Commit | Line | Data |
---|---|---|
a0d0e21e | 1 | /* handy.h |
a687059c | 2 | * |
1129b882 NC |
3 | * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1999, 2000, |
4 | * 2001, 2002, 2004, 2005, 2006, 2007, 2008 by Larry Wall and others | |
a687059c | 5 | * |
6e21c824 LW |
6 | * You may distribute under the terms of either the GNU General Public |
7 | * License or the Artistic License, as specified in the README file. | |
8d063cd8 | 8 | * |
8d063cd8 LW |
9 | */ |
10 | ||
85e6fe83 | 11 | #if !defined(__STDC__) |
378cc40b LW |
12 | #ifdef NULL |
13 | #undef NULL | |
14 | #endif | |
a687059c LW |
15 | #ifndef I286 |
16 | # define NULL 0 | |
17 | #else | |
18 | # define NULL 0L | |
19 | #endif | |
85e6fe83 LW |
20 | #endif |
21 | ||
24792b8d NC |
22 | #ifndef PERL_CORE |
23 | # define Null(type) ((type)NULL) | |
954c1994 GS |
24 | |
25 | /* | |
ccfc67b7 | 26 | =head1 Handy Values |
954c1994 | 27 | |
3609ea0d | 28 | =for apidoc AmU||Nullch |
24792b8d | 29 | Null character pointer. (No longer available when C<PERL_CORE> is defined.) |
2307c6d0 | 30 | |
954c1994 | 31 | =for apidoc AmU||Nullsv |
24792b8d | 32 | Null SV pointer. (No longer available when C<PERL_CORE> is defined.) |
954c1994 GS |
33 | |
34 | =cut | |
35 | */ | |
36 | ||
24792b8d NC |
37 | # define Nullch Null(char*) |
38 | # define Nullfp Null(PerlIO*) | |
39 | # define Nullsv Null(SV*) | |
40 | #endif | |
8d063cd8 | 41 | |
641d3f0b | 42 | #ifdef TRUE |
43 | #undef TRUE | |
44 | #endif | |
45 | #ifdef FALSE | |
46 | #undef FALSE | |
47 | #endif | |
48 | #define TRUE (1) | |
49 | #define FALSE (0) | |
50 | ||
cf3f0ffb DM |
51 | /* The MUTABLE_*() macros cast pointers to the types shown, in such a way |
52 | * (compiler permitting) that casting away const-ness will give a warning; | |
53 | * e.g.: | |
54 | * | |
55 | * const SV *sv = ...; | |
56 | * AV *av1 = (AV*)sv; <== BAD: the const has been silently cast away | |
57 | * AV *av2 = MUTABLE_AV(sv); <== GOOD: it may warn | |
58 | */ | |
59 | ||
b1bc3f34 | 60 | #if defined(__GNUC__) && !defined(PERL_GCC_BRACE_GROUPS_FORBIDDEN) |
f065df04 | 61 | # define MUTABLE_PTR(p) ({ void *_p = (p); _p; }) |
b1bc3f34 NC |
62 | #else |
63 | # define MUTABLE_PTR(p) ((void *) (p)) | |
64 | #endif | |
65 | ||
a062e10d | 66 | #define MUTABLE_AV(p) ((AV *)MUTABLE_PTR(p)) |
ea726b52 | 67 | #define MUTABLE_CV(p) ((CV *)MUTABLE_PTR(p)) |
159b6efe | 68 | #define MUTABLE_GV(p) ((GV *)MUTABLE_PTR(p)) |
dbebbdb4 | 69 | #define MUTABLE_HV(p) ((HV *)MUTABLE_PTR(p)) |
a45c7426 | 70 | #define MUTABLE_IO(p) ((IO *)MUTABLE_PTR(p)) |
b1bc3f34 | 71 | #define MUTABLE_SV(p) ((SV *)MUTABLE_PTR(p)) |
27d4fb96 | 72 | |
73 | /* XXX Configure ought to have a test for a boolean type, if I can | |
74 | just figure out all the headers such a test needs. | |
75 | Andy Dougherty August 1996 | |
76 | */ | |
8e84507e | 77 | /* bool is built-in for g++-2.6.3 and later, which might be used |
c1d22f6b GS |
78 | for extensions. <_G_config.h> defines _G_HAVE_BOOL, but we can't |
79 | be sure _G_config.h will be included before this file. _G_config.h | |
8e84507e | 80 | also defines _G_HAVE_BOOL for both gcc and g++, but only g++ |
c1d22f6b GS |
81 | actually has bool. Hence, _G_HAVE_BOOL is pretty useless for us. |
82 | g++ can be identified by __GNUG__. | |
83 | Andy Dougherty February 2000 | |
5d94fbed | 84 | */ |
3609ea0d | 85 | #ifdef __GNUG__ /* GNU g++ has bool built-in */ |
5d94fbed | 86 | # ifndef HAS_BOOL |
c1d22f6b | 87 | # define HAS_BOOL 1 |
5d94fbed | 88 | # endif |
5d94fbed AD |
89 | #endif |
90 | ||
641d3f0b | 91 | /* The NeXT dynamic loader headers will not build with the bool macro |
92 | So declare them now to clear confusion. | |
93 | */ | |
8f1f23e8 | 94 | #if defined(NeXT) || defined(__NeXT__) |
641d3f0b | 95 | # undef FALSE |
96 | # undef TRUE | |
97 | typedef enum bool { FALSE = 0, TRUE = 1 } bool; | |
98 | # define ENUM_BOOL 1 | |
99 | # ifndef HAS_BOOL | |
100 | # define HAS_BOOL 1 | |
101 | # endif /* !HAS_BOOL */ | |
8f1f23e8 | 102 | #endif /* NeXT || __NeXT__ */ |
641d3f0b | 103 | |
5d94fbed | 104 | #ifndef HAS_BOOL |
61bb5906 | 105 | # if defined(UTS) || defined(VMS) |
5d94fbed AD |
106 | # define bool int |
107 | # else | |
108 | # define bool char | |
109 | # endif | |
c1d22f6b | 110 | # define HAS_BOOL 1 |
a687059c | 111 | #endif |
0d3e774c | 112 | |
f2338a2e DM |
113 | /* a simple (bool) cast may not do the right thing: if bool is defined |
114 | * as char for example, then the cast from int is implementation-defined | |
115 | */ | |
116 | ||
117 | #define cBOOL(cbool) ((bool)!!(cbool)) | |
118 | ||
46c6c7e2 | 119 | /* Try to figure out __func__ or __FUNCTION__ equivalent, if any. |
e352bcff JH |
120 | * XXX Should really be a Configure probe, with HAS__FUNCTION__ |
121 | * and FUNCTION__ as results. | |
122 | * XXX Similarly, a Configure probe for __FILE__ and __LINE__ is needed. */ | |
46c6c7e2 JH |
123 | #if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) || (defined(__SUNPRO_C)) /* C99 or close enough. */ |
124 | # define FUNCTION__ __func__ | |
125 | #else | |
126 | # if (defined(_MSC_VER) && _MSC_VER < 1300) || /* Pre-MSVC 7.0 has neither __func__ nor __FUNCTION and no good workarounds, either. */ \ | |
be73ebb0 | 127 | (defined(__DECC_VER)) /* Tru64 or VMS, and strict C89 being used, but not modern enough cc (in Tur64, -c99 not known, only -std1). */ |
46c6c7e2 JH |
128 | # define FUNCTION__ "" |
129 | # else | |
130 | # define FUNCTION__ __FUNCTION__ /* Common extension. */ | |
131 | # endif | |
132 | #endif | |
133 | ||
27d4fb96 | 134 | /* XXX A note on the perl source internal type system. The |
135 | original intent was that I32 be *exactly* 32 bits. | |
136 | ||
137 | Currently, we only guarantee that I32 is *at least* 32 bits. | |
138 | Specifically, if int is 64 bits, then so is I32. (This is the case | |
139 | for the Cray.) This has the advantage of meshing nicely with | |
140 | standard library calls (where we pass an I32 and the library is | |
141 | expecting an int), but the disadvantage that an I32 is not 32 bits. | |
142 | Andy Dougherty August 1996 | |
24fef2a7 | 143 | |
dc45a647 MB |
144 | There is no guarantee that there is *any* integral type with |
145 | exactly 32 bits. It is perfectly legal for a system to have | |
146 | sizeof(short) == sizeof(int) == sizeof(long) == 8. | |
693762b4 | 147 | |
dc45a647 MB |
148 | Similarly, there is no guarantee that I16 and U16 have exactly 16 |
149 | bits. | |
693762b4 | 150 | |
8e84507e NIS |
151 | For dealing with issues that may arise from various 32/64-bit |
152 | systems, we will ask Configure to check out | |
8175356b | 153 | |
3609ea0d JH |
154 | SHORTSIZE == sizeof(short) |
155 | INTSIZE == sizeof(int) | |
156 | LONGSIZE == sizeof(long) | |
dc45a647 | 157 | LONGLONGSIZE == sizeof(long long) (if HAS_LONG_LONG) |
3609ea0d | 158 | PTRSIZE == sizeof(void *) |
dc45a647 MB |
159 | DOUBLESIZE == sizeof(double) |
160 | LONG_DOUBLESIZE == sizeof(long double) (if HAS_LONG_DOUBLE). | |
8175356b | 161 | |
27d4fb96 | 162 | */ |
163 | ||
69512466 JH |
164 | #ifdef I_INTTYPES /* e.g. Linux has int64_t without <inttypes.h> */ |
165 | # include <inttypes.h> | |
dd0eed91 JH |
166 | # ifdef INT32_MIN_BROKEN |
167 | # undef INT32_MIN | |
168 | # define INT32_MIN (-2147483647-1) | |
169 | # endif | |
170 | # ifdef INT64_MIN_BROKEN | |
171 | # undef INT64_MIN | |
172 | # define INT64_MIN (-9223372036854775807LL-1) | |
173 | # endif | |
69512466 JH |
174 | #endif |
175 | ||
8175356b JH |
176 | typedef I8TYPE I8; |
177 | typedef U8TYPE U8; | |
178 | typedef I16TYPE I16; | |
179 | typedef U16TYPE U16; | |
180 | typedef I32TYPE I32; | |
181 | typedef U32TYPE U32; | |
6b8eaf93 JH |
182 | #ifdef PERL_CORE |
183 | # ifdef HAS_QUAD | |
8175356b JH |
184 | typedef I64TYPE I64; |
185 | typedef U64TYPE U64; | |
6b8eaf93 JH |
186 | # endif |
187 | #endif /* PERL_CORE */ | |
8175356b | 188 | |
69512466 JH |
189 | #if defined(HAS_QUAD) && defined(USE_64_BIT_INT) |
190 | # ifndef UINT64_C /* usually from <inttypes.h> */ | |
191 | # if defined(HAS_LONG_LONG) && QUADKIND == QUAD_IS_LONG_LONG | |
192 | # define INT64_C(c) CAT2(c,LL) | |
193 | # define UINT64_C(c) CAT2(c,ULL) | |
194 | # else | |
195 | # if LONGSIZE == 8 && QUADKIND == QUAD_IS_LONG | |
196 | # define INT64_C(c) CAT2(c,L) | |
197 | # define UINT64_C(c) CAT2(c,UL) | |
198 | # else | |
fa58a56f S |
199 | # if defined(_WIN64) && defined(_MSC_VER) |
200 | # define INT64_C(c) CAT2(c,I64) | |
201 | # define UINT64_C(c) CAT2(c,UI64) | |
202 | # else | |
203 | # define INT64_C(c) ((I64TYPE)(c)) | |
204 | # define UINT64_C(c) ((U64TYPE)(c)) | |
205 | # endif | |
69512466 JH |
206 | # endif |
207 | # endif | |
e8c95190 JH |
208 | # endif |
209 | #endif | |
dc750f2e | 210 | |
5086dff9 MB |
211 | /* HMB H.Merijn Brand - a placeholder for preparing Configure patches: |
212 | * | |
213 | * USE_DTRACE HAS_PSEUDOFORK HAS_TIMEGM LOCALTIME_R_NEEDS_TZSET | |
214 | * GMTIME_MAX GMTIME_MIN LOCALTIME_MAX LOCALTIME_MIN | |
215 | * HAS_CTIME64 HAS_LOCALTIME64 HAS_GMTIME64 HAS_DIFFTIME64 | |
216 | * HAS_MKTIME64 HAS_ASCTIME64 HAS_GETADDRINFO HAS_GETNAMEINFO | |
7636ea95 | 217 | * HAS_INETNTOP HAS_INETPTON CHARBITS HAS_PRCTL |
5086dff9 MB |
218 | * Not (yet) used at top level, but mention them for metaconfig |
219 | */ | |
e8c95190 | 220 | |
a22e52b9 JH |
221 | /* Mention I8SIZE, U8SIZE, I16SIZE, U16SIZE, I32SIZE, U32SIZE, |
222 | I64SIZE, and U64SIZE here so that metaconfig pulls them in. */ | |
223 | ||
d8668976 | 224 | #if defined(UINT8_MAX) && defined(INT16_MAX) && defined(INT32_MAX) |
5ff3f7a4 | 225 | |
5ff3f7a4 GS |
226 | /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type. |
227 | Please search CHAR_MAX in perl.h for further details. */ | |
228 | #define U8_MAX UINT8_MAX | |
229 | #define U8_MIN UINT8_MIN | |
230 | ||
5ff3f7a4 GS |
231 | #define I16_MAX INT16_MAX |
232 | #define I16_MIN INT16_MIN | |
233 | #define U16_MAX UINT16_MAX | |
234 | #define U16_MIN UINT16_MIN | |
235 | ||
5ff3f7a4 GS |
236 | #define I32_MAX INT32_MAX |
237 | #define I32_MIN INT32_MIN | |
0e983133 GS |
238 | #ifndef UINT32_MAX_BROKEN /* e.g. HP-UX with gcc messes this up */ |
239 | # define U32_MAX UINT32_MAX | |
240 | #else | |
241 | # define U32_MAX 4294967295U | |
242 | #endif | |
5ff3f7a4 GS |
243 | #define U32_MIN UINT32_MIN |
244 | ||
245 | #else | |
246 | ||
5c9fa16e KA |
247 | /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type. |
248 | Please search CHAR_MAX in perl.h for further details. */ | |
27d4fb96 | 249 | #define U8_MAX PERL_UCHAR_MAX |
250 | #define U8_MIN PERL_UCHAR_MIN | |
79072805 | 251 | |
27d4fb96 | 252 | #define I16_MAX PERL_SHORT_MAX |
253 | #define I16_MIN PERL_SHORT_MIN | |
254 | #define U16_MAX PERL_USHORT_MAX | |
255 | #define U16_MIN PERL_USHORT_MIN | |
79072805 | 256 | |
c4f23d77 | 257 | #if LONGSIZE > 4 |
27d4fb96 | 258 | # define I32_MAX PERL_INT_MAX |
259 | # define I32_MIN PERL_INT_MIN | |
260 | # define U32_MAX PERL_UINT_MAX | |
261 | # define U32_MIN PERL_UINT_MIN | |
79072805 | 262 | #else |
27d4fb96 | 263 | # define I32_MAX PERL_LONG_MAX |
264 | # define I32_MIN PERL_LONG_MIN | |
265 | # define U32_MAX PERL_ULONG_MAX | |
266 | # define U32_MIN PERL_ULONG_MIN | |
79072805 LW |
267 | #endif |
268 | ||
5ff3f7a4 GS |
269 | #endif |
270 | ||
58a9a5d5 | 271 | /* log(2) is pretty close to 0.30103, just in case anyone is grepping for it */ |
fc36a67e | 272 | #define BIT_DIGITS(N) (((N)*146)/485 + 1) /* log2(10) =~ 146/485 */ |
273 | #define TYPE_DIGITS(T) BIT_DIGITS(sizeof(T) * 8) | |
274 | #define TYPE_CHARS(T) (TYPE_DIGITS(T) + 2) /* sign, NUL */ | |
275 | ||
ff68c719 | 276 | #define Ctl(ch) ((ch) & 037) |
8d063cd8 | 277 | |
3fe05580 MHM |
278 | /* |
279 | =head1 SV-Body Allocation | |
280 | ||
281 | =for apidoc Ama|SV*|newSVpvs|const char* s | |
282 | Like C<newSVpvn>, but takes a literal string instead of a string/length pair. | |
283 | ||
84bafc02 NC |
284 | =for apidoc Ama|SV*|newSVpvs_flags|const char* s|U32 flags |
285 | Like C<newSVpvn_flags>, but takes a literal string instead of a string/length | |
286 | pair. | |
287 | ||
3fe05580 MHM |
288 | =for apidoc Ama|SV*|newSVpvs_share|const char* s |
289 | Like C<newSVpvn_share>, but takes a literal string instead of a string/length | |
290 | pair and omits the hash parameter. | |
291 | ||
7b4e7469 | 292 | =for apidoc Am|void|sv_catpvs|SV* sv|const char* s |
3fe05580 MHM |
293 | Like C<sv_catpvn>, but takes a literal string instead of a string/length pair. |
294 | ||
7b4e7469 | 295 | =for apidoc Am|void|sv_setpvs|SV* sv|const char* s |
3fe05580 MHM |
296 | Like C<sv_setpvn>, but takes a literal string instead of a string/length pair. |
297 | ||
298 | =head1 Memory Management | |
299 | ||
300 | =for apidoc Ama|char*|savepvs|const char* s | |
301 | Like C<savepvn>, but takes a literal string instead of a string/length pair. | |
302 | ||
303 | =head1 GV Functions | |
304 | ||
305 | =for apidoc Am|HV*|gv_stashpvs|const char* name|I32 create | |
306 | Like C<gv_stashpvn>, but takes a literal string instead of a string/length pair. | |
307 | ||
308 | =head1 Hash Manipulation Functions | |
309 | ||
310 | =for apidoc Am|SV**|hv_fetchs|HV* tb|const char* key|I32 lval | |
311 | Like C<hv_fetch>, but takes a literal string instead of a string/length pair. | |
312 | ||
313 | =for apidoc Am|SV**|hv_stores|HV* tb|const char* key|NULLOK SV* val | |
314 | Like C<hv_store>, but takes a literal string instead of a string/length pair | |
315 | and omits the hash parameter. | |
316 | ||
510966aa Z |
317 | =head1 Lexer interface |
318 | ||
319 | =for apidoc Amx|void|lex_stuff_pvs|const char *pv|U32 flags | |
320 | ||
321 | Like L</lex_stuff_pvn>, but takes a literal string instead of a | |
322 | string/length pair. | |
323 | ||
3fe05580 MHM |
324 | =cut |
325 | */ | |
326 | ||
65177ae0 | 327 | /* concatenating with "" ensures that only literal strings are accepted as argument */ |
26bfd924 | 328 | #define STR_WITH_LEN(s) ("" s ""), (sizeof(s)-1) |
2efa8cc7 | 329 | |
ba3a79e7 GA |
330 | /* note that STR_WITH_LEN() can't be used as argument to macros or functions that |
331 | * under some configurations might be macros, which means that it requires the full | |
332 | * Perl_xxx(aTHX_ ...) form for any API calls where it's used. | |
333 | */ | |
334 | ||
335 | /* STR_WITH_LEN() shortcuts */ | |
336 | #define newSVpvs(str) Perl_newSVpvn(aTHX_ STR_WITH_LEN(str)) | |
84bafc02 NC |
337 | #define newSVpvs_flags(str,flags) \ |
338 | Perl_newSVpvn_flags(aTHX_ STR_WITH_LEN(str), flags) | |
ba3a79e7 GA |
339 | #define newSVpvs_share(str) Perl_newSVpvn_share(aTHX_ STR_WITH_LEN(str), 0) |
340 | #define sv_catpvs(sv, str) Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC) | |
3fe05580 | 341 | #define sv_setpvs(sv, str) Perl_sv_setpvn(aTHX_ sv, STR_WITH_LEN(str)) |
ba3a79e7 GA |
342 | #define savepvs(str) Perl_savepvn(aTHX_ STR_WITH_LEN(str)) |
343 | #define gv_stashpvs(str, create) Perl_gv_stashpvn(aTHX_ STR_WITH_LEN(str), create) | |
5c1737d1 | 344 | #define gv_fetchpvs(namebeg, add, sv_type) Perl_gv_fetchpvn_flags(aTHX_ STR_WITH_LEN(namebeg), add, sv_type) |
a038e571 NC |
345 | #define hv_fetchs(hv,key,lval) \ |
346 | ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \ | |
347 | (lval) ? (HV_FETCH_JUST_SV | HV_FETCH_LVALUE) \ | |
348 | : HV_FETCH_JUST_SV, NULL, 0)) | |
349 | ||
350 | #define hv_stores(hv,key,val) \ | |
351 | ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \ | |
352 | (HV_FETCH_ISSTORE|HV_FETCH_JUST_SV), (val), 0)) | |
ba3a79e7 | 353 | |
510966aa Z |
354 | #define lex_stuff_pvs(pv,flags) Perl_lex_stuff_pvn(aTHX_ STR_WITH_LEN(pv), flags) |
355 | ||
b96d8cd9 NC |
356 | #define get_cvs(str, flags) \ |
357 | Perl_get_cvn_flags(aTHX_ STR_WITH_LEN(str), (flags)) | |
5c1737d1 | 358 | |
954c1994 | 359 | /* |
ccfc67b7 JH |
360 | =head1 Miscellaneous Functions |
361 | ||
954c1994 GS |
362 | =for apidoc Am|bool|strNE|char* s1|char* s2 |
363 | Test two strings to see if they are different. Returns true or | |
364 | false. | |
365 | ||
366 | =for apidoc Am|bool|strEQ|char* s1|char* s2 | |
367 | Test two strings to see if they are equal. Returns true or false. | |
368 | ||
369 | =for apidoc Am|bool|strLT|char* s1|char* s2 | |
370 | Test two strings to see if the first, C<s1>, is less than the second, | |
371 | C<s2>. Returns true or false. | |
372 | ||
373 | =for apidoc Am|bool|strLE|char* s1|char* s2 | |
374 | Test two strings to see if the first, C<s1>, is less than or equal to the | |
375 | second, C<s2>. Returns true or false. | |
376 | ||
377 | =for apidoc Am|bool|strGT|char* s1|char* s2 | |
378 | Test two strings to see if the first, C<s1>, is greater than the second, | |
379 | C<s2>. Returns true or false. | |
380 | ||
381 | =for apidoc Am|bool|strGE|char* s1|char* s2 | |
382 | Test two strings to see if the first, C<s1>, is greater than or equal to | |
383 | the second, C<s2>. Returns true or false. | |
384 | ||
385 | =for apidoc Am|bool|strnNE|char* s1|char* s2|STRLEN len | |
386 | Test two strings to see if they are different. The C<len> parameter | |
387 | indicates the number of bytes to compare. Returns true or false. (A | |
388 | wrapper for C<strncmp>). | |
389 | ||
390 | =for apidoc Am|bool|strnEQ|char* s1|char* s2|STRLEN len | |
391 | Test two strings to see if they are equal. The C<len> parameter indicates | |
392 | the number of bytes to compare. Returns true or false. (A wrapper for | |
393 | C<strncmp>). | |
394 | ||
395 | =cut | |
396 | */ | |
397 | ||
8d063cd8 LW |
398 | #define strNE(s1,s2) (strcmp(s1,s2)) |
399 | #define strEQ(s1,s2) (!strcmp(s1,s2)) | |
400 | #define strLT(s1,s2) (strcmp(s1,s2) < 0) | |
401 | #define strLE(s1,s2) (strcmp(s1,s2) <= 0) | |
402 | #define strGT(s1,s2) (strcmp(s1,s2) > 0) | |
403 | #define strGE(s1,s2) (strcmp(s1,s2) >= 0) | |
404 | #define strnNE(s1,s2,l) (strncmp(s1,s2,l)) | |
405 | #define strnEQ(s1,s2,l) (!strncmp(s1,s2,l)) | |
378cc40b | 406 | |
36477c24 | 407 | #ifdef HAS_MEMCMP |
408 | # define memNE(s1,s2,l) (memcmp(s1,s2,l)) | |
409 | # define memEQ(s1,s2,l) (!memcmp(s1,s2,l)) | |
410 | #else | |
411 | # define memNE(s1,s2,l) (bcmp(s1,s2,l)) | |
412 | # define memEQ(s1,s2,l) (!bcmp(s1,s2,l)) | |
413 | #endif | |
414 | ||
568a785a | 415 | #define memEQs(s1, l, s2) \ |
8e6497b6 | 416 | (sizeof(s2)-1 == l && memEQ(s1, ("" s2 ""), (sizeof(s2)-1))) |
568a785a NC |
417 | #define memNEs(s1, l, s2) !memEQs(s1, l, s2) |
418 | ||
bbce6d69 | 419 | /* |
420 | * Character classes. | |
421 | * | |
422 | * Unfortunately, the introduction of locales means that we | |
423 | * can't trust isupper(), etc. to tell the truth. And when | |
424 | * it comes to /\w+/ with tainting enabled, we *must* be able | |
425 | * to trust our character classes. | |
426 | * | |
427 | * Therefore, the default tests in the text of Perl will be | |
428 | * independent of locale. Any code that wants to depend on | |
429 | * the current locale will use the tests that begin with "lc". | |
430 | */ | |
431 | ||
2304df62 AD |
432 | #ifdef HAS_SETLOCALE /* XXX Is there a better test for this? */ |
433 | # ifndef CTYPE256 | |
434 | # define CTYPE256 | |
435 | # endif | |
436 | #endif | |
437 | ||
954c1994 | 438 | /* |
ccfc67b7 JH |
439 | |
440 | =head1 Character classes | |
06dc89af KW |
441 | The functions in this section operate using the character set of the platform |
442 | Perl is running on, and are unaffected by locale. For ASCII platforms, they | |
443 | will all return false for characters outside the ASCII range. For EBCDIC | |
444 | platforms, they use the code page of the platform. The code pages that Perl | |
445 | knows about all have 8-bit characters, so most of these functions will return | |
446 | true for more characters than on ASCII platforms. | |
ccfc67b7 | 447 | |
954c1994 | 448 | =for apidoc Am|bool|isALPHA|char ch |
2787a470 | 449 | Returns a boolean indicating whether the specified character is an |
06dc89af | 450 | alphabetic character in the platform's native character set. |
954c1994 | 451 | |
954c1994 | 452 | =for apidoc Am|bool|isDIGIT|char ch |
2787a470 | 453 | Returns a boolean indicating whether the specified character is a |
06dc89af | 454 | digit in the platform's native character set. |
954c1994 | 455 | |
0c82b6df | 456 | =for apidoc Am|bool|isLOWER|char ch |
2787a470 | 457 | Returns a boolean indicating whether the specified character is a |
0c82b6df KW |
458 | lowercase character in the platform's native character set. |
459 | ||
c99e91e9 | 460 | =for apidoc Am|bool|isOCTAL|char ch |
2787a470 | 461 | Returns a boolean indicating whether the specified character is an |
06dc89af | 462 | octal digit, [0-7] in the platform's native character set. |
c99e91e9 | 463 | |
0c82b6df | 464 | =for apidoc Am|bool|isSPACE|char ch |
2787a470 | 465 | Returns a boolean indicating whether the specified character is a |
0c82b6df KW |
466 | whitespace character in the platform's native character set. |
467 | ||
954c1994 | 468 | =for apidoc Am|bool|isUPPER|char ch |
2787a470 | 469 | Returns a boolean indicating whether the specified character is an |
06dc89af | 470 | uppercase character in the platform's native character set. |
954c1994 | 471 | |
8eea39dd KW |
472 | =for apidoc Am|bool|isWORDCHAR|char ch |
473 | Returns a boolean indicating whether the specified character is a | |
474 | character that is any of: alphabetic, numeric, or an underscore. This is the | |
475 | same as what C<\w> matches in a regular expression. | |
476 | C<isALNUM()> is a synonym provided for backward compatibility. Note that it | |
477 | does not have the standard C language meaning of alphanumeric, since it matches | |
478 | an underscore and the standard meaning does not. | |
479 | ||
e7c1e6c1 KW |
480 | =head1 Character case changing |
481 | ||
954c1994 | 482 | =for apidoc Am|char|toUPPER|char ch |
06dc89af KW |
483 | Converts the specified character to uppercase in the platform's native |
484 | character set, if possible; otherwise returns the input character itself. | |
954c1994 GS |
485 | |
486 | =for apidoc Am|char|toLOWER|char ch | |
06dc89af KW |
487 | Converts the specified character to lowercase in the platform's native |
488 | character set, if possible; otherwise returns the input character itself. | |
954c1994 GS |
489 | |
490 | =cut | |
353c9b6f KW |
491 | |
492 | NOTE: Since some of these are macros, there is no check in those that the | |
493 | parameter is a char or U8. This means that if called with a larger width | |
494 | parameter, casts can silently truncate and yield wrong results. | |
495 | ||
62fa66b6 KW |
496 | Also note that these macros are repeated in Devel::PPPort, so should also be |
497 | patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc | |
498 | ||
954c1994 GS |
499 | */ |
500 | ||
cf301eb7 KW |
501 | /* FITS_IN_8_BITS(c) returns true if c occupies no more than 8 bits. It is |
502 | * designed to be hopefully bomb-proof, making sure that no bits of | |
503 | * information are lost even on a 64-bit machine, but to get the compiler to | |
504 | * optimize it out if possible. This is because Configure makes sure that the | |
505 | * machine has an 8-bit byte, so if c is stored in a byte, the sizeof() | |
506 | * guarantees that this evaluates to a constant true at compile time. The use | |
507 | * of the mask instead of '< 256' keeps gcc from complaining that it is alway | |
d0dcc402 KW |
508 | * true, when c's storage class is a byte. Use U64TYPE because U64 is known |
509 | * only in the perl core, and this macro can be called from outside that */ | |
cf301eb7 | 510 | #ifdef HAS_QUAD |
d0dcc402 | 511 | # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U64TYPE)(c) & 0xFF) == (U64TYPE)(c))) |
cf301eb7 KW |
512 | #else |
513 | # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U32)(c) & 0xFF) == (U32)(c))) | |
514 | #endif | |
515 | ||
ec79cfbb KW |
516 | #define isASCII(c) (FITS_IN_8_BITS(c) ? NATIVE_TO_UNI((U8) c) <= 127 : 0) |
517 | #define isASCII_A(c) isASCII(c) | |
6515cb76 | 518 | |
25d1bb79 KW |
519 | /* include the Latin1 lookup table */ |
520 | #include "l1_char_class_tab.h" | |
521 | ||
ec79cfbb | 522 | /* ASCII range only */ |
25d1bb79 KW |
523 | #define isALNUMC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_A)) |
524 | #define isALPHA_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_A)) | |
525 | #define isBLANK_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_A)) | |
526 | #define isCNTRL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_A)) | |
527 | #define isDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_DIGIT_A)) | |
528 | #define isGRAPH_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_A)) | |
529 | #define isLOWER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_A)) | |
530 | #define isPRINT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_A)) | |
531 | #define isPSXSPC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_A)) | |
532 | #define isPUNCT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_A)) | |
533 | #define isSPACE_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_A)) | |
534 | #define isUPPER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_A)) | |
535 | #define isWORDCHAR_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_A)) | |
536 | #define isXDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_XDIGIT_A)) | |
ec79cfbb KW |
537 | |
538 | /* Latin1 definitions */ | |
cb233ae3 KW |
539 | /* ALPHAU includes Unicode semantics for latin1 characters. It has an extra |
540 | * >= AA test to speed up ASCII-only tests at the expense of the others */ | |
a9fccb9a KW |
541 | /* XXX decide whether to document the ALPHAU, ALNUMU and isSPACE_L1 functions. |
542 | * Most of these should be implemented as table lookup for speed */ | |
ec79cfbb KW |
543 | #define isALPHAU(c) (isALPHA_A(c) || (NATIVE_TO_UNI((U8) c) >= 0xAA \ |
544 | && ((NATIVE_TO_UNI((U8) c) >= 0xC0 \ | |
cb233ae3 KW |
545 | && NATIVE_TO_UNI((U8) c) != 0xD7 && NATIVE_TO_UNI((U8) c) != 0xF7) \ |
546 | || NATIVE_TO_UNI((U8) c) == 0xAA \ | |
547 | || NATIVE_TO_UNI((U8) c) == 0xB5 \ | |
548 | || NATIVE_TO_UNI((U8) c) == 0xBA))) | |
ec79cfbb KW |
549 | #define isSPACE_L1(c) (isSPACE(c) \ |
550 | || (NATIVE_TO_UNI(c) == 0x85 || NATIVE_TO_UNI(c) == 0xA0)) | |
551 | #define isWORDCHAR_L1(c) (isDIGIT(c) || isALPHAU(c) || (c) == '_') | |
cb233ae3 | 552 | |
ec79cfbb KW |
553 | /* Same macro in non-EBCDIC and EBCDIC. Called macros may evaluate |
554 | * differently between the two */ | |
555 | #define isALNUM(c) isWORDCHAR(c) | |
556 | #define isALNUMU(c) isWORDCHAR_L1(c) | |
cb233ae3 | 557 | /* continuation character for legal NAME in \N{NAME} */ |
ec79cfbb | 558 | #define isCHARNAME_CONT(c) (isWORDCHAR_L1(c) || (c) == ' ' || (c) == '-' || (c) == '(' || (c) == ')' || (c) == ':' || NATIVE_TO_UNI((U8) c) == 0xA0) |
7d79683e | 559 | #define isIDFIRST(c) (isALPHA(c) || (c) == '_') |
ec79cfbb KW |
560 | #define isOCTAL_A(c) ((c) >= '0' && (c) <= '7') |
561 | #define isOCTAL(c) isOCTAL_A(c) | |
49323284 | 562 | #define isWORDCHAR(c) (isALPHA(c) || isDIGIT(c) || (c) == '_') |
ec79cfbb | 563 | |
9d116dd7 | 564 | #ifdef EBCDIC |
fae4c845 | 565 | /* In EBCDIC we do not do locales: therefore can use native functions */ |
b8c5462f | 566 | # define isALNUMC(c) isalnum(c) |
271b5cb3 | 567 | # define isALPHA(c) isalpha(c) |
0bc70f7f | 568 | # define isBLANK(c) ((c) == ' ' || (c) == '\t' || NATIVE_TO_UNI(c) == 0xA0) |
b8c5462f | 569 | # define isCNTRL(c) iscntrl(c) |
9916188f | 570 | # define isDIGIT(c) isdigit(c) |
b8c5462f | 571 | # define isGRAPH(c) isgraph(c) |
7d79683e | 572 | # define isLOWER(c) islower(c) |
9d116dd7 | 573 | # define isPRINT(c) isprint(c) |
56e4b611 | 574 | # define isPSXSPC(c) isspace(c) |
b8c5462f | 575 | # define isPUNCT(c) ispunct(c) |
174b5e76 | 576 | # define isSPACE(c) (isPSXSPC(c) && (c) != '\v') |
7d79683e | 577 | # define isUPPER(c) isupper(c) |
b8c5462f | 578 | # define isXDIGIT(c) isxdigit(c) |
9d116dd7 | 579 | # define toLOWER(c) tolower(c) |
7d79683e | 580 | # define toUPPER(c) toupper(c) |
ec79cfbb KW |
581 | #else /* Not EBCDIC: ASCII-only matching */ |
582 | # define isALNUMC(c) isALNUMC_A(c) | |
271b5cb3 | 583 | # define isALPHA(c) isALPHA_A(c) |
ec79cfbb KW |
584 | # define isBLANK(c) isBLANK_A(c) |
585 | # define isCNTRL(c) isCNTRL_A(c) | |
586 | # define isDIGIT(c) isDIGIT_A(c) | |
587 | # define isGRAPH(c) isGRAPH_A(c) | |
588 | # define isLOWER(c) isLOWER_A(c) | |
589 | # define isPRINT(c) isPRINT_A(c) | |
590 | # define isPSXSPC(c) isPSXSPC_A(c) | |
591 | # define isPUNCT(c) isPUNCT_A(c) | |
592 | # define isSPACE(c) isSPACE_A(c) | |
593 | # define isUPPER(c) isUPPER_A(c) | |
594 | # define isXDIGIT(c) isXDIGIT_A(c) | |
00f254e2 | 595 | |
12e948b5 | 596 | /* ASCII casing. */ |
9d116dd7 | 597 | # define toLOWER(c) (isUPPER(c) ? (c) + ('a' - 'A') : (c)) |
7d79683e | 598 | # define toUPPER(c) (isLOWER(c) ? (c) - ('a' - 'A') : (c)) |
9d116dd7 | 599 | #endif |
bbce6d69 | 600 | |
1a0901db KW |
601 | |
602 | /* Use table lookup for speed; return error character for input | |
603 | * out-of-range */ | |
604 | #define toLOWER_LATIN1(c) (FITS_IN_8_BITS(c) \ | |
605 | ? UNI_TO_NATIVE(PL_latin1_lc[ \ | |
606 | NATIVE_TO_UNI( (U8) (c)) ]) \ | |
607 | : UNICODE_REPLACEMENT) | |
608 | /* Modified uc. Is correct uc except for three non-ascii chars which are | |
609 | * all mapped to one of them, and these need special handling; error | |
610 | * character for input out-of-range */ | |
611 | #define toUPPER_LATIN1_MOD(c) (FITS_IN_8_BITS(c) \ | |
612 | ? UNI_TO_NATIVE(PL_mod_latin1_uc[ \ | |
613 | NATIVE_TO_UNI( (U8) (c)) ]) \ | |
614 | : UNICODE_REPLACEMENT) | |
615 | ||
bbce6d69 | 616 | #ifdef USE_NEXT_CTYPE |
617 | ||
618 | # define isALNUM_LC(c) \ | |
37bd1396 | 619 | (NXIsAlNum((unsigned int)(c)) || (char)(c) == '_') |
ff68c719 | 620 | # define isIDFIRST_LC(c) \ |
621 | (NXIsAlpha((unsigned int)(c)) || (char)(c) == '_') | |
622 | # define isALPHA_LC(c) NXIsAlpha((unsigned int)(c)) | |
623 | # define isSPACE_LC(c) NXIsSpace((unsigned int)(c)) | |
624 | # define isDIGIT_LC(c) NXIsDigit((unsigned int)(c)) | |
625 | # define isUPPER_LC(c) NXIsUpper((unsigned int)(c)) | |
626 | # define isLOWER_LC(c) NXIsLower((unsigned int)(c)) | |
37bd1396 | 627 | # define isALNUMC_LC(c) NXIsAlNum((unsigned int)(c)) |
b8c5462f JH |
628 | # define isCNTRL_LC(c) NXIsCntrl((unsigned int)(c)) |
629 | # define isGRAPH_LC(c) NXIsGraph((unsigned int)(c)) | |
ff68c719 | 630 | # define isPRINT_LC(c) NXIsPrint((unsigned int)(c)) |
b8c5462f | 631 | # define isPUNCT_LC(c) NXIsPunct((unsigned int)(c)) |
ff68c719 | 632 | # define toUPPER_LC(c) NXToUpper((unsigned int)(c)) |
633 | # define toLOWER_LC(c) NXToLower((unsigned int)(c)) | |
bbce6d69 | 634 | |
635 | #else /* !USE_NEXT_CTYPE */ | |
b8c5462f | 636 | |
bbce6d69 | 637 | # if defined(CTYPE256) || (!defined(isascii) && !defined(HAS_ISASCII)) |
638 | ||
b8c5462f | 639 | # define isALNUM_LC(c) (isalnum((unsigned char)(c)) || (char)(c) == '_') |
ff68c719 | 640 | # define isIDFIRST_LC(c) (isalpha((unsigned char)(c)) || (char)(c) == '_') |
bbce6d69 | 641 | # define isALPHA_LC(c) isalpha((unsigned char)(c)) |
642 | # define isSPACE_LC(c) isspace((unsigned char)(c)) | |
643 | # define isDIGIT_LC(c) isdigit((unsigned char)(c)) | |
644 | # define isUPPER_LC(c) isupper((unsigned char)(c)) | |
645 | # define isLOWER_LC(c) islower((unsigned char)(c)) | |
b8c5462f JH |
646 | # define isALNUMC_LC(c) isalnum((unsigned char)(c)) |
647 | # define isCNTRL_LC(c) iscntrl((unsigned char)(c)) | |
648 | # define isGRAPH_LC(c) isgraph((unsigned char)(c)) | |
bbce6d69 | 649 | # define isPRINT_LC(c) isprint((unsigned char)(c)) |
b8c5462f | 650 | # define isPUNCT_LC(c) ispunct((unsigned char)(c)) |
bbce6d69 | 651 | # define toUPPER_LC(c) toupper((unsigned char)(c)) |
652 | # define toLOWER_LC(c) tolower((unsigned char)(c)) | |
653 | ||
654 | # else | |
655 | ||
3609ea0d | 656 | # define isALNUM_LC(c) (isascii(c) && (isalnum(c) || (c) == '_')) |
bbce6d69 | 657 | # define isIDFIRST_LC(c) (isascii(c) && (isalpha(c) || (c) == '_')) |
658 | # define isALPHA_LC(c) (isascii(c) && isalpha(c)) | |
659 | # define isSPACE_LC(c) (isascii(c) && isspace(c)) | |
660 | # define isDIGIT_LC(c) (isascii(c) && isdigit(c)) | |
661 | # define isUPPER_LC(c) (isascii(c) && isupper(c)) | |
662 | # define isLOWER_LC(c) (isascii(c) && islower(c)) | |
b8c5462f JH |
663 | # define isALNUMC_LC(c) (isascii(c) && isalnum(c)) |
664 | # define isCNTRL_LC(c) (isascii(c) && iscntrl(c)) | |
665 | # define isGRAPH_LC(c) (isascii(c) && isgraph(c)) | |
bbce6d69 | 666 | # define isPRINT_LC(c) (isascii(c) && isprint(c)) |
b8c5462f | 667 | # define isPUNCT_LC(c) (isascii(c) && ispunct(c)) |
bbce6d69 | 668 | # define toUPPER_LC(c) toupper(c) |
669 | # define toLOWER_LC(c) tolower(c) | |
670 | ||
671 | # endif | |
a0d0e21e | 672 | #endif /* USE_NEXT_CTYPE */ |
55204971 | 673 | |
aaa51d5e JF |
674 | #define isPSXSPC_LC(c) (isSPACE_LC(c) || (c) == '\v') |
675 | #define isBLANK_LC(c) isBLANK(c) /* could be wrong */ | |
676 | ||
a0ed51b3 LW |
677 | #define isALNUM_uni(c) is_uni_alnum(c) |
678 | #define isIDFIRST_uni(c) is_uni_idfirst(c) | |
679 | #define isALPHA_uni(c) is_uni_alpha(c) | |
680 | #define isSPACE_uni(c) is_uni_space(c) | |
681 | #define isDIGIT_uni(c) is_uni_digit(c) | |
682 | #define isUPPER_uni(c) is_uni_upper(c) | |
683 | #define isLOWER_uni(c) is_uni_lower(c) | |
b8c5462f JH |
684 | #define isASCII_uni(c) is_uni_ascii(c) |
685 | #define isCNTRL_uni(c) is_uni_cntrl(c) | |
686 | #define isGRAPH_uni(c) is_uni_graph(c) | |
a0ed51b3 | 687 | #define isPRINT_uni(c) is_uni_print(c) |
b8c5462f JH |
688 | #define isPUNCT_uni(c) is_uni_punct(c) |
689 | #define isXDIGIT_uni(c) is_uni_xdigit(c) | |
a2a2844f JH |
690 | #define toUPPER_uni(c,s,l) to_uni_upper(c,s,l) |
691 | #define toTITLE_uni(c,s,l) to_uni_title(c,s,l) | |
692 | #define toLOWER_uni(c,s,l) to_uni_lower(c,s,l) | |
b8d68ded | 693 | #define toFOLD_uni(c,s,l) to_uni_fold(c,s,l) |
a0ed51b3 | 694 | |
aaa51d5e JF |
695 | #define isPSXSPC_uni(c) (isSPACE_uni(c) ||(c) == '\f') |
696 | #define isBLANK_uni(c) isBLANK(c) /* could be wrong */ | |
697 | ||
9041c2e3 NIS |
698 | #define isALNUM_LC_uvchr(c) (c < 256 ? isALNUM_LC(c) : is_uni_alnum_lc(c)) |
699 | #define isIDFIRST_LC_uvchr(c) (c < 256 ? isIDFIRST_LC(c) : is_uni_idfirst_lc(c)) | |
700 | #define isALPHA_LC_uvchr(c) (c < 256 ? isALPHA_LC(c) : is_uni_alpha_lc(c)) | |
701 | #define isSPACE_LC_uvchr(c) (c < 256 ? isSPACE_LC(c) : is_uni_space_lc(c)) | |
702 | #define isDIGIT_LC_uvchr(c) (c < 256 ? isDIGIT_LC(c) : is_uni_digit_lc(c)) | |
703 | #define isUPPER_LC_uvchr(c) (c < 256 ? isUPPER_LC(c) : is_uni_upper_lc(c)) | |
704 | #define isLOWER_LC_uvchr(c) (c < 256 ? isLOWER_LC(c) : is_uni_lower_lc(c)) | |
9041c2e3 NIS |
705 | #define isCNTRL_LC_uvchr(c) (c < 256 ? isCNTRL_LC(c) : is_uni_cntrl_lc(c)) |
706 | #define isGRAPH_LC_uvchr(c) (c < 256 ? isGRAPH_LC(c) : is_uni_graph_lc(c)) | |
707 | #define isPRINT_LC_uvchr(c) (c < 256 ? isPRINT_LC(c) : is_uni_print_lc(c)) | |
708 | #define isPUNCT_LC_uvchr(c) (c < 256 ? isPUNCT_LC(c) : is_uni_punct_lc(c)) | |
a0ed51b3 | 709 | |
aaa51d5e JF |
710 | #define isPSXSPC_LC_uni(c) (isSPACE_LC_uni(c) ||(c) == '\f') |
711 | #define isBLANK_LC_uni(c) isBLANK(c) /* could be wrong */ | |
712 | ||
a0ed51b3 | 713 | #define isALNUM_utf8(p) is_utf8_alnum(p) |
82686b01 JH |
714 | /* The ID_Start of Unicode is quite limiting: it assumes a L-class |
715 | * character (meaning that you cannot have, say, a CJK character). | |
716 | * Instead, let's allow ID_Continue but not digits. */ | |
717 | #define isIDFIRST_utf8(p) (is_utf8_idcont(p) && !is_utf8_digit(p)) | |
a0ed51b3 LW |
718 | #define isALPHA_utf8(p) is_utf8_alpha(p) |
719 | #define isSPACE_utf8(p) is_utf8_space(p) | |
720 | #define isDIGIT_utf8(p) is_utf8_digit(p) | |
721 | #define isUPPER_utf8(p) is_utf8_upper(p) | |
722 | #define isLOWER_utf8(p) is_utf8_lower(p) | |
b8c5462f JH |
723 | #define isASCII_utf8(p) is_utf8_ascii(p) |
724 | #define isCNTRL_utf8(p) is_utf8_cntrl(p) | |
725 | #define isGRAPH_utf8(p) is_utf8_graph(p) | |
a0ed51b3 | 726 | #define isPRINT_utf8(p) is_utf8_print(p) |
b8c5462f JH |
727 | #define isPUNCT_utf8(p) is_utf8_punct(p) |
728 | #define isXDIGIT_utf8(p) is_utf8_xdigit(p) | |
a2a2844f JH |
729 | #define toUPPER_utf8(p,s,l) to_utf8_upper(p,s,l) |
730 | #define toTITLE_utf8(p,s,l) to_utf8_title(p,s,l) | |
731 | #define toLOWER_utf8(p,s,l) to_utf8_lower(p,s,l) | |
a0ed51b3 | 732 | |
aaa51d5e JF |
733 | #define isPSXSPC_utf8(c) (isSPACE_utf8(c) ||(c) == '\f') |
734 | #define isBLANK_utf8(c) isBLANK(c) /* could be wrong */ | |
735 | ||
9041c2e3 NIS |
736 | #define isALNUM_LC_utf8(p) isALNUM_LC_uvchr(utf8_to_uvchr(p, 0)) |
737 | #define isIDFIRST_LC_utf8(p) isIDFIRST_LC_uvchr(utf8_to_uvchr(p, 0)) | |
738 | #define isALPHA_LC_utf8(p) isALPHA_LC_uvchr(utf8_to_uvchr(p, 0)) | |
739 | #define isSPACE_LC_utf8(p) isSPACE_LC_uvchr(utf8_to_uvchr(p, 0)) | |
740 | #define isDIGIT_LC_utf8(p) isDIGIT_LC_uvchr(utf8_to_uvchr(p, 0)) | |
741 | #define isUPPER_LC_utf8(p) isUPPER_LC_uvchr(utf8_to_uvchr(p, 0)) | |
742 | #define isLOWER_LC_utf8(p) isLOWER_LC_uvchr(utf8_to_uvchr(p, 0)) | |
743 | #define isALNUMC_LC_utf8(p) isALNUMC_LC_uvchr(utf8_to_uvchr(p, 0)) | |
744 | #define isCNTRL_LC_utf8(p) isCNTRL_LC_uvchr(utf8_to_uvchr(p, 0)) | |
745 | #define isGRAPH_LC_utf8(p) isGRAPH_LC_uvchr(utf8_to_uvchr(p, 0)) | |
746 | #define isPRINT_LC_utf8(p) isPRINT_LC_uvchr(utf8_to_uvchr(p, 0)) | |
747 | #define isPUNCT_LC_utf8(p) isPUNCT_LC_uvchr(utf8_to_uvchr(p, 0)) | |
a0ed51b3 | 748 | |
aaa51d5e JF |
749 | #define isPSXSPC_LC_utf8(c) (isSPACE_LC_utf8(c) ||(c) == '\f') |
750 | #define isBLANK_LC_utf8(c) isBLANK(c) /* could be wrong */ | |
751 | ||
6ae9f32a KW |
752 | /* This conversion works both ways, strangely enough. On EBCDIC platforms, |
753 | * CTRL-@ is 0, CTRL-A is 1, etc, just like on ASCII */ | |
754 | # define toCTRL(c) (toUPPER(NATIVE_TO_UNI(c)) ^ 64) | |
bbce6d69 | 755 | |
dea28490 JJ |
756 | /* Line numbers are unsigned, 32 bits. */ |
757 | typedef U32 line_t; | |
dea28490 | 758 | #define NOLINE ((line_t) 4294967295UL) |
378cc40b | 759 | |
91152fc1 DG |
760 | /* Helpful alias for version prescan */ |
761 | #define is_LAX_VERSION(a,b) \ | |
762 | (a != Perl_prescan_version(aTHX_ a, FALSE, b, NULL, NULL, NULL, NULL)) | |
763 | ||
764 | #define is_STRICT_VERSION(a,b) \ | |
765 | (a != Perl_prescan_version(aTHX_ a, TRUE, b, NULL, NULL, NULL, NULL)) | |
766 | ||
767 | #define BADVERSION(a,b,c) \ | |
768 | if (b) { \ | |
769 | *b = c; \ | |
770 | } \ | |
771 | return a; | |
8c52afec | 772 | |
8e84507e | 773 | /* |
ccfc67b7 JH |
774 | =head1 Memory Management |
775 | ||
a02a5408 | 776 | =for apidoc Am|void|Newx|void* ptr|int nitems|type |
954c1994 GS |
777 | The XSUB-writer's interface to the C C<malloc> function. |
778 | ||
c5008215 JC |
779 | In 5.9.3, Newx() and friends replace the older New() API, and drops |
780 | the first parameter, I<x>, a debug aid which allowed callers to identify | |
37b8b4c9 | 781 | themselves. This aid has been superseded by a new build option, |
c5008215 JC |
782 | PERL_MEM_LOG (see L<perlhack/PERL_MEM_LOG>). The older API is still |
783 | there for use in XS modules supporting older perls. | |
784 | ||
a02a5408 | 785 | =for apidoc Am|void|Newxc|void* ptr|int nitems|type|cast |
954c1994 | 786 | The XSUB-writer's interface to the C C<malloc> function, with |
c5008215 | 787 | cast. See also C<Newx>. |
954c1994 | 788 | |
a02a5408 | 789 | =for apidoc Am|void|Newxz|void* ptr|int nitems|type |
954c1994 | 790 | The XSUB-writer's interface to the C C<malloc> function. The allocated |
c5008215 | 791 | memory is zeroed with C<memzero>. See also C<Newx>. |
a02a5408 | 792 | |
954c1994 GS |
793 | =for apidoc Am|void|Renew|void* ptr|int nitems|type |
794 | The XSUB-writer's interface to the C C<realloc> function. | |
795 | ||
796 | =for apidoc Am|void|Renewc|void* ptr|int nitems|type|cast | |
797 | The XSUB-writer's interface to the C C<realloc> function, with | |
798 | cast. | |
799 | ||
49b8b560 | 800 | =for apidoc Am|void|Safefree|void* ptr |
954c1994 GS |
801 | The XSUB-writer's interface to the C C<free> function. |
802 | ||
803 | =for apidoc Am|void|Move|void* src|void* dest|int nitems|type | |
804 | The XSUB-writer's interface to the C C<memmove> function. The C<src> is the | |
805 | source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is | |
806 | the type. Can do overlapping moves. See also C<Copy>. | |
807 | ||
e90e2364 NC |
808 | =for apidoc Am|void *|MoveD|void* src|void* dest|int nitems|type |
809 | Like C<Move> but returns dest. Useful for encouraging compilers to tail-call | |
810 | optimise. | |
811 | ||
954c1994 GS |
812 | =for apidoc Am|void|Copy|void* src|void* dest|int nitems|type |
813 | The XSUB-writer's interface to the C C<memcpy> function. The C<src> is the | |
814 | source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is | |
815 | the type. May fail on overlapping copies. See also C<Move>. | |
816 | ||
e90e2364 NC |
817 | =for apidoc Am|void *|CopyD|void* src|void* dest|int nitems|type |
818 | ||
819 | Like C<Copy> but returns dest. Useful for encouraging compilers to tail-call | |
820 | optimise. | |
821 | ||
954c1994 GS |
822 | =for apidoc Am|void|Zero|void* dest|int nitems|type |
823 | ||
824 | The XSUB-writer's interface to the C C<memzero> function. The C<dest> is the | |
825 | destination, C<nitems> is the number of items, and C<type> is the type. | |
826 | ||
e90e2364 NC |
827 | =for apidoc Am|void *|ZeroD|void* dest|int nitems|type |
828 | ||
829 | Like C<Zero> but returns dest. Useful for encouraging compilers to tail-call | |
830 | optimise. | |
831 | ||
954c1994 | 832 | =for apidoc Am|void|StructCopy|type src|type dest|type |
4375e838 | 833 | This is an architecture-independent macro to copy one structure to another. |
954c1994 | 834 | |
7e337ee0 JH |
835 | =for apidoc Am|void|PoisonWith|void* dest|int nitems|type|U8 byte |
836 | ||
837 | Fill up memory with a byte pattern (a byte repeated over and over | |
838 | again) that hopefully catches attempts to access uninitialized memory. | |
839 | ||
840 | =for apidoc Am|void|PoisonNew|void* dest|int nitems|type | |
841 | ||
842 | PoisonWith(0xAB) for catching access to allocated but uninitialized memory. | |
843 | ||
1c12ffb4 | 844 | =for apidoc Am|void|PoisonFree|void* dest|int nitems|type |
7e337ee0 JH |
845 | |
846 | PoisonWith(0xEF) for catching access to freed memory. | |
847 | ||
9965345d JH |
848 | =for apidoc Am|void|Poison|void* dest|int nitems|type |
849 | ||
7e337ee0 | 850 | PoisonWith(0xEF) for catching access to freed memory. |
9965345d JH |
851 | |
852 | =cut */ | |
954c1994 | 853 | |
561b68a9 SH |
854 | /* Maintained for backwards-compatibility only. Use newSV() instead. */ |
855 | #ifndef PERL_CORE | |
ff06c60c | 856 | #define NEWSV(x,len) newSV(len) |
561b68a9 | 857 | #endif |
ff06c60c | 858 | |
19a94d75 JH |
859 | #define MEM_SIZE_MAX ((MEM_SIZE)~0) |
860 | ||
2720f793 JH |
861 | /* The +0.0 in MEM_WRAP_CHECK_ is an attempt to foil |
862 | * overly eager compilers that will bleat about e.g. | |
863 | * (U16)n > (size_t)~0/sizeof(U16) always being false. */ | |
27d5b266 | 864 | #ifdef PERL_MALLOC_WRAP |
fabdb6c0 | 865 | #define MEM_WRAP_CHECK(n,t) MEM_WRAP_CHECK_1(n,t,PL_memory_wrap) |
27d5b266 | 866 | #define MEM_WRAP_CHECK_1(n,t,a) \ |
f1f66076 | 867 | (void)(sizeof(t) > 1 && ((MEM_SIZE)(n)+0.0) > MEM_SIZE_MAX/sizeof(t) && (Perl_croak_nocontext("%s",(a)),0)) |
8b44ba4c | 868 | #define MEM_WRAP_CHECK_(n,t) MEM_WRAP_CHECK(n,t), |
27d5b266 | 869 | |
f1f66076 | 870 | #define PERL_STRLEN_ROUNDUP(n) ((void)(((n) > MEM_SIZE_MAX - 2 * PERL_STRLEN_ROUNDUP_QUANTUM) ? (Perl_croak_nocontext("%s",PL_memory_wrap),0):0),((n-1+PERL_STRLEN_ROUNDUP_QUANTUM)&~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM-1))) |
1936d2a7 | 871 | |
27d5b266 JH |
872 | #else |
873 | ||
410319be NC |
874 | #define MEM_WRAP_CHECK(n,t) |
875 | #define MEM_WRAP_CHECK_1(n,t,a) | |
876 | #define MEM_WRAP_CHECK_2(n,t,a,b) | |
8b44ba4c NC |
877 | #define MEM_WRAP_CHECK_(n,t) |
878 | ||
6fff79ce | 879 | #define PERL_STRLEN_ROUNDUP(n) (((n-1+PERL_STRLEN_ROUNDUP_QUANTUM)&~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM-1))) |
27d5b266 | 880 | |
1936d2a7 | 881 | #endif |
8b44ba4c | 882 | |
fe4f188c | 883 | #ifdef PERL_MEM_LOG |
46c6c7e2 | 884 | /* |
9f653bb5 | 885 | * If PERL_MEM_LOG is defined, all Newx()s, Renew()s, and Safefree()s |
46c6c7e2 JH |
886 | * go through functions, which are handy for debugging breakpoints, but |
887 | * which more importantly get the immediate calling environment (file and | |
e352bcff JH |
888 | * line number, and C function name if available) passed in. This info can |
889 | * then be used for logging the calls, for which one gets a sample | |
73d1d973 | 890 | * implementation unless -DPERL_MEM_LOG_NOIMPL is also defined. |
3609ea0d | 891 | * |
46c6c7e2 | 892 | * Known problems: |
94e892a6 | 893 | * - not all memory allocs get logged, only those |
46c6c7e2 | 894 | * that go through Newx() and derivatives (while all |
94e892a6 | 895 | * Safefrees do get logged) |
46c6c7e2 JH |
896 | * - __FILE__ and __LINE__ do not work everywhere |
897 | * - __func__ or __FUNCTION__ even less so | |
898 | * - I think more goes on after the perlio frees but | |
899 | * the thing is that STDERR gets closed (as do all | |
900 | * the file descriptors) | |
901 | * - no deeper calling stack than the caller of the Newx() | |
902 | * or the kind, but do I look like a C reflection/introspection | |
903 | * utility to you? | |
904 | * - the function prototypes for the logging functions | |
905 | * probably should maybe be somewhere else than handy.h | |
906 | * - one could consider inlining (macrofying) the logging | |
907 | * for speed, but I am too lazy | |
908 | * - one could imagine recording the allocations in a hash, | |
909 | * (keyed by the allocation address?), and maintain that | |
910 | * through reallocs and frees, but how to do that without | |
911 | * any News() happening...? | |
73d1d973 | 912 | * - lots of -Ddefines to get useful/controllable output |
b953482e | 913 | * - lots of ENV reads |
46c6c7e2 JH |
914 | */ |
915 | ||
12754f92 | 916 | PERL_EXPORT_C Malloc_t Perl_mem_log_alloc(const UV n, const UV typesize, const char *type_name, Malloc_t newalloc, const char *filename, const int linenumber, const char *funcname); |
46c6c7e2 | 917 | |
12754f92 | 918 | PERL_EXPORT_C Malloc_t Perl_mem_log_realloc(const UV n, const UV typesize, const char *type_name, Malloc_t oldalloc, Malloc_t newalloc, const char *filename, const int linenumber, const char *funcname); |
46c6c7e2 | 919 | |
12754f92 | 920 | PERL_EXPORT_C Malloc_t Perl_mem_log_free(Malloc_t oldalloc, const char *filename, const int linenumber, const char *funcname); |
46c6c7e2 | 921 | |
0b0ab801 | 922 | # ifdef PERL_CORE |
73d1d973 | 923 | # ifndef PERL_MEM_LOG_NOIMPL |
0b0ab801 MHM |
924 | enum mem_log_type { |
925 | MLT_ALLOC, | |
926 | MLT_REALLOC, | |
d7a2c63c MHM |
927 | MLT_FREE, |
928 | MLT_NEW_SV, | |
929 | MLT_DEL_SV | |
0b0ab801 MHM |
930 | }; |
931 | # endif | |
12754f92 | 932 | # if defined(PERL_IN_SV_C) /* those are only used in sv.c */ |
d7a2c63c MHM |
933 | void Perl_mem_log_new_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname); |
934 | void Perl_mem_log_del_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname); | |
12754f92 | 935 | # endif |
0b0ab801 MHM |
936 | # endif |
937 | ||
fe4f188c JH |
938 | #endif |
939 | ||
940 | #ifdef PERL_MEM_LOG | |
d1401ee9 MHM |
941 | #define MEM_LOG_ALLOC(n,t,a) Perl_mem_log_alloc(n,sizeof(t),STRINGIFY(t),a,__FILE__,__LINE__,FUNCTION__) |
942 | #define MEM_LOG_REALLOC(n,t,v,a) Perl_mem_log_realloc(n,sizeof(t),STRINGIFY(t),v,a,__FILE__,__LINE__,FUNCTION__) | |
46c6c7e2 | 943 | #define MEM_LOG_FREE(a) Perl_mem_log_free(a,__FILE__,__LINE__,FUNCTION__) |
fe4f188c JH |
944 | #endif |
945 | ||
946 | #ifndef MEM_LOG_ALLOC | |
947 | #define MEM_LOG_ALLOC(n,t,a) (a) | |
948 | #endif | |
949 | #ifndef MEM_LOG_REALLOC | |
950 | #define MEM_LOG_REALLOC(n,t,v,a) (a) | |
951 | #endif | |
952 | #ifndef MEM_LOG_FREE | |
953 | #define MEM_LOG_FREE(a) (a) | |
954 | #endif | |
955 | ||
d1401ee9 MHM |
956 | #define Newx(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t)))))) |
957 | #define Newxc(v,n,t,c) (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t)))))) | |
958 | #define Newxz(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safecalloc((n),sizeof(t))))) | |
a6f6820f NC |
959 | |
960 | #ifndef PERL_CORE | |
a02a5408 JC |
961 | /* pre 5.9.x compatibility */ |
962 | #define New(x,v,n,t) Newx(v,n,t) | |
963 | #define Newc(x,v,n,t,c) Newxc(v,n,t,c) | |
4541904d | 964 | #define Newz(x,v,n,t) Newxz(v,n,t) |
a6f6820f | 965 | #endif |
a02a5408 | 966 | |
ff68c719 | 967 | #define Renew(v,n,t) \ |
d1401ee9 | 968 | (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t)))))) |
ff68c719 | 969 | #define Renewc(v,n,t,c) \ |
d1401ee9 | 970 | (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t)))))) |
94010e71 NC |
971 | |
972 | #ifdef PERL_POISON | |
973 | #define Safefree(d) \ | |
06c0cc96 | 974 | ((d) ? (void)(safefree(MEM_LOG_FREE((Malloc_t)(d))), Poison(&(d), 1, Malloc_t)) : (void) 0) |
94010e71 | 975 | #else |
fe4f188c | 976 | #define Safefree(d) safefree(MEM_LOG_FREE((Malloc_t)(d))) |
94010e71 | 977 | #endif |
55497cff | 978 | |
8b44ba4c NC |
979 | #define Move(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memmove((char*)(d),(const char*)(s), (n) * sizeof(t))) |
980 | #define Copy(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memcpy((char*)(d),(const char*)(s), (n) * sizeof(t))) | |
981 | #define Zero(d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memzero((char*)(d), (n) * sizeof(t))) | |
55497cff | 982 | |
8b44ba4c NC |
983 | #define MoveD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memmove((char*)(d),(const char*)(s), (n) * sizeof(t))) |
984 | #define CopyD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memcpy((char*)(d),(const char*)(s), (n) * sizeof(t))) | |
e90e2364 | 985 | #ifdef HAS_MEMSET |
8b44ba4c | 986 | #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t))) |
e90e2364 | 987 | #else |
8b44ba4c NC |
988 | /* Using bzero(), which returns void. */ |
989 | #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t)),d) | |
e90e2364 NC |
990 | #endif |
991 | ||
7e337ee0 JH |
992 | #define PoisonWith(d,n,t,b) (MEM_WRAP_CHECK_(n,t) (void)memset((char*)(d), (U8)(b), (n) * sizeof(t))) |
993 | #define PoisonNew(d,n,t) PoisonWith(d,n,t,0xAB) | |
994 | #define PoisonFree(d,n,t) PoisonWith(d,n,t,0xEF) | |
995 | #define Poison(d,n,t) PoisonFree(d,n,t) | |
27d5b266 | 996 | |
2304df62 | 997 | #ifdef USE_STRUCT_COPY |
ff68c719 | 998 | #define StructCopy(s,d,t) (*((t*)(d)) = *((t*)(s))) |
bee1dbe2 LW |
999 | #else |
1000 | #define StructCopy(s,d,t) Copy(s,d,1,t) | |
1001 | #endif | |
2cc61e15 | 1002 | |
622913ab | 1003 | #define C_ARRAY_LENGTH(a) (sizeof(a)/sizeof((a)[0])) |
cd431fde | 1004 | #define C_ARRAY_END(a) (a) + (sizeof(a)/sizeof((a)[0])) |
622913ab | 1005 | |
2cc61e15 DD |
1006 | #ifdef NEED_VA_COPY |
1007 | # ifdef va_copy | |
1008 | # define Perl_va_copy(s, d) va_copy(d, s) | |
2cc61e15 | 1009 | # else |
a1866d1b JH |
1010 | # if defined(__va_copy) |
1011 | # define Perl_va_copy(s, d) __va_copy(d, s) | |
1012 | # else | |
1013 | # define Perl_va_copy(s, d) Copy(s, d, 1, va_list) | |
1014 | # endif | |
2cc61e15 DD |
1015 | # endif |
1016 | #endif | |
1017 | ||
472d47bc SB |
1018 | /* convenience debug macros */ |
1019 | #ifdef USE_ITHREADS | |
1020 | #define pTHX_FORMAT "Perl interpreter: 0x%p" | |
1021 | #define pTHX__FORMAT ", Perl interpreter: 0x%p" | |
f54cb97a AL |
1022 | #define pTHX_VALUE_ (void *)my_perl, |
1023 | #define pTHX_VALUE (void *)my_perl | |
1024 | #define pTHX__VALUE_ ,(void *)my_perl, | |
1025 | #define pTHX__VALUE ,(void *)my_perl | |
472d47bc | 1026 | #else |
3609ea0d | 1027 | #define pTHX_FORMAT |
472d47bc | 1028 | #define pTHX__FORMAT |
3609ea0d | 1029 | #define pTHX_VALUE_ |
472d47bc | 1030 | #define pTHX_VALUE |
3609ea0d | 1031 | #define pTHX__VALUE_ |
472d47bc SB |
1032 | #define pTHX__VALUE |
1033 | #endif /* USE_ITHREADS */ | |
3609ea0d | 1034 | |
2acdbac1 NC |
1035 | /* Perl_deprecate was not part of the public API, and did not have a deprecate() |
1036 | shortcut macro defined without -DPERL_CORE. Neither codesearch.google.com nor | |
1037 | CPAN::Unpack show any users outside the core. */ | |
1038 | #ifdef PERL_CORE | |
d1d15184 | 1039 | # define deprecate(s) Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), "Use of " s " is deprecated") |
2acdbac1 NC |
1040 | #endif |
1041 | ||
e9a8c099 MHM |
1042 | /* |
1043 | * Local variables: | |
1044 | * c-indentation-style: bsd | |
1045 | * c-basic-offset: 4 | |
1046 | * indent-tabs-mode: t | |
1047 | * End: | |
1048 | * | |
1049 | * ex: set ts=8 sts=4 sw=4 noet: | |
1050 | */ |