|
1 /* |
|
2 * |
|
3 * Copyright (c) 2003 |
|
4 * John Maddock |
|
5 * |
|
6 * Use, modification and distribution are subject to the |
|
7 * Boost Software License, Version 1.0. (See accompanying file |
|
8 * LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) |
|
9 * |
|
10 */ |
|
11 |
|
12 /* |
|
13 * LOCATION: see http://www.boost.org for most recent version. |
|
14 * FILE u32regex_token_iterator.hpp |
|
15 * VERSION see <boost/version.hpp> |
|
16 * DESCRIPTION: Provides u32regex_token_iterator implementation. |
|
17 */ |
|
18 |
|
19 #ifndef BOOST_REGEX_V4_U32REGEX_TOKEN_ITERATOR_HPP |
|
20 #define BOOST_REGEX_V4_U32REGEX_TOKEN_ITERATOR_HPP |
|
21 |
|
22 #if (BOOST_WORKAROUND(__BORLANDC__, >= 0x560) && BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT(0x570)))\ |
|
23 || BOOST_WORKAROUND(BOOST_MSVC, < 1300) \ |
|
24 || BOOST_WORKAROUND(__MWERKS__, BOOST_TESTED_AT(0x3003)) |
|
25 // |
|
26 // Borland C++ Builder 6, and Visual C++ 6, |
|
27 // can't cope with the array template constructor |
|
28 // so we have a template member that will accept any type as |
|
29 // argument, and then assert that is really is an array: |
|
30 // |
|
31 #include <boost/static_assert.hpp> |
|
32 #include <boost/type_traits/is_array.hpp> |
|
33 #endif |
|
34 |
|
35 namespace boost{ |
|
36 |
|
37 #ifdef BOOST_HAS_ABI_HEADERS |
|
38 # include BOOST_ABI_PREFIX |
|
39 #endif |
|
40 #if BOOST_WORKAROUND(BOOST_MSVC, > 1300) |
|
41 # pragma warning(push) |
|
42 # pragma warning(disable:4700) |
|
43 #endif |
|
44 |
|
45 template <class BidirectionalIterator> |
|
46 class u32regex_token_iterator_implementation |
|
47 { |
|
48 typedef u32regex regex_type; |
|
49 typedef sub_match<BidirectionalIterator> value_type; |
|
50 |
|
51 match_results<BidirectionalIterator> what; // current match |
|
52 BidirectionalIterator end; // end of search area |
|
53 BidirectionalIterator base; // start of search area |
|
54 const regex_type re; // the expression |
|
55 match_flag_type flags; // match flags |
|
56 value_type result; // the current string result |
|
57 int N; // the current sub-expression being enumerated |
|
58 std::vector<int> subs; // the sub-expressions to enumerate |
|
59 |
|
60 public: |
|
61 u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, int sub, match_flag_type f) |
|
62 : end(last), re(*p), flags(f){ subs.push_back(sub); } |
|
63 u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, const std::vector<int>& v, match_flag_type f) |
|
64 : end(last), re(*p), flags(f), subs(v){} |
|
65 #if BOOST_WORKAROUND(BOOST_MSVC, < 1300) |
|
66 // can't reliably get this to work.... |
|
67 #elif (BOOST_WORKAROUND(__BORLANDC__, >= 0x560) && BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT(0x570)))\ |
|
68 || BOOST_WORKAROUND(BOOST_MSVC, < 1300) \ |
|
69 || BOOST_WORKAROUND(__MWERKS__, BOOST_TESTED_AT(0x3003)) \ |
|
70 || BOOST_WORKAROUND(__HP_aCC, < 60700) |
|
71 template <class T> |
|
72 u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, const T& submatches, match_flag_type f) |
|
73 : end(last), re(*p), flags(f) |
|
74 { |
|
75 // assert that T really is an array: |
|
76 BOOST_STATIC_ASSERT(::boost::is_array<T>::value); |
|
77 const std::size_t array_size = sizeof(T) / sizeof(submatches[0]); |
|
78 for(std::size_t i = 0; i < array_size; ++i) |
|
79 { |
|
80 subs.push_back(submatches[i]); |
|
81 } |
|
82 } |
|
83 #else |
|
84 template <std::size_t CN> |
|
85 u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, const int (&submatches)[CN], match_flag_type f) |
|
86 : end(last), re(*p), flags(f) |
|
87 { |
|
88 for(std::size_t i = 0; i < CN; ++i) |
|
89 { |
|
90 subs.push_back(submatches[i]); |
|
91 } |
|
92 } |
|
93 #endif |
|
94 |
|
95 bool init(BidirectionalIterator first) |
|
96 { |
|
97 base = first; |
|
98 N = 0; |
|
99 if(u32regex_search(first, end, what, re, flags, base) == true) |
|
100 { |
|
101 N = 0; |
|
102 result = ((subs[N] == -1) ? what.prefix() : what[(int)subs[N]]); |
|
103 return true; |
|
104 } |
|
105 else if((subs[N] == -1) && (first != end)) |
|
106 { |
|
107 result.first = first; |
|
108 result.second = end; |
|
109 result.matched = (first != end); |
|
110 N = -1; |
|
111 return true; |
|
112 } |
|
113 return false; |
|
114 } |
|
115 bool compare(const u32regex_token_iterator_implementation& that) |
|
116 { |
|
117 if(this == &that) return true; |
|
118 return (&re.get_data() == &that.re.get_data()) |
|
119 && (end == that.end) |
|
120 && (flags == that.flags) |
|
121 && (N == that.N) |
|
122 && (what[0].first == that.what[0].first) |
|
123 && (what[0].second == that.what[0].second); |
|
124 } |
|
125 const value_type& get() |
|
126 { return result; } |
|
127 bool next() |
|
128 { |
|
129 if(N == -1) |
|
130 return false; |
|
131 if(N+1 < (int)subs.size()) |
|
132 { |
|
133 ++N; |
|
134 result =((subs[N] == -1) ? what.prefix() : what[subs[N]]); |
|
135 return true; |
|
136 } |
|
137 //if(what.prefix().first != what[0].second) |
|
138 // flags |= match_prev_avail | regex_constants::match_not_bob; |
|
139 BidirectionalIterator last_end(what[0].second); |
|
140 if(u32regex_search(last_end, end, what, re, ((what[0].first == what[0].second) ? flags | regex_constants::match_not_initial_null : flags), base)) |
|
141 { |
|
142 N =0; |
|
143 result =((subs[N] == -1) ? what.prefix() : what[subs[N]]); |
|
144 return true; |
|
145 } |
|
146 else if((last_end != end) && (subs[0] == -1)) |
|
147 { |
|
148 N =-1; |
|
149 result.first = last_end; |
|
150 result.second = end; |
|
151 result.matched = (last_end != end); |
|
152 return true; |
|
153 } |
|
154 return false; |
|
155 } |
|
156 private: |
|
157 u32regex_token_iterator_implementation& operator=(const u32regex_token_iterator_implementation&); |
|
158 }; |
|
159 |
|
160 template <class BidirectionalIterator> |
|
161 class u32regex_token_iterator |
|
162 #ifndef BOOST_NO_STD_ITERATOR |
|
163 : public std::iterator< |
|
164 std::forward_iterator_tag, |
|
165 sub_match<BidirectionalIterator>, |
|
166 typename re_detail::regex_iterator_traits<BidirectionalIterator>::difference_type, |
|
167 const sub_match<BidirectionalIterator>*, |
|
168 const sub_match<BidirectionalIterator>& > |
|
169 #endif |
|
170 { |
|
171 private: |
|
172 typedef u32regex_token_iterator_implementation<BidirectionalIterator> impl; |
|
173 typedef shared_ptr<impl> pimpl; |
|
174 public: |
|
175 typedef u32regex regex_type; |
|
176 typedef sub_match<BidirectionalIterator> value_type; |
|
177 typedef typename re_detail::regex_iterator_traits<BidirectionalIterator>::difference_type |
|
178 difference_type; |
|
179 typedef const value_type* pointer; |
|
180 typedef const value_type& reference; |
|
181 typedef std::forward_iterator_tag iterator_category; |
|
182 |
|
183 u32regex_token_iterator(){} |
|
184 u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re, |
|
185 int submatch = 0, match_flag_type m = match_default) |
|
186 : pdata(new impl(&re, b, submatch, m)) |
|
187 { |
|
188 if(!pdata->init(a)) |
|
189 pdata.reset(); |
|
190 } |
|
191 u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re, |
|
192 const std::vector<int>& submatches, match_flag_type m = match_default) |
|
193 : pdata(new impl(&re, b, submatches, m)) |
|
194 { |
|
195 if(!pdata->init(a)) |
|
196 pdata.reset(); |
|
197 } |
|
198 #if BOOST_WORKAROUND(BOOST_MSVC, < 1300) |
|
199 // can't reliably get this to work.... |
|
200 #elif (BOOST_WORKAROUND(__BORLANDC__, >= 0x560) && BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT(0x570)))\ |
|
201 || BOOST_WORKAROUND(BOOST_MSVC, < 1300) \ |
|
202 || BOOST_WORKAROUND(__MWERKS__, BOOST_TESTED_AT(0x3003)) \ |
|
203 || BOOST_WORKAROUND(__HP_aCC, < 60700) |
|
204 template <class T> |
|
205 u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re, |
|
206 const T& submatches, match_flag_type m = match_default) |
|
207 : pdata(new impl(&re, b, submatches, m)) |
|
208 { |
|
209 if(!pdata->init(a)) |
|
210 pdata.reset(); |
|
211 } |
|
212 #else |
|
213 template <std::size_t N> |
|
214 u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re, |
|
215 const int (&submatches)[N], match_flag_type m = match_default) |
|
216 : pdata(new impl(&re, b, submatches, m)) |
|
217 { |
|
218 if(!pdata->init(a)) |
|
219 pdata.reset(); |
|
220 } |
|
221 #endif |
|
222 u32regex_token_iterator(const u32regex_token_iterator& that) |
|
223 : pdata(that.pdata) {} |
|
224 u32regex_token_iterator& operator=(const u32regex_token_iterator& that) |
|
225 { |
|
226 pdata = that.pdata; |
|
227 return *this; |
|
228 } |
|
229 bool operator==(const u32regex_token_iterator& that)const |
|
230 { |
|
231 if((pdata.get() == 0) || (that.pdata.get() == 0)) |
|
232 return pdata.get() == that.pdata.get(); |
|
233 return pdata->compare(*(that.pdata.get())); |
|
234 } |
|
235 bool operator!=(const u32regex_token_iterator& that)const |
|
236 { return !(*this == that); } |
|
237 const value_type& operator*()const |
|
238 { return pdata->get(); } |
|
239 const value_type* operator->()const |
|
240 { return &(pdata->get()); } |
|
241 u32regex_token_iterator& operator++() |
|
242 { |
|
243 cow(); |
|
244 if(0 == pdata->next()) |
|
245 { |
|
246 pdata.reset(); |
|
247 } |
|
248 return *this; |
|
249 } |
|
250 u32regex_token_iterator operator++(int) |
|
251 { |
|
252 u32regex_token_iterator result(*this); |
|
253 ++(*this); |
|
254 return result; |
|
255 } |
|
256 private: |
|
257 |
|
258 pimpl pdata; |
|
259 |
|
260 void cow() |
|
261 { |
|
262 // copy-on-write |
|
263 if(pdata.get() && !pdata.unique()) |
|
264 { |
|
265 pdata.reset(new impl(*(pdata.get()))); |
|
266 } |
|
267 } |
|
268 }; |
|
269 |
|
270 typedef u32regex_token_iterator<const char*> utf8regex_token_iterator; |
|
271 typedef u32regex_token_iterator<const UChar*> utf16regex_token_iterator; |
|
272 typedef u32regex_token_iterator<const UChar32*> utf32regex_token_iterator; |
|
273 |
|
274 // construction from an integral sub_match state_id: |
|
275 inline u32regex_token_iterator<const char*> make_u32regex_token_iterator(const char* p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default) |
|
276 { |
|
277 return u32regex_token_iterator<const char*>(p, p+std::strlen(p), e, submatch, m); |
|
278 } |
|
279 #ifndef BOOST_NO_WREGEX |
|
280 inline u32regex_token_iterator<const wchar_t*> make_u32regex_token_iterator(const wchar_t* p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default) |
|
281 { |
|
282 return u32regex_token_iterator<const wchar_t*>(p, p+std::wcslen(p), e, submatch, m); |
|
283 } |
|
284 #endif |
|
285 #if !defined(U_WCHAR_IS_UTF16) && (U_SIZEOF_WCHAR_T != 2) |
|
286 inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UChar* p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default) |
|
287 { |
|
288 return u32regex_token_iterator<const UChar*>(p, p+u_strlen(p), e, submatch, m); |
|
289 } |
|
290 #endif |
|
291 template <class charT, class Traits, class Alloc> |
|
292 inline u32regex_token_iterator<typename std::basic_string<charT, Traits, Alloc>::const_iterator> make_u32regex_token_iterator(const std::basic_string<charT, Traits, Alloc>& p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default) |
|
293 { |
|
294 typedef typename std::basic_string<charT, Traits, Alloc>::const_iterator iter_type; |
|
295 return u32regex_token_iterator<iter_type>(p.begin(), p.end(), e, m); |
|
296 } |
|
297 inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UnicodeString& s, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default) |
|
298 { |
|
299 return u32regex_token_iterator<const UChar*>(s.getBuffer(), s.getBuffer() + s.length(), e, submatch, m); |
|
300 } |
|
301 |
|
302 #if !BOOST_WORKAROUND(BOOST_MSVC, < 1300) |
|
303 // construction from a reference to an array: |
|
304 template <std::size_t N> |
|
305 inline u32regex_token_iterator<const char*> make_u32regex_token_iterator(const char* p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default) |
|
306 { |
|
307 return u32regex_token_iterator<const char*>(p, p+std::strlen(p), e, submatch, m); |
|
308 } |
|
309 #ifndef BOOST_NO_WREGEX |
|
310 template <std::size_t N> |
|
311 inline u32regex_token_iterator<const wchar_t*> make_u32regex_token_iterator(const wchar_t* p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default) |
|
312 { |
|
313 return u32regex_token_iterator<const wchar_t*>(p, p+std::wcslen(p), e, submatch, m); |
|
314 } |
|
315 #endif |
|
316 #if !defined(U_WCHAR_IS_UTF16) && (U_SIZEOF_WCHAR_T != 2) |
|
317 template <std::size_t N> |
|
318 inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UChar* p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default) |
|
319 { |
|
320 return u32regex_token_iterator<const UChar*>(p, p+u_strlen(p), e, m); |
|
321 } |
|
322 #endif |
|
323 template <class charT, class Traits, class Alloc, std::size_t N> |
|
324 inline u32regex_token_iterator<typename std::basic_string<charT, Traits, Alloc>::const_iterator> make_u32regex_token_iterator(const std::basic_string<charT, Traits, Alloc>& p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default) |
|
325 { |
|
326 typedef typename std::basic_string<charT, Traits, Alloc>::const_iterator iter_type; |
|
327 return u32regex_token_iterator<iter_type>(p.begin(), p.end(), e, m); |
|
328 } |
|
329 template <std::size_t N> |
|
330 inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UnicodeString& s, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default) |
|
331 { |
|
332 return u32regex_token_iterator<const UChar*>(s.getBuffer(), s.getBuffer() + s.length(), e, submatch, m); |
|
333 } |
|
334 #endif // BOOST_MSVC < 1300 |
|
335 |
|
336 // construction from a vector of sub_match state_id's: |
|
337 inline u32regex_token_iterator<const char*> make_u32regex_token_iterator(const char* p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default) |
|
338 { |
|
339 return u32regex_token_iterator<const char*>(p, p+std::strlen(p), e, submatch, m); |
|
340 } |
|
341 #ifndef BOOST_NO_WREGEX |
|
342 inline u32regex_token_iterator<const wchar_t*> make_u32regex_token_iterator(const wchar_t* p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default) |
|
343 { |
|
344 return u32regex_token_iterator<const wchar_t*>(p, p+std::wcslen(p), e, submatch, m); |
|
345 } |
|
346 #endif |
|
347 #if !defined(U_WCHAR_IS_UTF16) && (U_SIZEOF_WCHAR_T != 2) |
|
348 inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UChar* p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default) |
|
349 { |
|
350 return u32regex_token_iterator<const UChar*>(p, p+u_strlen(p), e, submatch, m); |
|
351 } |
|
352 #endif |
|
353 template <class charT, class Traits, class Alloc> |
|
354 inline u32regex_token_iterator<typename std::basic_string<charT, Traits, Alloc>::const_iterator> make_u32regex_token_iterator(const std::basic_string<charT, Traits, Alloc>& p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default) |
|
355 { |
|
356 typedef typename std::basic_string<charT, Traits, Alloc>::const_iterator iter_type; |
|
357 return u32regex_token_iterator<iter_type>(p.begin(), p.end(), e, m); |
|
358 } |
|
359 inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UnicodeString& s, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default) |
|
360 { |
|
361 return u32regex_token_iterator<const UChar*>(s.getBuffer(), s.getBuffer() + s.length(), e, submatch, m); |
|
362 } |
|
363 |
|
364 #if BOOST_WORKAROUND(BOOST_MSVC, == 1310) |
|
365 # pragma warning(pop) |
|
366 #endif |
|
367 #ifdef BOOST_HAS_ABI_HEADERS |
|
368 # include BOOST_ABI_SUFFIX |
|
369 #endif |
|
370 |
|
371 } // namespace boost |
|
372 |
|
373 #endif // BOOST_REGEX_V4_REGEX_TOKEN_ITERATOR_HPP |
|
374 |
|
375 |
|
376 |
|
377 |