1 /* $NetBSD: chartype.h,v 1.10 2011/11/16 01:45:10 christos Exp $ */
2
3 /*-
4 * Copyright (c) 2009 The NetBSD Foundation, Inc.
5 * All rights reserved.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
15 * 3. All advertising materials mentioning features or use of this software
16 * must display the following acknowledgement:
17 * This product includes software developed by the NetBSD
18 * Foundation, Inc. and its contributors.
19 * 4. Neither the name of The NetBSD Foundation nor the names of its
20 * contributors may be used to endorse or promote products derived
21 * from this software without specific prior written permission.
22 *
23 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
24 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
25 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
26 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
27 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
28 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
29 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
30 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
31 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
32 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
33 * POSSIBILITY OF SUCH DAMAGE.
34 */
35
36 #ifndef _h_chartype_f
37 #define _h_chartype_f
38
39
40
41 #ifdef WIDECHAR
42 /*
43 * The test here in the original version of the source had a white-list
44 * of platforms and gave an "#error" if there wa an attempt to build
45 * otherwise. The list did not include FreeBSD and so for utterly
46 * frivolous reasons the code would not compile there. I have changed the
47 * error to a warning. If __GNUC__ is defined there should be a "#warning"
48 * directive that can be used, otherwise I hope that use of a (probably)
49 * undefined #pragma will cause the compiler to say something.
50 * At present I find it hard to see the possibility of modern machines that
51 * do not use the first 127 codepoints as expected. A LONG while in the past
52 * one might have had EBCDIC. ACN May 2016
53 *
54 * Aha - I have done some more research. In some cases especially in CJK
55 * markets wide strings might be encoded with each wchar_t holding just
56 * a concatenation of bytes from an earlier multi-byte extended representation,
57 * and that could perhaps lead to the presence of bytes in the range 0-127
58 * not standing for ASCII characters. That would relate to ISO 2022. The
59 * code here in libedit really wants wchar_t to be a 32-bit type so that
60 * *ALL* Unicode characters can be represented. Where that is the case it
61 * will be able to accept UTF-8 input. On platforms where wchar_t is only
62 * a 16-bit type it will only be possible to cope with characters in the
63 * basic multilingual plane. The most common symbols NOT present there are
64 * perhaps some mathematical symbols and alphabets (including Gothic) and
65 * a load of Emojis. So those who use them and try (for instance) using copy
66 * and paste to insert them input input should be aware that on some platforms
67 * they will not be well supported.
68 */
69
70 /* Ideally we should also test the value of the define to see if it
71 * supports non-BMP code points without requiring UTF-16, but nothing
72 * seems to actually advertise this properly, despite Unicode 3.1 having
73 * been around since 2001... */
74 #if !defined(__NetBSD__) && !defined(__sun) && \
75 !(defined(__APPLE__) && defined(__MACH__)) && \
76 !(defined(__FreeBSD__) || defined(__DragonFly__)) && !defined(__CYGWIN__)
77 #ifndef __STDC_ISO_10646__
78 /* In many places it is assumed that the first 127 code points are ASCII
79 * compatible, so ensure wchar_t indeed does ISO 10646 and not some other
80 * funky encoding that could break us in weird and wonderful ways. */
81 #ifdef __GNUC__
82 #warning wchar_t must store ISO 10646 characters
83 #else
84 #pragra warn_the_user wchar_t must store ISO 10646 characters
85 #endif
86 #endif
87 #endif
88
89 /* Oh for a <uchar.h> with char32_t and __STDC_UTF_32__ in it...
90 * ref: ISO/IEC DTR 19769
91 */
92 #if WCHAR_MAX < INT32_MAX
93 #ifdef __GNUC__
94 /*
95 * Note that the "#warning" directive is not part of standard C. However
96 * it is supported by gcc and clang (and clang, at least on a Macintosh,
97 * predefined __GNUC__). So I will enable this just in cases where I am
98 * confident that it will be a warning and not an error.
99 */
100 #warning Build environment does not support non-BMP characters
101 #else
102 /*
103 * On other compilers I am going to hope that there is no built-in
104 * pragma called "warn_the_user" and that the compiler will report
105 * when an undefined pragma is encountered. I might have hoped that
106 * "#pragma warning ..." would be the idiom to use, but there are cases
107 * where that is used to control when the compiler generates warnings rather
108 * than to generate one instantly.
109 */
110 #pragma warn_the_user "Build environment does not support non-BMP characters"
111 #endif
112 #endif
113
114 #ifndef HAVE_WCSDUP
115 wchar_t *wcsdup(const wchar_t *s);
116 #endif
117
118 #define ct_mbtowc mbtowc
119 #define ct_mbtowc_reset mbtowc(0,0,(size_t)0)
120 #define ct_wctomb wctomb
121 #define ct_wctomb_reset wctomb(0,0)
122 #define ct_wcstombs wcstombs
123 #define ct_mbstowcs mbstowcs
124
125 #define Char wchar_t
126 #define Int wint_t
127 #define FUN(prefix,rest) prefix ## _w ## rest
128 #define FUNW(type) type ## _w
129 #define TYPE(type) type ## W
130 #define FSTR "%ls"
131 #define STR(x) L ## x
132 #define UC(c) c
133 #define Isalpha(x) iswalpha(x)
134 #define Isalnum(x) iswalnum(x)
135 #define Isgraph(x) iswgraph(x)
136 #define Isspace(x) iswspace(x)
137 #define Isdigit(x) iswdigit(x)
138 #define Iscntrl(x) iswcntrl(x)
139 #define Isprint(x) iswprint(x)
140
141 #define Isupper(x) iswupper(x)
142 #define Islower(x) iswlower(x)
143 #define Toupper(x) towupper(x)
144 #define Tolower(x) towlower(x)
145
146 #define IsASCII(x) (x < 0x100)
147
148 #define Strlen(x) wcslen(x)
149 #define Strchr(s,c) wcschr(s,c)
150 #define Strrchr(s,c) wcsrchr(s,c)
151 #define Strstr(s,v) wcsstr(s,v)
152 #define Strdup(x) wcsdup(x)
153 #define Strcpy(d,s) wcscpy(d,s)
154 #define Strncpy(d,s,n) wcsncpy(d,s,n)
155 #define Strncat(d,s,n) wcsncat(d,s,n)
156
157 #define Strcmp(s,v) wcscmp(s,v)
158 #define Strncmp(s,v,n) wcsncmp(s,v,n)
159 #define Strcspn(s,r) wcscspn(s,r)
160
161 #define Strtol(p,e,b) wcstol(p,e,b)
162
163 static inline int
Width(wchar_t c)164 Width(wchar_t c)
165 {
166 int w = wcwidth(c);
167 return w < 0 ? 0 : w;
168 }
169
170 #else /* NARROW */
171
172 #define ct_mbtowc error
173 #define ct_mbtowc_reset
174 #define ct_wctomb error
175 #define ct_wctomb_reset
176 #define ct_wcstombs(a, b, c) (strncpy(a, b, c), strlen(a))
177 #define ct_mbstowcs(a, b, c) (strncpy(a, b, c), strlen(a))
178
179 #define Char char
180 #define Int int
181 #define FUN(prefix,rest) prefix ## _ ## rest
182 #define FUNW(type) type
183 #define TYPE(type) type
184 #define FSTR "%s"
185 #define STR(x) x
186 #define UC(c) (unsigned char)(c)
187
188 #define Isalpha(x) isalpha((unsigned char)x)
189 #define Isalnum(x) isalnum((unsigned char)x)
190 #define Isgraph(x) isgraph((unsigned char)x)
191 #define Isspace(x) isspace((unsigned char)x)
192 #define Isdigit(x) isdigit((unsigned char)x)
193 #define Iscntrl(x) iscntrl((unsigned char)x)
194 #define Isprint(x) isprint((unsigned char)x)
195
196 #define Isupper(x) isupper((unsigned char)x)
197 #define Islower(x) islower((unsigned char)x)
198 #define Toupper(x) toupper((unsigned char)x)
199 #define Tolower(x) tolower((unsigned char)x)
200
201 #define IsASCII(x) isascii((unsigned char)x)
202
203 #define Strlen(x) strlen(x)
204 #define Strchr(s,c) strchr(s,c)
205 #define Strrchr(s,c) strrchr(s,c)
206 #define Strstr(s,v) strstr(s,v)
207 #define Strdup(x) strdup(x)
208 #define Strcpy(d,s) strcpy(d,s)
209 #define Strncpy(d,s,n) strncpy(d,s,n)
210 #define Strncat(d,s,n) strncat(d,s,n)
211
212 #define Strcmp(s,v) strcmp(s,v)
213 #define Strncmp(s,v,n) strncmp(s,v,n)
214 #define Strcspn(s,r) strcspn(s,r)
215
216 #define Strtol(p,e,b) strtol(p,e,b)
217
218 #define Width(c) 1
219
220 #endif
221
222
223 #ifdef WIDECHAR
224 /*
225 * Conversion buffer
226 */
227 typedef struct ct_buffer_t {
228 char *cbuff;
229 size_t csize;
230 Char *wbuff;
231 size_t wsize;
232 } ct_buffer_t;
233
234 #define ct_encode_string __ct_encode_string
235 /* Encode a wide-character string and return the UTF-8 encoded result. */
236 public char *ct_encode_string(const Char *, ct_buffer_t *);
237
238 #define ct_decode_string __ct_decode_string
239 /* Decode a (multi)?byte string and return the wide-character string result. */
240 public Char *ct_decode_string(const char *, ct_buffer_t *);
241
242 /* Decode a (multi)?byte argv string array.
243 * The pointer returned must be free()d when done. */
244 protected Char **ct_decode_argv(int, const char *[], ct_buffer_t *);
245
246 /* Resizes the conversion buffer(s) if needed. */
247 protected void ct_conv_buff_resize(ct_buffer_t *, size_t, size_t);
248 protected ssize_t ct_encode_char(char *, size_t, Char);
249 protected size_t ct_enc_width(Char);
250
251 #define ct_free_argv(s) el_free(s)
252
253 #else
254 #define ct_encode_string(s, b) (s)
255 #define ct_decode_string(s, b) (s)
256 #define ct_decode_argv(l, s, b) (s)
257 #define ct_conv_buff_resize(b, os, ns)
258 #define ct_encode_char(d, l, s) (*d = s, 1)
259 #define ct_free_argv(s)
260 #endif
261
262 #ifndef NARROWCHAR
263 /* Encode a characted into the destination buffer, provided there is sufficent
264 * buffer space available. Returns the number of bytes used up (zero if the
265 * character cannot be encoded, -1 if there was not enough space available). */
266
267 /* The maximum buffer size to hold the most unwieldly visual representation,
268 * in this case \U+nnnnn. */
269 #define VISUAL_WIDTH_MAX ((size_t)8)
270
271 /* The terminal is thought of in terms of X columns by Y lines. In the cases
272 * where a wide character takes up more than one column, the adjacent
273 * occupied column entries will contain this faux character. */
274 #define MB_FILL_CHAR ((Char)-1)
275
276 /* Visual width of character c, taking into account ^? , \0177 and \U+nnnnn
277 * style visual expansions. */
278 protected int ct_visual_width(Char);
279
280 /* Turn the given character into the appropriate visual format, matching
281 * the width given by ct_visual_width(). Returns the number of characters used
282 * up, or -1 if insufficient space. Buffer length is in count of Char's. */
283 protected ssize_t ct_visual_char(Char *, size_t, Char);
284
285 /* Convert the given string into visual format, using the ct_visual_char()
286 * function. Uses a static buffer, so not threadsafe. */
287 protected const Char *ct_visual_string(const Char *);
288
289
290 /* printable character, use ct_visual_width() to find out display width */
291 #define CHTYPE_PRINT ( 0)
292 /* control character found inside the ASCII portion of the charset */
293 #define CHTYPE_ASCIICTL (-1)
294 /* a \t */
295 #define CHTYPE_TAB (-2)
296 /* a \n */
297 #define CHTYPE_NL (-3)
298 /* non-printable character */
299 #define CHTYPE_NONPRINT (-4)
300 /* classification of character c, as one of the above defines */
301 protected int ct_chr_class(Char c);
302 #endif
303
304
305 #endif /* _chartype_f */
306