286{
287 wxArrayString tokens;
288 wxStringTokenizer tk(str, delims, mode);
289 while ( tk.HasMoreTokens() )
290 {
291 tokens.Add(tk.GetNextToken());
292 }
293
294 return tokens;
295}