1 /* Iterating through multibyte strings: macros for multi-byte encodings.
2 Copyright (C) 2001, 2005, 2007 Free Software Foundation, Inc.
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 3 of the License, or
7 (at your option) any later version.
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.
14 You should have received a copy of the GNU General Public License
15 along with this program. If not, see <http://www.gnu.org/licenses/>. */
17 /* Written by Bruno Haible <bruno@clisp.org>. */
19 /* The macros in this file implement forward iteration through a
22 With these macros, an iteration loop that looks like
25 for (iter = buf; iter < buf + buflen; iter++)
33 for (mbi_init (iter, buf, buflen); mbi_avail (iter); mbi_advance (iter))
35 do_something (mbi_cur_ptr (iter), mb_len (mbi_cur (iter)));
38 The benefit of these macros over plain use of mbrtowc is:
39 - Handling of invalid multibyte sequences is possible without
40 making the code more complicated, while still preserving the
41 invalid multibyte sequences.
44 is a type usable for variable declarations.
46 mbi_init (iter, startptr, length)
47 initializes the iterator, starting at startptr and crossing length bytes.
50 returns true if there are more multibyte chracters available before
51 the end of string is reached. In this case, mbi_cur (iter) is
52 initialized to the next multibyte chracter.
55 advances the iterator by one multibyte character.
58 returns the current multibyte character, of type mbchar_t. All the
59 macros defined in mbchar.h can be used on it.
62 return a pointer to the beginning of the current multibyte character.
64 mbi_reloc (iter, ptrdiff)
65 relocates iterator when the string is moved by ptrdiff bytes.
67 mbi_copy (&destiter, &srciter)
68 copies srciter to destiter.
70 Here are the function prototypes of the macros.
72 extern void mbi_init (mbi_iterator_t iter,
73 const char *startptr, size_t length);
74 extern bool mbi_avail (mbi_iterator_t iter);
75 extern void mbi_advance (mbi_iterator_t iter);
76 extern mbchar_t mbi_cur (mbi_iterator_t iter);
77 extern const char * mbi_cur_ptr (mbi_iterator_t iter);
78 extern void mbi_reloc (mbi_iterator_t iter, ptrdiff_t ptrdiff);
79 extern void mbi_copy (mbi_iterator_t *new, const mbi_iterator_t *old);
90 /* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before
92 BSD/OS 4.1 has a bug: <stdio.h> and <time.h> must be included before
102 const char *limit; /* pointer to end of string */
103 bool in_shift; /* true if next byte may not be interpreted as ASCII */
104 mbstate_t state; /* if in_shift: current shift state */
105 bool next_done; /* true if mbi_avail has already filled the following */
106 struct mbchar cur; /* the current character:
107 const char *cur.ptr pointer to current character
108 The following are only valid after mbi_avail.
109 size_t cur.bytes number of bytes of current character
110 bool cur.wc_valid true if wc is a valid wide character
111 wchar_t cur.wc if wc_valid: the current character
116 mbiter_multi_next (struct mbiter_multi *iter)
122 /* Handle most ASCII characters quickly, without calling mbrtowc(). */
123 if (is_basic (*iter->cur.ptr))
125 /* These characters are part of the basic character set. ISO C 99
126 guarantees that their wide character code is identical to their
129 iter->cur.wc = *iter->cur.ptr;
130 iter->cur.wc_valid = true;
134 assert (mbsinit (&iter->state));
135 iter->in_shift = true;
137 iter->cur.bytes = mbrtowc (&iter->cur.wc, iter->cur.ptr,
138 iter->limit - iter->cur.ptr, &iter->state);
139 if (iter->cur.bytes == (size_t) -1)
141 /* An invalid multibyte sequence was encountered. */
143 iter->cur.wc_valid = false;
144 /* Whether to set iter->in_shift = false and reset iter->state
145 or not is not very important; the string is bogus anyway. */
147 else if (iter->cur.bytes == (size_t) -2)
149 /* An incomplete multibyte character at the end. */
150 iter->cur.bytes = iter->limit - iter->cur.ptr;
151 iter->cur.wc_valid = false;
152 /* Whether to set iter->in_shift = false and reset iter->state
153 or not is not important; the string end is reached anyway. */
157 if (iter->cur.bytes == 0)
159 /* A null wide character was encountered. */
161 assert (*iter->cur.ptr == '\0');
162 assert (iter->cur.wc == 0);
164 iter->cur.wc_valid = true;
166 /* When in the initial state, we can go back treating ASCII
167 characters more quickly. */
168 if (mbsinit (&iter->state))
169 iter->in_shift = false;
172 iter->next_done = true;
176 mbiter_multi_reloc (struct mbiter_multi *iter, ptrdiff_t ptrdiff)
178 iter->cur.ptr += ptrdiff;
179 iter->limit += ptrdiff;
183 mbiter_multi_copy (struct mbiter_multi *new_iter, const struct mbiter_multi *old_iter)
185 new_iter->limit = old_iter->limit;
186 if ((new_iter->in_shift = old_iter->in_shift))
187 memcpy (&new_iter->state, &old_iter->state, sizeof (mbstate_t));
189 memset (&new_iter->state, 0, sizeof (mbstate_t));
190 new_iter->next_done = old_iter->next_done;
191 mb_copy (&new_iter->cur, &old_iter->cur);
194 /* Iteration macros. */
195 typedef struct mbiter_multi mbi_iterator_t;
196 #define mbi_init(iter, startptr, length) \
197 ((iter).cur.ptr = (startptr), (iter).limit = (iter).cur.ptr + (length), \
198 (iter).in_shift = false, memset (&(iter).state, '\0', sizeof (mbstate_t)), \
199 (iter).next_done = false)
200 #define mbi_avail(iter) \
201 ((iter).cur.ptr < (iter).limit && (mbiter_multi_next (&(iter)), true))
202 #define mbi_advance(iter) \
203 ((iter).cur.ptr += (iter).cur.bytes, (iter).next_done = false)
205 /* Access to the current character. */
206 #define mbi_cur(iter) (iter).cur
207 #define mbi_cur_ptr(iter) (iter).cur.ptr
210 #define mbi_reloc(iter, ptrdiff) mbiter_multi_reloc (&iter, ptrdiff)
212 /* Copying an iterator. */
213 #define mbi_copy mbiter_multi_copy
215 #endif /* _MBITER_H */