Commit | Line | Data |
---|---|---|
a0d0e21e | 1 | /* handy.h |
a687059c | 2 | * |
1129b882 | 3 | * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1999, 2000, |
da5d8dbb | 4 | * 2001, 2002, 2004, 2005, 2006, 2007, 2008, 2012 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 | |
bd31be4b NC |
73 | #ifdef I_STDBOOL |
74 | # include <stdbool.h> | |
75 | # ifndef HAS_BOOL | |
76 | # define HAS_BOOL 1 | |
77 | # endif | |
78 | #endif | |
79 | ||
8e84507e | 80 | /* bool is built-in for g++-2.6.3 and later, which might be used |
c1d22f6b GS |
81 | for extensions. <_G_config.h> defines _G_HAVE_BOOL, but we can't |
82 | be sure _G_config.h will be included before this file. _G_config.h | |
8e84507e | 83 | also defines _G_HAVE_BOOL for both gcc and g++, but only g++ |
c1d22f6b GS |
84 | actually has bool. Hence, _G_HAVE_BOOL is pretty useless for us. |
85 | g++ can be identified by __GNUG__. | |
86 | Andy Dougherty February 2000 | |
5d94fbed | 87 | */ |
3609ea0d | 88 | #ifdef __GNUG__ /* GNU g++ has bool built-in */ |
5d94fbed | 89 | # ifndef HAS_BOOL |
c1d22f6b | 90 | # define HAS_BOOL 1 |
5d94fbed | 91 | # endif |
5d94fbed AD |
92 | #endif |
93 | ||
641d3f0b | 94 | /* The NeXT dynamic loader headers will not build with the bool macro |
95 | So declare them now to clear confusion. | |
96 | */ | |
8f1f23e8 | 97 | #if defined(NeXT) || defined(__NeXT__) |
641d3f0b | 98 | # undef FALSE |
99 | # undef TRUE | |
100 | typedef enum bool { FALSE = 0, TRUE = 1 } bool; | |
101 | # define ENUM_BOOL 1 | |
102 | # ifndef HAS_BOOL | |
103 | # define HAS_BOOL 1 | |
104 | # endif /* !HAS_BOOL */ | |
8f1f23e8 | 105 | #endif /* NeXT || __NeXT__ */ |
641d3f0b | 106 | |
5d94fbed | 107 | #ifndef HAS_BOOL |
61bb5906 | 108 | # if defined(UTS) || defined(VMS) |
5d94fbed AD |
109 | # define bool int |
110 | # else | |
111 | # define bool char | |
112 | # endif | |
c1d22f6b | 113 | # define HAS_BOOL 1 |
a687059c | 114 | #endif |
0d3e774c | 115 | |
f2338a2e DM |
116 | /* a simple (bool) cast may not do the right thing: if bool is defined |
117 | * as char for example, then the cast from int is implementation-defined | |
18f5643b | 118 | * (bool)!!(cbool) in a ternary triggers a bug in xlc on AIX |
f2338a2e DM |
119 | */ |
120 | ||
18f5643b | 121 | #define cBOOL(cbool) ((cbool) ? (bool)1 : (bool)0) |
f2338a2e | 122 | |
46c6c7e2 | 123 | /* Try to figure out __func__ or __FUNCTION__ equivalent, if any. |
e352bcff JH |
124 | * XXX Should really be a Configure probe, with HAS__FUNCTION__ |
125 | * and FUNCTION__ as results. | |
126 | * XXX Similarly, a Configure probe for __FILE__ and __LINE__ is needed. */ | |
46c6c7e2 JH |
127 | #if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) || (defined(__SUNPRO_C)) /* C99 or close enough. */ |
128 | # define FUNCTION__ __func__ | |
129 | #else | |
130 | # if (defined(_MSC_VER) && _MSC_VER < 1300) || /* Pre-MSVC 7.0 has neither __func__ nor __FUNCTION and no good workarounds, either. */ \ | |
be73ebb0 | 131 | (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 |
132 | # define FUNCTION__ "" |
133 | # else | |
134 | # define FUNCTION__ __FUNCTION__ /* Common extension. */ | |
135 | # endif | |
136 | #endif | |
137 | ||
27d4fb96 | 138 | /* XXX A note on the perl source internal type system. The |
139 | original intent was that I32 be *exactly* 32 bits. | |
140 | ||
141 | Currently, we only guarantee that I32 is *at least* 32 bits. | |
142 | Specifically, if int is 64 bits, then so is I32. (This is the case | |
143 | for the Cray.) This has the advantage of meshing nicely with | |
144 | standard library calls (where we pass an I32 and the library is | |
145 | expecting an int), but the disadvantage that an I32 is not 32 bits. | |
146 | Andy Dougherty August 1996 | |
24fef2a7 | 147 | |
dc45a647 MB |
148 | There is no guarantee that there is *any* integral type with |
149 | exactly 32 bits. It is perfectly legal for a system to have | |
150 | sizeof(short) == sizeof(int) == sizeof(long) == 8. | |
693762b4 | 151 | |
dc45a647 MB |
152 | Similarly, there is no guarantee that I16 and U16 have exactly 16 |
153 | bits. | |
693762b4 | 154 | |
8e84507e NIS |
155 | For dealing with issues that may arise from various 32/64-bit |
156 | systems, we will ask Configure to check out | |
8175356b | 157 | |
3609ea0d JH |
158 | SHORTSIZE == sizeof(short) |
159 | INTSIZE == sizeof(int) | |
160 | LONGSIZE == sizeof(long) | |
dc45a647 | 161 | LONGLONGSIZE == sizeof(long long) (if HAS_LONG_LONG) |
3609ea0d | 162 | PTRSIZE == sizeof(void *) |
dc45a647 MB |
163 | DOUBLESIZE == sizeof(double) |
164 | LONG_DOUBLESIZE == sizeof(long double) (if HAS_LONG_DOUBLE). | |
8175356b | 165 | |
27d4fb96 | 166 | */ |
167 | ||
69512466 JH |
168 | #ifdef I_INTTYPES /* e.g. Linux has int64_t without <inttypes.h> */ |
169 | # include <inttypes.h> | |
dd0eed91 JH |
170 | # ifdef INT32_MIN_BROKEN |
171 | # undef INT32_MIN | |
172 | # define INT32_MIN (-2147483647-1) | |
173 | # endif | |
174 | # ifdef INT64_MIN_BROKEN | |
175 | # undef INT64_MIN | |
176 | # define INT64_MIN (-9223372036854775807LL-1) | |
177 | # endif | |
69512466 JH |
178 | #endif |
179 | ||
8175356b JH |
180 | typedef I8TYPE I8; |
181 | typedef U8TYPE U8; | |
182 | typedef I16TYPE I16; | |
183 | typedef U16TYPE U16; | |
184 | typedef I32TYPE I32; | |
185 | typedef U32TYPE U32; | |
6b8eaf93 JH |
186 | #ifdef PERL_CORE |
187 | # ifdef HAS_QUAD | |
8175356b JH |
188 | typedef I64TYPE I64; |
189 | typedef U64TYPE U64; | |
6b8eaf93 JH |
190 | # endif |
191 | #endif /* PERL_CORE */ | |
8175356b | 192 | |
69512466 | 193 | #if defined(HAS_QUAD) && defined(USE_64_BIT_INT) |
80524f33 JL |
194 | # if defined(HAS_LONG_LONG) && QUADKIND == QUAD_IS_LONG_LONG |
195 | # define PeRl_INT64_C(c) CAT2(c,LL) | |
196 | # define PeRl_UINT64_C(c) CAT2(c,ULL) | |
197 | # else | |
198 | # if LONGSIZE == 8 && QUADKIND == QUAD_IS_LONG | |
199 | # define PeRl_INT64_C(c) CAT2(c,L) | |
200 | # define PeRl_UINT64_C(c) CAT2(c,UL) | |
69512466 | 201 | # else |
80524f33 JL |
202 | # if defined(_WIN64) && defined(_MSC_VER) |
203 | # define PeRl_INT64_C(c) CAT2(c,I64) | |
204 | # define PeRl_UINT64_C(c) CAT2(c,UI64) | |
69512466 | 205 | # else |
80524f33 JL |
206 | # define PeRl_INT64_C(c) ((I64TYPE)(c)) |
207 | # define PeRl_UINT64_C(c) ((U64TYPE)(c)) | |
69512466 JH |
208 | # endif |
209 | # endif | |
e8c95190 | 210 | # endif |
80524f33 JL |
211 | # ifndef UINT64_C |
212 | # define UINT64_C(c) PeRl_UINT64_C(c) | |
213 | # endif | |
214 | # ifndef INT64_C | |
215 | # define INT64_C(c) PeRl_INT64_C(c) | |
216 | # endif | |
e8c95190 | 217 | #endif |
dc750f2e | 218 | |
d8668976 | 219 | #if defined(UINT8_MAX) && defined(INT16_MAX) && defined(INT32_MAX) |
5ff3f7a4 | 220 | |
5ff3f7a4 GS |
221 | /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type. |
222 | Please search CHAR_MAX in perl.h for further details. */ | |
223 | #define U8_MAX UINT8_MAX | |
224 | #define U8_MIN UINT8_MIN | |
225 | ||
5ff3f7a4 GS |
226 | #define I16_MAX INT16_MAX |
227 | #define I16_MIN INT16_MIN | |
228 | #define U16_MAX UINT16_MAX | |
229 | #define U16_MIN UINT16_MIN | |
230 | ||
5ff3f7a4 GS |
231 | #define I32_MAX INT32_MAX |
232 | #define I32_MIN INT32_MIN | |
0e983133 GS |
233 | #ifndef UINT32_MAX_BROKEN /* e.g. HP-UX with gcc messes this up */ |
234 | # define U32_MAX UINT32_MAX | |
235 | #else | |
236 | # define U32_MAX 4294967295U | |
237 | #endif | |
5ff3f7a4 GS |
238 | #define U32_MIN UINT32_MIN |
239 | ||
240 | #else | |
241 | ||
5c9fa16e KA |
242 | /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type. |
243 | Please search CHAR_MAX in perl.h for further details. */ | |
27d4fb96 | 244 | #define U8_MAX PERL_UCHAR_MAX |
245 | #define U8_MIN PERL_UCHAR_MIN | |
79072805 | 246 | |
27d4fb96 | 247 | #define I16_MAX PERL_SHORT_MAX |
248 | #define I16_MIN PERL_SHORT_MIN | |
249 | #define U16_MAX PERL_USHORT_MAX | |
250 | #define U16_MIN PERL_USHORT_MIN | |
79072805 | 251 | |
c4f23d77 | 252 | #if LONGSIZE > 4 |
27d4fb96 | 253 | # define I32_MAX PERL_INT_MAX |
254 | # define I32_MIN PERL_INT_MIN | |
255 | # define U32_MAX PERL_UINT_MAX | |
256 | # define U32_MIN PERL_UINT_MIN | |
79072805 | 257 | #else |
27d4fb96 | 258 | # define I32_MAX PERL_LONG_MAX |
259 | # define I32_MIN PERL_LONG_MIN | |
260 | # define U32_MAX PERL_ULONG_MAX | |
261 | # define U32_MIN PERL_ULONG_MIN | |
79072805 LW |
262 | #endif |
263 | ||
5ff3f7a4 GS |
264 | #endif |
265 | ||
58a9a5d5 | 266 | /* log(2) is pretty close to 0.30103, just in case anyone is grepping for it */ |
fc36a67e | 267 | #define BIT_DIGITS(N) (((N)*146)/485 + 1) /* log2(10) =~ 146/485 */ |
268 | #define TYPE_DIGITS(T) BIT_DIGITS(sizeof(T) * 8) | |
269 | #define TYPE_CHARS(T) (TYPE_DIGITS(T) + 2) /* sign, NUL */ | |
270 | ||
ff68c719 | 271 | #define Ctl(ch) ((ch) & 037) |
8d063cd8 | 272 | |
3fe05580 MHM |
273 | /* |
274 | =head1 SV-Body Allocation | |
275 | ||
276 | =for apidoc Ama|SV*|newSVpvs|const char* s | |
277 | Like C<newSVpvn>, but takes a literal string instead of a string/length pair. | |
278 | ||
84bafc02 NC |
279 | =for apidoc Ama|SV*|newSVpvs_flags|const char* s|U32 flags |
280 | Like C<newSVpvn_flags>, but takes a literal string instead of a string/length | |
281 | pair. | |
282 | ||
3fe05580 MHM |
283 | =for apidoc Ama|SV*|newSVpvs_share|const char* s |
284 | Like C<newSVpvn_share>, but takes a literal string instead of a string/length | |
285 | pair and omits the hash parameter. | |
286 | ||
9dcc53ea Z |
287 | =for apidoc Am|void|sv_catpvs_flags|SV* sv|const char* s|I32 flags |
288 | Like C<sv_catpvn_flags>, but takes a literal string instead of a | |
289 | string/length pair. | |
290 | ||
291 | =for apidoc Am|void|sv_catpvs_nomg|SV* sv|const char* s | |
292 | Like C<sv_catpvn_nomg>, but takes a literal string instead of a | |
293 | string/length pair. | |
294 | ||
7b4e7469 | 295 | =for apidoc Am|void|sv_catpvs|SV* sv|const char* s |
3fe05580 MHM |
296 | Like C<sv_catpvn>, but takes a literal string instead of a string/length pair. |
297 | ||
9dcc53ea Z |
298 | =for apidoc Am|void|sv_catpvs_mg|SV* sv|const char* s |
299 | Like C<sv_catpvn_mg>, but takes a literal string instead of a | |
300 | string/length pair. | |
301 | ||
7b4e7469 | 302 | =for apidoc Am|void|sv_setpvs|SV* sv|const char* s |
3fe05580 MHM |
303 | Like C<sv_setpvn>, but takes a literal string instead of a string/length pair. |
304 | ||
9dcc53ea Z |
305 | =for apidoc Am|void|sv_setpvs_mg|SV* sv|const char* s |
306 | Like C<sv_setpvn_mg>, but takes a literal string instead of a | |
307 | string/length pair. | |
308 | ||
309 | =for apidoc Am|SV *|sv_setref_pvs|const char* s | |
310 | Like C<sv_setref_pvn>, but takes a literal string instead of a | |
311 | string/length pair. | |
312 | ||
3fe05580 MHM |
313 | =head1 Memory Management |
314 | ||
315 | =for apidoc Ama|char*|savepvs|const char* s | |
316 | Like C<savepvn>, but takes a literal string instead of a string/length pair. | |
317 | ||
9dcc53ea Z |
318 | =for apidoc Ama|char*|savesharedpvs|const char* s |
319 | A version of C<savepvs()> which allocates the duplicate string in memory | |
320 | which is shared between threads. | |
321 | ||
3fe05580 MHM |
322 | =head1 GV Functions |
323 | ||
324 | =for apidoc Am|HV*|gv_stashpvs|const char* name|I32 create | |
325 | Like C<gv_stashpvn>, but takes a literal string instead of a string/length pair. | |
326 | ||
327 | =head1 Hash Manipulation Functions | |
328 | ||
329 | =for apidoc Am|SV**|hv_fetchs|HV* tb|const char* key|I32 lval | |
330 | Like C<hv_fetch>, but takes a literal string instead of a string/length pair. | |
331 | ||
332 | =for apidoc Am|SV**|hv_stores|HV* tb|const char* key|NULLOK SV* val | |
333 | Like C<hv_store>, but takes a literal string instead of a string/length pair | |
334 | and omits the hash parameter. | |
335 | ||
510966aa Z |
336 | =head1 Lexer interface |
337 | ||
338 | =for apidoc Amx|void|lex_stuff_pvs|const char *pv|U32 flags | |
339 | ||
340 | Like L</lex_stuff_pvn>, but takes a literal string instead of a | |
341 | string/length pair. | |
342 | ||
3fe05580 MHM |
343 | =cut |
344 | */ | |
345 | ||
65177ae0 | 346 | /* concatenating with "" ensures that only literal strings are accepted as argument */ |
26bfd924 | 347 | #define STR_WITH_LEN(s) ("" s ""), (sizeof(s)-1) |
2efa8cc7 | 348 | |
ba3a79e7 GA |
349 | /* note that STR_WITH_LEN() can't be used as argument to macros or functions that |
350 | * under some configurations might be macros, which means that it requires the full | |
351 | * Perl_xxx(aTHX_ ...) form for any API calls where it's used. | |
352 | */ | |
353 | ||
354 | /* STR_WITH_LEN() shortcuts */ | |
355 | #define newSVpvs(str) Perl_newSVpvn(aTHX_ STR_WITH_LEN(str)) | |
84bafc02 NC |
356 | #define newSVpvs_flags(str,flags) \ |
357 | Perl_newSVpvn_flags(aTHX_ STR_WITH_LEN(str), flags) | |
ba3a79e7 | 358 | #define newSVpvs_share(str) Perl_newSVpvn_share(aTHX_ STR_WITH_LEN(str), 0) |
9dcc53ea Z |
359 | #define sv_catpvs_flags(sv, str, flags) \ |
360 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), flags) | |
361 | #define sv_catpvs_nomg(sv, str) \ | |
362 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), 0) | |
363 | #define sv_catpvs(sv, str) \ | |
364 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC) | |
365 | #define sv_catpvs_mg(sv, str) \ | |
366 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC|SV_SMAGIC) | |
3fe05580 | 367 | #define sv_setpvs(sv, str) Perl_sv_setpvn(aTHX_ sv, STR_WITH_LEN(str)) |
9dcc53ea Z |
368 | #define sv_setpvs_mg(sv, str) Perl_sv_setpvn_mg(aTHX_ sv, STR_WITH_LEN(str)) |
369 | #define sv_setref_pvs(rv, classname, str) \ | |
370 | Perl_sv_setref_pvn(aTHX_ rv, classname, STR_WITH_LEN(str)) | |
ba3a79e7 | 371 | #define savepvs(str) Perl_savepvn(aTHX_ STR_WITH_LEN(str)) |
9dcc53ea Z |
372 | #define savesharedpvs(str) Perl_savesharedpvn(aTHX_ STR_WITH_LEN(str)) |
373 | #define gv_stashpvs(str, create) \ | |
374 | Perl_gv_stashpvn(aTHX_ STR_WITH_LEN(str), create) | |
375 | #define gv_fetchpvs(namebeg, add, sv_type) \ | |
376 | Perl_gv_fetchpvn_flags(aTHX_ STR_WITH_LEN(namebeg), add, sv_type) | |
377 | #define gv_fetchpvn(namebeg, len, add, sv_type) \ | |
378 | Perl_gv_fetchpvn_flags(aTHX_ namebeg, len, add, sv_type) | |
379 | #define sv_catxmlpvs(dsv, str, utf8) \ | |
380 | Perl_sv_catxmlpvn(aTHX_ dsv, STR_WITH_LEN(str), utf8) | |
a038e571 NC |
381 | #define hv_fetchs(hv,key,lval) \ |
382 | ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \ | |
383 | (lval) ? (HV_FETCH_JUST_SV | HV_FETCH_LVALUE) \ | |
384 | : HV_FETCH_JUST_SV, NULL, 0)) | |
385 | ||
386 | #define hv_stores(hv,key,val) \ | |
387 | ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \ | |
388 | (HV_FETCH_ISSTORE|HV_FETCH_JUST_SV), (val), 0)) | |
ba3a79e7 | 389 | |
510966aa Z |
390 | #define lex_stuff_pvs(pv,flags) Perl_lex_stuff_pvn(aTHX_ STR_WITH_LEN(pv), flags) |
391 | ||
b96d8cd9 NC |
392 | #define get_cvs(str, flags) \ |
393 | Perl_get_cvn_flags(aTHX_ STR_WITH_LEN(str), (flags)) | |
5c1737d1 | 394 | |
954c1994 | 395 | /* |
ccfc67b7 JH |
396 | =head1 Miscellaneous Functions |
397 | ||
954c1994 GS |
398 | =for apidoc Am|bool|strNE|char* s1|char* s2 |
399 | Test two strings to see if they are different. Returns true or | |
400 | false. | |
401 | ||
402 | =for apidoc Am|bool|strEQ|char* s1|char* s2 | |
403 | Test two strings to see if they are equal. Returns true or false. | |
404 | ||
405 | =for apidoc Am|bool|strLT|char* s1|char* s2 | |
406 | Test two strings to see if the first, C<s1>, is less than the second, | |
407 | C<s2>. Returns true or false. | |
408 | ||
409 | =for apidoc Am|bool|strLE|char* s1|char* s2 | |
410 | Test two strings to see if the first, C<s1>, is less than or equal to the | |
411 | second, C<s2>. Returns true or false. | |
412 | ||
413 | =for apidoc Am|bool|strGT|char* s1|char* s2 | |
414 | Test two strings to see if the first, C<s1>, is greater than the second, | |
415 | C<s2>. Returns true or false. | |
416 | ||
417 | =for apidoc Am|bool|strGE|char* s1|char* s2 | |
418 | Test two strings to see if the first, C<s1>, is greater than or equal to | |
419 | the second, C<s2>. Returns true or false. | |
420 | ||
421 | =for apidoc Am|bool|strnNE|char* s1|char* s2|STRLEN len | |
422 | Test two strings to see if they are different. The C<len> parameter | |
423 | indicates the number of bytes to compare. Returns true or false. (A | |
424 | wrapper for C<strncmp>). | |
425 | ||
426 | =for apidoc Am|bool|strnEQ|char* s1|char* s2|STRLEN len | |
427 | Test two strings to see if they are equal. The C<len> parameter indicates | |
428 | the number of bytes to compare. Returns true or false. (A wrapper for | |
429 | C<strncmp>). | |
430 | ||
431 | =cut | |
432 | */ | |
433 | ||
8d063cd8 LW |
434 | #define strNE(s1,s2) (strcmp(s1,s2)) |
435 | #define strEQ(s1,s2) (!strcmp(s1,s2)) | |
436 | #define strLT(s1,s2) (strcmp(s1,s2) < 0) | |
437 | #define strLE(s1,s2) (strcmp(s1,s2) <= 0) | |
438 | #define strGT(s1,s2) (strcmp(s1,s2) > 0) | |
439 | #define strGE(s1,s2) (strcmp(s1,s2) >= 0) | |
440 | #define strnNE(s1,s2,l) (strncmp(s1,s2,l)) | |
441 | #define strnEQ(s1,s2,l) (!strncmp(s1,s2,l)) | |
378cc40b | 442 | |
36477c24 | 443 | #ifdef HAS_MEMCMP |
444 | # define memNE(s1,s2,l) (memcmp(s1,s2,l)) | |
445 | # define memEQ(s1,s2,l) (!memcmp(s1,s2,l)) | |
446 | #else | |
447 | # define memNE(s1,s2,l) (bcmp(s1,s2,l)) | |
448 | # define memEQ(s1,s2,l) (!bcmp(s1,s2,l)) | |
449 | #endif | |
450 | ||
568a785a | 451 | #define memEQs(s1, l, s2) \ |
8e6497b6 | 452 | (sizeof(s2)-1 == l && memEQ(s1, ("" s2 ""), (sizeof(s2)-1))) |
568a785a NC |
453 | #define memNEs(s1, l, s2) !memEQs(s1, l, s2) |
454 | ||
bbce6d69 | 455 | /* |
456 | * Character classes. | |
457 | * | |
458 | * Unfortunately, the introduction of locales means that we | |
459 | * can't trust isupper(), etc. to tell the truth. And when | |
460 | * it comes to /\w+/ with tainting enabled, we *must* be able | |
461 | * to trust our character classes. | |
462 | * | |
463 | * Therefore, the default tests in the text of Perl will be | |
464 | * independent of locale. Any code that wants to depend on | |
465 | * the current locale will use the tests that begin with "lc". | |
466 | */ | |
467 | ||
2304df62 AD |
468 | #ifdef HAS_SETLOCALE /* XXX Is there a better test for this? */ |
469 | # ifndef CTYPE256 | |
470 | # define CTYPE256 | |
471 | # endif | |
472 | #endif | |
473 | ||
954c1994 | 474 | /* |
ccfc67b7 JH |
475 | |
476 | =head1 Character classes | |
8a58bdcf KW |
477 | There are three variants for all the functions in this section. The base ones |
478 | operate using the character set of the platform Perl is running on. The ones | |
479 | with an C<_A> suffix operate on the ASCII character set, and the ones with an | |
480 | C<_L1> suffix operate on the full Latin1 character set. All are unaffected by | |
181ab5fd | 481 | locale and by C<use bytes>. |
8a58bdcf KW |
482 | |
483 | For ASCII platforms, the base function with no suffix and the one with the | |
484 | C<_A> suffix are identical. The function with the C<_L1> suffix imposes the | |
485 | Latin-1 character set onto the platform. That is, the code points that are | |
486 | ASCII are unaffected, since ASCII is a subset of Latin-1. But the non-ASCII | |
487 | code points are treated as if they are Latin-1 characters. For example, | |
488 | C<isSPACE_L1()> will return true when called with the code point 0xA0, which is | |
489 | the Latin-1 NO-BREAK SPACE. | |
490 | ||
491 | For EBCDIC platforms, the base function with no suffix and the one with the | |
492 | C<_L1> suffix should be identical, since, as of this writing, the EBCDIC code | |
493 | pages that Perl knows about all are equivalent to Latin-1. The function that | |
494 | ends in an C<_A> suffix will not return true unless the specified character also | |
495 | has an ASCII equivalent. | |
ccfc67b7 | 496 | |
954c1994 | 497 | =for apidoc Am|bool|isALPHA|char ch |
2787a470 | 498 | Returns a boolean indicating whether the specified character is an |
06dc89af | 499 | alphabetic character in the platform's native character set. |
8a58bdcf KW |
500 | See the L<top of this section|/Character classes> for an explanation of variants |
501 | C<isALPHA_A> and C<isALPHA_L1>. | |
502 | ||
503 | =for apidoc Am|bool|isASCII|char ch | |
504 | Returns a boolean indicating whether the specified character is one of the 128 | |
505 | characters in the ASCII character set. On non-ASCII platforms, it is if this | |
506 | character corresponds to an ASCII character. Variants C<isASCII_A()> and | |
507 | C<isASCII_L1()> are identical to C<isASCII()>. | |
954c1994 | 508 | |
954c1994 | 509 | =for apidoc Am|bool|isDIGIT|char ch |
2787a470 | 510 | Returns a boolean indicating whether the specified character is a |
06dc89af | 511 | digit in the platform's native character set. |
8a58bdcf | 512 | Variants C<isDIGIT_A> and C<isDIGIT_L1> are identical to C<isDIGIT>. |
954c1994 | 513 | |
0c82b6df | 514 | =for apidoc Am|bool|isLOWER|char ch |
2787a470 | 515 | Returns a boolean indicating whether the specified character is a |
0c82b6df | 516 | lowercase character in the platform's native character set. |
8a58bdcf KW |
517 | See the L<top of this section|/Character classes> for an explanation of variants |
518 | C<isLOWER_A> and C<isLOWER_L1>. | |
0c82b6df | 519 | |
c99e91e9 | 520 | =for apidoc Am|bool|isOCTAL|char ch |
2787a470 | 521 | Returns a boolean indicating whether the specified character is an |
06dc89af | 522 | octal digit, [0-7] in the platform's native character set. |
8a58bdcf | 523 | Variants C<isOCTAL_A> and C<isOCTAL_L1> are identical to C<isOCTAL>. |
c99e91e9 | 524 | |
0c82b6df | 525 | =for apidoc Am|bool|isSPACE|char ch |
2787a470 | 526 | Returns a boolean indicating whether the specified character is a |
8a58bdcf KW |
527 | whitespace character in the platform's native character set. This is the same |
528 | as what C<\s> matches in a regular expression. | |
529 | See the L<top of this section|/Character classes> for an explanation of variants | |
530 | C<isSPACE_A> and C<isSPACE_L1>. | |
0c82b6df | 531 | |
954c1994 | 532 | =for apidoc Am|bool|isUPPER|char ch |
2787a470 | 533 | Returns a boolean indicating whether the specified character is an |
06dc89af | 534 | uppercase character in the platform's native character set. |
8a58bdcf KW |
535 | See the L<top of this section|/Character classes> for an explanation of variants |
536 | C<isUPPER_A> and C<isUPPER_L1>. | |
954c1994 | 537 | |
8eea39dd KW |
538 | =for apidoc Am|bool|isWORDCHAR|char ch |
539 | Returns a boolean indicating whether the specified character is a | |
540 | character that is any of: alphabetic, numeric, or an underscore. This is the | |
541 | same as what C<\w> matches in a regular expression. | |
542 | C<isALNUM()> is a synonym provided for backward compatibility. Note that it | |
543 | does not have the standard C language meaning of alphanumeric, since it matches | |
544 | an underscore and the standard meaning does not. | |
8a58bdcf KW |
545 | See the L<top of this section|/Character classes> for an explanation of variants |
546 | C<isWORDCHAR_A> and C<isWORDCHAR_L1>. | |
547 | ||
548 | =for apidoc Am|bool|isXDIGIT|char ch | |
549 | Returns a boolean indicating whether the specified character is a hexadecimal | |
550 | digit, [0-9A-Fa-f]. Variants C<isXDIGIT_A()> and C<isXDIGIT_L1()> are | |
551 | identical to C<isXDIGIT()>. | |
8eea39dd | 552 | |
e7c1e6c1 KW |
553 | =head1 Character case changing |
554 | ||
954c1994 | 555 | =for apidoc Am|char|toUPPER|char ch |
06dc89af KW |
556 | Converts the specified character to uppercase in the platform's native |
557 | character set, if possible; otherwise returns the input character itself. | |
954c1994 GS |
558 | |
559 | =for apidoc Am|char|toLOWER|char ch | |
06dc89af KW |
560 | Converts the specified character to lowercase in the platform's native |
561 | character set, if possible; otherwise returns the input character itself. | |
954c1994 GS |
562 | |
563 | =cut | |
353c9b6f | 564 | |
8a58bdcf | 565 | Note that these macros are repeated in Devel::PPPort, so should also be |
62fa66b6 KW |
566 | patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc |
567 | ||
954c1994 GS |
568 | */ |
569 | ||
7c062697 KW |
570 | /* Specify the widest unsigned type on the platform. Use U64TYPE because U64 |
571 | * is known only in the perl core, and this macro can be called from outside | |
572 | * that */ | |
573 | #ifdef HAS_QUAD | |
574 | # define WIDEST_UTYPE U64TYPE | |
575 | #else | |
576 | # define WIDEST_UTYPE U32 | |
577 | #endif | |
578 | ||
3912bc88 KW |
579 | /* FITS_IN_8_BITS(c) returns true if c doesn't have a bit set other than in |
580 | * the lower 8. It is designed to be hopefully bomb-proof, making sure that no | |
581 | * bits of information are lost even on a 64-bit machine, but to get the | |
582 | * compiler to optimize it out if possible. This is because Configure makes | |
583 | * sure that the machine has an 8-bit byte, so if c is stored in a byte, the | |
584 | * sizeof() guarantees that this evaluates to a constant true at compile time. | |
220c71bf KW |
585 | */ |
586 | #define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || !(((WIDEST_UTYPE)(c)) & ~0xFF)) | |
cf301eb7 | 587 | |
41f43cc2 | 588 | #ifdef EBCDIC |
ff7eba81 | 589 | # define isASCII(c) (FITS_IN_8_BITS(c) && (NATIVE_TO_UNI((U8) (c)) < 128)) |
41f43cc2 KW |
590 | #else |
591 | # define isASCII(c) ((WIDEST_UTYPE)(c) < 128) | |
592 | #endif | |
593 | ||
ec79cfbb | 594 | #define isASCII_A(c) isASCII(c) |
3509bb53 | 595 | #define isASCII_L1(c) isASCII(c) |
6515cb76 | 596 | |
c2da0b36 KW |
597 | /* The lower 3 bits in both the ASCII and EBCDIC representations of '0' are 0, |
598 | * and the 8 possible permutations of those bits exactly comprise the 8 octal | |
599 | * digits */ | |
3d33b3e3 | 600 | #define isOCTAL_A(c) cBOOL(FITS_IN_8_BITS(c) && (0xF8 & (c)) == '0') |
c2da0b36 KW |
601 | |
602 | ||
ec79cfbb | 603 | /* ASCII range only */ |
9fb1bf9d | 604 | #ifdef H_PERL /* If have access to perl.h, lookup in its table */ |
181ab5fd KW |
605 | /* Bits for PL_charclass[]. These use names used in l1_char_class_tab.h but |
606 | * their actual definitions are here. If that has a name not used here, it | |
607 | * won't compile. */ | |
96ac0975 NC |
608 | # define _CC_ALNUMC_A (1<<0) |
609 | # define _CC_ALNUMC_L1 (1<<1) | |
610 | # define _CC_ALPHA_A (1<<2) | |
611 | # define _CC_ALPHA_L1 (1<<3) | |
612 | # define _CC_BLANK_A (1<<4) | |
613 | # define _CC_BLANK_L1 (1<<5) | |
614 | # define _CC_CHARNAME_CONT (1<<6) | |
615 | # define _CC_CNTRL_A (1<<7) | |
616 | # define _CC_CNTRL_L1 (1<<8) | |
617 | # define _CC_DIGIT_A (1<<9) | |
618 | # define _CC_GRAPH_A (1<<10) | |
619 | # define _CC_GRAPH_L1 (1<<11) | |
620 | # define _CC_IDFIRST_A (1<<12) | |
621 | # define _CC_IDFIRST_L1 (1<<13) | |
622 | # define _CC_LOWER_A (1<<14) | |
623 | # define _CC_LOWER_L1 (1<<15) | |
96ac0975 NC |
624 | # define _CC_PRINT_A (1<<17) |
625 | # define _CC_PRINT_L1 (1<<18) | |
626 | # define _CC_PSXSPC_A (1<<19) | |
627 | # define _CC_PSXSPC_L1 (1<<20) | |
628 | # define _CC_PUNCT_A (1<<21) | |
629 | # define _CC_PUNCT_L1 (1<<22) | |
630 | # define _CC_SPACE_A (1<<23) | |
631 | # define _CC_SPACE_L1 (1<<24) | |
632 | # define _CC_UPPER_A (1<<25) | |
633 | # define _CC_UPPER_L1 (1<<26) | |
634 | # define _CC_WORDCHAR_A (1<<27) | |
635 | # define _CC_WORDCHAR_L1 (1<<28) | |
636 | # define _CC_XDIGIT_A (1<<29) | |
637 | # define _CC_NONLATIN1_FOLD (1<<30) | |
2703178d | 638 | # define _CC_QUOTEMETA (1U<<31) /* 1U keeps Solaris from griping */ |
c2da0b36 | 639 | /* Unused: (1<<16) |
98d360d4 | 640 | * If more are needed, can give up some of the above. The first ones to go |
c2da0b36 KW |
641 | * would be those that require just two tests to verify; either there are two |
642 | * code points, like BLANK_A, or it occupies a single range like DIGIT_A, | |
643 | * UPPER_A, and LOWER_A. Also consider the ones that can be replaced with two | |
644 | * tests and an additional mask, so | |
645 | * | |
646 | * #define isCNTRL_A cBOOL(FITS_IN_8_BITS(c) \ | |
647 | * && (( ! (~0x1F & NATIVE_TO_UNI(c)])) \ | |
648 | * || UNLIKELY(NATIVE_TO_UNI(c) == 0x7f))) | |
649 | * | |
650 | * This takes advantage of the contiguous block of these with the first one's | |
651 | * representation having the lower order bits all zero;, except the DELETE must | |
652 | * be tested specially. A similar pattern can be used for for isCNTRL_L1, | |
653 | * isPRINT_A, and isPRINT_L1 | |
96ac0975 NC |
654 | */ |
655 | ||
656 | # ifdef DOINIT | |
657 | EXTCONST U32 PL_charclass[] = { | |
658 | # include "l1_char_class_tab.h" | |
659 | }; | |
660 | ||
661 | # else /* ! DOINIT */ | |
662 | EXTCONST U32 PL_charclass[]; | |
663 | # endif | |
664 | ||
9fb1bf9d KW |
665 | # define isALNUMC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_A)) |
666 | # define isALPHA_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_A)) | |
667 | # define isBLANK_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_A)) | |
668 | # define isCNTRL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_A)) | |
669 | # define isDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_DIGIT_A)) | |
670 | # define isGRAPH_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_A)) | |
671 | # define isIDFIRST_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_A)) | |
672 | # define isLOWER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_A)) | |
9fb1bf9d KW |
673 | # define isPRINT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_A)) |
674 | # define isPSXSPC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_A)) | |
675 | # define isPUNCT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_A)) | |
676 | # define isSPACE_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_A)) | |
677 | # define isUPPER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_A)) | |
678 | # define isWORDCHAR_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_A)) | |
679 | # define isXDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_XDIGIT_A)) | |
fb9bfbf7 KW |
680 | /* Either participates in a fold with a character above 255, or is a |
681 | * multi-char fold */ | |
e2e75538 | 682 | # define _HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(c) ((! cBOOL(FITS_IN_8_BITS(c))) || (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_NONLATIN1_FOLD)) |
25093c95 | 683 | # define _isQUOTEMETA(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_QUOTEMETA)) |
9fb1bf9d | 684 | #else /* No perl.h. */ |
9fb1bf9d KW |
685 | # ifdef EBCDIC |
686 | # define isALNUMC_A(c) (isASCII(c) && isALNUMC(c)) | |
687 | # define isALPHA_A(c) (isASCII(c) && isALPHA(c)) | |
688 | # define isBLANK_A(c) (isASCII(c) && isBLANK(c)) | |
689 | # define isCNTRL_A(c) (isASCII(c) && isCNTRL(c)) | |
690 | # define isDIGIT_A(c) (isASCII(c) && isDIGIT(c)) | |
691 | # define isGRAPH_A(c) (isASCII(c) && isGRAPH(c)) | |
692 | # define isIDFIRST_A(c) (isASCII(c) && isIDFIRST(c)) | |
693 | # define isLOWER_A(c) (isASCII(c) && isLOWER(c)) | |
694 | # define isPRINT_A(c) (isASCII(c) && isPRINT(c)) | |
695 | # define isPSXSPC_A(c) (isASCII(c) && isPSXSPC(c)) | |
696 | # define isPUNCT_A(c) (isASCII(c) && isPUNCT(c)) | |
697 | # define isSPACE_A(c) (isASCII(c) && isSPACE(c)) | |
698 | # define isUPPER_A(c) (isASCII(c) && isUPPER(c)) | |
699 | # define isWORDCHAR_A(c) (isASCII(c) && isWORDCHAR(c)) | |
700 | # define isXDIGIT_A(c) (isASCII(c) && isXDIGIT(c)) | |
701 | # else /* ASCII platform, no perl.h */ | |
702 | # define isALNUMC_A(c) (isALPHA_A(c) || isDIGIT_A(c)) | |
703 | # define isALPHA_A(c) (isUPPER_A(c) || isLOWER_A(c)) | |
704 | # define isBLANK_A(c) ((c) == ' ' || (c) == '\t') | |
ff7eba81 | 705 | # define isCNTRL_A(c) (FITS_IN_8_BITS(c) && ((U8) (c) < ' ' || (c) == 127)) |
2181da29 | 706 | # define isDIGIT_A(c) ((c) <= '9' && (c) >= '0') |
9fb1bf9d KW |
707 | # define isGRAPH_A(c) (isWORDCHAR_A(c) || isPUNCT_A(c)) |
708 | # define isIDFIRST_A(c) (isALPHA_A(c) || (c) == '_') | |
709 | # define isLOWER_A(c) ((c) >= 'a' && (c) <= 'z') | |
710 | # define isPRINT_A(c) (((c) >= 32 && (c) < 127)) | |
711 | # define isPSXSPC_A(c) (isSPACE_A(c) || (c) == '\v') | |
712 | # define isPUNCT_A(c) (((c) >= 33 && (c) <= 47) || ((c) >= 58 && (c) <= 64) || ((c) >= 91 && (c) <= 96) || ((c) >= 123 && (c) <= 126)) | |
713 | # define isSPACE_A(c) ((c) == ' ' || (c) == '\t' || (c) == '\n' || (c) =='\r' || (c) == '\f') | |
2181da29 | 714 | # define isUPPER_A(c) ((c) <= 'Z' && (c) >= 'A') |
9fb1bf9d | 715 | # define isWORDCHAR_A(c) (isALPHA_A(c) || isDIGIT_A(c) || (c) == '_') |
2181da29 | 716 | # define isXDIGIT_A(c) (isDIGIT_A(c) || ((c) >= 'a' && (c) <= 'f') || ((c) <= 'F' && (c) >= 'A')) |
9fb1bf9d KW |
717 | # endif |
718 | #endif /* ASCII range definitions */ | |
ec79cfbb KW |
719 | |
720 | /* Latin1 definitions */ | |
8a58bdcf KW |
721 | #ifdef H_PERL |
722 | # define isALNUMC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_L1)) | |
723 | # define isALPHA_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_L1)) | |
724 | # define isBLANK_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_L1)) | |
725 | /* continuation character for legal NAME in \N{NAME} */ | |
726 | # define isCHARNAME_CONT(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CHARNAME_CONT)) | |
727 | # define isCNTRL_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_L1)) | |
728 | # define isGRAPH_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_L1)) | |
729 | # define isIDFIRST_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_L1)) | |
730 | # define isLOWER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_L1)) | |
731 | # define isPRINT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_L1)) | |
732 | # define isPSXSPC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_L1)) | |
733 | # define isPUNCT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_L1)) | |
734 | # define isSPACE_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_L1)) | |
735 | # define isUPPER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_L1)) | |
736 | # define isWORDCHAR_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_L1)) | |
737 | #else /* No access to perl.h. Only a few provided here, just in case needed | |
738 | * for backwards compatibility */ | |
739 | /* ALPHAU includes Unicode semantics for latin1 characters. It has an extra | |
740 | * >= AA test to speed up ASCII-only tests at the expense of the others */ | |
741 | # define isALPHA_L1(c) (isALPHA(c) || (NATIVE_TO_UNI((U8) c) >= 0xAA \ | |
742 | && ((NATIVE_TO_UNI((U8) c) >= 0xC0 \ | |
743 | && NATIVE_TO_UNI((U8) c) != 0xD7 && NATIVE_TO_UNI((U8) c) != 0xF7) \ | |
744 | || NATIVE_TO_UNI((U8) c) == 0xAA \ | |
745 | || NATIVE_TO_UNI((U8) c) == 0xB5 \ | |
746 | || NATIVE_TO_UNI((U8) c) == 0xBA))) | |
747 | # define isCHARNAME_CONT(c) (isALNUM_L1(c) || (c) == ' ' || (c) == '-' || (c) == '(' || (c) == ')' || (c) == ':' || NATIVE_TO_UNI((U8) c) == 0xA0) | |
748 | #endif | |
749 | ||
750 | /* Macros for backwards compatibility and for completeness when the ASCII and | |
751 | * Latin1 values are identical */ | |
ec79cfbb KW |
752 | #define isALNUM(c) isWORDCHAR(c) |
753 | #define isALNUMU(c) isWORDCHAR_L1(c) | |
8a58bdcf KW |
754 | #define isALPHAU(c) isALPHA_L1(c) |
755 | #define isDIGIT_L1(c) isDIGIT_A(c) | |
756 | #define isOCTAL(c) isOCTAL_A(c) | |
757 | #define isOCTAL_L1(c) isOCTAL_A(c) | |
758 | #define isXDIGIT_L1(c) isXDIGIT_A(c) | |
759 | ||
760 | /* Macros that differ between EBCDIC and ASCII. Where C89 defines a function, | |
761 | * that is used in the EBCDIC form, because in EBCDIC we do not do locales: | |
762 | * therefore can use native functions. For those where C89 doesn't define a | |
763 | * function, use our function, assuming that the EBCDIC code page is isomorphic | |
764 | * with Latin1, which the three currently recognized by Perl are. Some libc's | |
765 | * have an isblank(), but it's not guaranteed. */ | |
9d116dd7 | 766 | #ifdef EBCDIC |
b8c5462f | 767 | # define isALNUMC(c) isalnum(c) |
271b5cb3 | 768 | # define isALPHA(c) isalpha(c) |
0bc70f7f | 769 | # define isBLANK(c) ((c) == ' ' || (c) == '\t' || NATIVE_TO_UNI(c) == 0xA0) |
b8c5462f | 770 | # define isCNTRL(c) iscntrl(c) |
9916188f | 771 | # define isDIGIT(c) isdigit(c) |
b8c5462f | 772 | # define isGRAPH(c) isgraph(c) |
8a58bdcf | 773 | # define isIDFIRST(c) (isALPHA(c) || (c) == '_') |
7d79683e | 774 | # define isLOWER(c) islower(c) |
9d116dd7 | 775 | # define isPRINT(c) isprint(c) |
56e4b611 | 776 | # define isPSXSPC(c) isspace(c) |
b8c5462f | 777 | # define isPUNCT(c) ispunct(c) |
174b5e76 | 778 | # define isSPACE(c) (isPSXSPC(c) && (c) != '\v') |
7d79683e | 779 | # define isUPPER(c) isupper(c) |
b8c5462f | 780 | # define isXDIGIT(c) isxdigit(c) |
8a58bdcf | 781 | # define isWORDCHAR(c) (isalnum(c) || (c) == '_') |
9d116dd7 | 782 | # define toLOWER(c) tolower(c) |
7d79683e | 783 | # define toUPPER(c) toupper(c) |
ec79cfbb | 784 | #else /* Not EBCDIC: ASCII-only matching */ |
e58b1dde | 785 | # define isALNUMC(c) isALNUMC_A(c) /* Mnemonic: "C's alnum" = alpha + digit */ |
271b5cb3 | 786 | # define isALPHA(c) isALPHA_A(c) |
ec79cfbb KW |
787 | # define isBLANK(c) isBLANK_A(c) |
788 | # define isCNTRL(c) isCNTRL_A(c) | |
789 | # define isDIGIT(c) isDIGIT_A(c) | |
790 | # define isGRAPH(c) isGRAPH_A(c) | |
8a58bdcf | 791 | # define isIDFIRST(c) isIDFIRST_A(c) |
ec79cfbb KW |
792 | # define isLOWER(c) isLOWER_A(c) |
793 | # define isPRINT(c) isPRINT_A(c) | |
794 | # define isPSXSPC(c) isPSXSPC_A(c) | |
795 | # define isPUNCT(c) isPUNCT_A(c) | |
796 | # define isSPACE(c) isSPACE_A(c) | |
797 | # define isUPPER(c) isUPPER_A(c) | |
8a58bdcf | 798 | # define isWORDCHAR(c) isWORDCHAR_A(c) |
ec79cfbb | 799 | # define isXDIGIT(c) isXDIGIT_A(c) |
00f254e2 | 800 | |
8a58bdcf KW |
801 | /* ASCII casing. These could also be written as |
802 | #define toLOWER(c) (isASCII(c) ? toLOWER_LATIN1(c) : (c)) | |
803 | #define toUPPER(c) (isASCII(c) ? toUPPER_LATIN1_MOD(c) : (c)) | |
804 | which uses table lookup and mask instead of subtraction. (This would | |
805 | work because the _MOD does not apply in the ASCII range) */ | |
9d116dd7 | 806 | # define toLOWER(c) (isUPPER(c) ? (c) + ('a' - 'A') : (c)) |
7d79683e | 807 | # define toUPPER(c) (isLOWER(c) ? (c) - ('a' - 'A') : (c)) |
9d116dd7 | 808 | #endif |
bbce6d69 | 809 | |
1a0901db KW |
810 | |
811 | /* Use table lookup for speed; return error character for input | |
812 | * out-of-range */ | |
813 | #define toLOWER_LATIN1(c) (FITS_IN_8_BITS(c) \ | |
814 | ? UNI_TO_NATIVE(PL_latin1_lc[ \ | |
815 | NATIVE_TO_UNI( (U8) (c)) ]) \ | |
816 | : UNICODE_REPLACEMENT) | |
817 | /* Modified uc. Is correct uc except for three non-ascii chars which are | |
818 | * all mapped to one of them, and these need special handling; error | |
819 | * character for input out-of-range */ | |
820 | #define toUPPER_LATIN1_MOD(c) (FITS_IN_8_BITS(c) \ | |
821 | ? UNI_TO_NATIVE(PL_mod_latin1_uc[ \ | |
822 | NATIVE_TO_UNI( (U8) (c)) ]) \ | |
823 | : UNICODE_REPLACEMENT) | |
824 | ||
bbce6d69 | 825 | #ifdef USE_NEXT_CTYPE |
826 | ||
827 | # define isALNUM_LC(c) \ | |
37bd1396 | 828 | (NXIsAlNum((unsigned int)(c)) || (char)(c) == '_') |
ff68c719 | 829 | # define isIDFIRST_LC(c) \ |
830 | (NXIsAlpha((unsigned int)(c)) || (char)(c) == '_') | |
831 | # define isALPHA_LC(c) NXIsAlpha((unsigned int)(c)) | |
07315176 | 832 | # define isASCII_LC(c) isASCII((unsigned int)(c)) |
67addccf | 833 | # define isBLANK_LC(c) isBLANK((unsigned int)(c)) |
ff68c719 | 834 | # define isSPACE_LC(c) NXIsSpace((unsigned int)(c)) |
835 | # define isDIGIT_LC(c) NXIsDigit((unsigned int)(c)) | |
836 | # define isUPPER_LC(c) NXIsUpper((unsigned int)(c)) | |
837 | # define isLOWER_LC(c) NXIsLower((unsigned int)(c)) | |
37bd1396 | 838 | # define isALNUMC_LC(c) NXIsAlNum((unsigned int)(c)) |
b8c5462f JH |
839 | # define isCNTRL_LC(c) NXIsCntrl((unsigned int)(c)) |
840 | # define isGRAPH_LC(c) NXIsGraph((unsigned int)(c)) | |
ff68c719 | 841 | # define isPRINT_LC(c) NXIsPrint((unsigned int)(c)) |
b8c5462f | 842 | # define isPUNCT_LC(c) NXIsPunct((unsigned int)(c)) |
ff68c719 | 843 | # define toUPPER_LC(c) NXToUpper((unsigned int)(c)) |
844 | # define toLOWER_LC(c) NXToLower((unsigned int)(c)) | |
bbce6d69 | 845 | |
846 | #else /* !USE_NEXT_CTYPE */ | |
b8c5462f | 847 | |
bbce6d69 | 848 | # if defined(CTYPE256) || (!defined(isascii) && !defined(HAS_ISASCII)) |
849 | ||
851131d3 KW |
850 | /* Note that the foo_LC() macros in this case generally are defined only on |
851 | * code points 0-256, and give undefined, unwarned results if called with | |
852 | * values outside that range */ | |
853 | ||
b8c5462f | 854 | # define isALNUM_LC(c) (isalnum((unsigned char)(c)) || (char)(c) == '_') |
ff68c719 | 855 | # define isIDFIRST_LC(c) (isalpha((unsigned char)(c)) || (char)(c) == '_') |
bbce6d69 | 856 | # define isALPHA_LC(c) isalpha((unsigned char)(c)) |
07315176 KW |
857 | # ifdef HAS_ISASCII |
858 | # define isASCII_LC(c) isascii((unsigned char)(c)) | |
859 | # else | |
860 | # define isASCII_LC(c) isASCII((unsigned char)(c)) | |
861 | # endif | |
67addccf KW |
862 | # ifdef HAS_ISBLANK |
863 | # define isBLANK_LC(c) isblank((unsigned char)(c)) | |
864 | # else | |
865 | # define isBLANK_LC(c) isBLANK((unsigned char)(c)) | |
866 | # endif | |
bbce6d69 | 867 | # define isSPACE_LC(c) isspace((unsigned char)(c)) |
868 | # define isDIGIT_LC(c) isdigit((unsigned char)(c)) | |
869 | # define isUPPER_LC(c) isupper((unsigned char)(c)) | |
870 | # define isLOWER_LC(c) islower((unsigned char)(c)) | |
b8c5462f JH |
871 | # define isALNUMC_LC(c) isalnum((unsigned char)(c)) |
872 | # define isCNTRL_LC(c) iscntrl((unsigned char)(c)) | |
873 | # define isGRAPH_LC(c) isgraph((unsigned char)(c)) | |
bbce6d69 | 874 | # define isPRINT_LC(c) isprint((unsigned char)(c)) |
b8c5462f | 875 | # define isPUNCT_LC(c) ispunct((unsigned char)(c)) |
bbce6d69 | 876 | # define toUPPER_LC(c) toupper((unsigned char)(c)) |
877 | # define toLOWER_LC(c) tolower((unsigned char)(c)) | |
878 | ||
879 | # else | |
880 | ||
3609ea0d | 881 | # define isALNUM_LC(c) (isascii(c) && (isalnum(c) || (c) == '_')) |
bbce6d69 | 882 | # define isIDFIRST_LC(c) (isascii(c) && (isalpha(c) || (c) == '_')) |
883 | # define isALPHA_LC(c) (isascii(c) && isalpha(c)) | |
07315176 | 884 | # define isASCII_LC(c) isascii(c) |
67addccf KW |
885 | # ifdef HAS_ISBLANK |
886 | # define isBLANK_LC(c) (isascii(c) && isblank(c)) | |
887 | # else | |
888 | # define isBLANK_LC(c) isBLANK(c) | |
889 | # endif | |
bbce6d69 | 890 | # define isSPACE_LC(c) (isascii(c) && isspace(c)) |
891 | # define isDIGIT_LC(c) (isascii(c) && isdigit(c)) | |
892 | # define isUPPER_LC(c) (isascii(c) && isupper(c)) | |
893 | # define isLOWER_LC(c) (isascii(c) && islower(c)) | |
b8c5462f JH |
894 | # define isALNUMC_LC(c) (isascii(c) && isalnum(c)) |
895 | # define isCNTRL_LC(c) (isascii(c) && iscntrl(c)) | |
896 | # define isGRAPH_LC(c) (isascii(c) && isgraph(c)) | |
bbce6d69 | 897 | # define isPRINT_LC(c) (isascii(c) && isprint(c)) |
b8c5462f | 898 | # define isPUNCT_LC(c) (isascii(c) && ispunct(c)) |
bbce6d69 | 899 | # define toUPPER_LC(c) toupper(c) |
900 | # define toLOWER_LC(c) tolower(c) | |
901 | ||
902 | # endif | |
a0d0e21e | 903 | #endif /* USE_NEXT_CTYPE */ |
55204971 | 904 | |
aaa51d5e | 905 | #define isPSXSPC_LC(c) (isSPACE_LC(c) || (c) == '\v') |
aaa51d5e | 906 | |
eb0052d1 KW |
907 | /* For use in the macros just below. If the input is Latin1, use the Latin1 |
908 | * (_L1) version of the macro; otherwise use the function. Won't compile if | |
909 | * 'c' isn't unsigned, as won't match function prototype. The macros do bounds | |
910 | * checking, so have duplicate checks here, so could create versions of the | |
911 | * macros that don't, but experiments show that gcc optimizes them out anyway. | |
912 | */ | |
913 | #define generic_uni(macro, function, c) ((c) < 256 \ | |
914 | ? CAT2(macro, _L1)(c) \ | |
915 | : function(c)) | |
181ab5fd | 916 | /* Note that all ignore 'use bytes' */ |
eb0052d1 KW |
917 | |
918 | #define isALNUM_uni(c) generic_uni(isWORDCHAR, is_uni_alnum, c) | |
bdd8600f | 919 | #define isBLANK_uni(c) generic_uni(isBLANK, is_uni_blank, c) |
eb0052d1 KW |
920 | #define isIDFIRST_uni(c) generic_uni(isIDFIRST, is_uni_idfirst, c) |
921 | #define isALPHA_uni(c) generic_uni(isALPHA, is_uni_alpha, c) | |
922 | #define isSPACE_uni(c) generic_uni(isSPACE, is_uni_space, c) | |
923 | #define isDIGIT_uni(c) generic_uni(isDIGIT, is_uni_digit, c) | |
924 | #define isUPPER_uni(c) generic_uni(isUPPER, is_uni_upper, c) | |
925 | #define isLOWER_uni(c) generic_uni(isLOWER, is_uni_lower, c) | |
141d8bad | 926 | #define isASCII_uni(c) isASCII(c) |
7b952154 | 927 | /* All controls are in Latin1 */ |
ff7eba81 | 928 | #define isCNTRL_uni(c) ((c) < 256 && isCNTRL_L1(c)) |
eb0052d1 KW |
929 | #define isGRAPH_uni(c) generic_uni(isGRAPH, is_uni_graph, c) |
930 | #define isPRINT_uni(c) generic_uni(isPRINT, is_uni_print, c) | |
931 | #define isPUNCT_uni(c) generic_uni(isPUNCT, is_uni_punct, c) | |
932 | #define isXDIGIT_uni(c) generic_uni(isXDIGIT, is_uni_xdigit, c) | |
a2a2844f JH |
933 | #define toUPPER_uni(c,s,l) to_uni_upper(c,s,l) |
934 | #define toTITLE_uni(c,s,l) to_uni_title(c,s,l) | |
935 | #define toLOWER_uni(c,s,l) to_uni_lower(c,s,l) | |
b8d68ded | 936 | #define toFOLD_uni(c,s,l) to_uni_fold(c,s,l) |
a0ed51b3 | 937 | |
eb0052d1 KW |
938 | /* Posix and regular space differ only in U+000B, which is in Latin1 */ |
939 | #define isPSXSPC_uni(c) ((c) < 256 ? isPSXSPC_L1(c) : isSPACE_uni(c)) | |
aaa51d5e | 940 | |
9041c2e3 NIS |
941 | #define isALNUM_LC_uvchr(c) (c < 256 ? isALNUM_LC(c) : is_uni_alnum_lc(c)) |
942 | #define isIDFIRST_LC_uvchr(c) (c < 256 ? isIDFIRST_LC(c) : is_uni_idfirst_lc(c)) | |
943 | #define isALPHA_LC_uvchr(c) (c < 256 ? isALPHA_LC(c) : is_uni_alpha_lc(c)) | |
944 | #define isSPACE_LC_uvchr(c) (c < 256 ? isSPACE_LC(c) : is_uni_space_lc(c)) | |
945 | #define isDIGIT_LC_uvchr(c) (c < 256 ? isDIGIT_LC(c) : is_uni_digit_lc(c)) | |
946 | #define isUPPER_LC_uvchr(c) (c < 256 ? isUPPER_LC(c) : is_uni_upper_lc(c)) | |
947 | #define isLOWER_LC_uvchr(c) (c < 256 ? isLOWER_LC(c) : is_uni_lower_lc(c)) | |
9041c2e3 NIS |
948 | #define isCNTRL_LC_uvchr(c) (c < 256 ? isCNTRL_LC(c) : is_uni_cntrl_lc(c)) |
949 | #define isGRAPH_LC_uvchr(c) (c < 256 ? isGRAPH_LC(c) : is_uni_graph_lc(c)) | |
950 | #define isPRINT_LC_uvchr(c) (c < 256 ? isPRINT_LC(c) : is_uni_print_lc(c)) | |
951 | #define isPUNCT_LC_uvchr(c) (c < 256 ? isPUNCT_LC(c) : is_uni_punct_lc(c)) | |
a0ed51b3 | 952 | |
aaa51d5e JF |
953 | #define isPSXSPC_LC_uni(c) (isSPACE_LC_uni(c) ||(c) == '\f') |
954 | #define isBLANK_LC_uni(c) isBLANK(c) /* could be wrong */ | |
955 | ||
c8362b00 | 956 | /* For use in the macros just below. If the input is ASCII, use the ASCII (_A) |
7d7a6efc KW |
957 | * version of the macro; if the input is in the upper Latin1 range, use the |
958 | * Latin1 (_L1) version of the macro, after converting from utf8; otherwise use | |
959 | * the function. This relies on the fact that ASCII characters have the same | |
960 | * representation whether utf8 or not */ | |
961 | #define generic_utf8(macro, function, p) (isASCII(*(p)) \ | |
57a0fb10 | 962 | ? CAT2(CAT2(macro,_),A)(*(p)) \ |
7d7a6efc KW |
963 | : (UTF8_IS_DOWNGRADEABLE_START(*(p))) \ |
964 | ? CAT2(macro, _L1) \ | |
965 | (TWO_BYTE_UTF8_TO_UNI(*(p), \ | |
966 | *((p)+1))) \ | |
967 | : function(p)) | |
c8362b00 | 968 | |
4b88fb76 KW |
969 | /* Note that all assume that the utf8 has been validated, and ignore 'use |
970 | * bytes' */ | |
181ab5fd | 971 | |
c8362b00 | 972 | #define isALNUM_utf8(p) generic_utf8(isWORDCHAR, is_utf8_alnum, p) |
c11ff943 KW |
973 | /* To prevent S_scan_word in toke.c from hanging, we have to make sure that |
974 | * IDFIRST is an alnum. See | |
c8362b00 | 975 | * http://rt.perl.org/rt3/Ticket/Display.html?id=74022 for more detail than you |
9479a769 KW |
976 | * ever wanted to know about. XXX It is unclear if this should extend to |
977 | * isIDFIRST_uni() which it hasn't so far. (In the ASCII range, there isn't a | |
978 | * difference.) This used to be not the XID version, but we decided to go with | |
979 | * the more modern Unicode definition */ | |
c8362b00 KW |
980 | #define isIDFIRST_utf8(p) (isASCII(*(p)) \ |
981 | ? isIDFIRST_A(*(p)) \ | |
7d7a6efc KW |
982 | : (UTF8_IS_DOWNGRADEABLE_START(*(p))) \ |
983 | ? isIDFIRST_L1(TWO_BYTE_UTF8_TO_UNI(*(p), \ | |
984 | *((p)+1)))\ | |
4244b029 | 985 | : Perl__is_utf8__perl_idstart(aTHX_ p)) |
c8362b00 KW |
986 | #define isIDCONT_utf8(p) generic_utf8(isWORDCHAR, is_utf8_xidcont, p) |
987 | #define isALPHA_utf8(p) generic_utf8(isALPHA, is_utf8_alpha, p) | |
bdd8600f | 988 | #define isBLANK_utf8(p) generic_utf8(isBLANK, is_utf8_blank, p) |
c8362b00 KW |
989 | #define isSPACE_utf8(p) generic_utf8(isSPACE, is_utf8_space, p) |
990 | #define isDIGIT_utf8(p) generic_utf8(isDIGIT, is_utf8_digit, p) | |
991 | #define isUPPER_utf8(p) generic_utf8(isUPPER, is_utf8_upper, p) | |
992 | #define isLOWER_utf8(p) generic_utf8(isLOWER, is_utf8_lower, p) | |
141d8bad | 993 | /* Because ASCII is invariant under utf8, the non-utf8 macro works */ |
bce3d7a4 | 994 | #define isASCII_utf8(p) isASCII(*p) |
c8362b00 KW |
995 | #define isCNTRL_utf8(p) generic_utf8(isCNTRL, is_utf8_cntrl, p) |
996 | #define isGRAPH_utf8(p) generic_utf8(isGRAPH, is_utf8_graph, p) | |
997 | #define isPRINT_utf8(p) generic_utf8(isPRINT, is_utf8_print, p) | |
998 | #define isPUNCT_utf8(p) generic_utf8(isPUNCT, is_utf8_punct, p) | |
999 | #define isXDIGIT_utf8(p) generic_utf8(isXDIGIT, is_utf8_xdigit, p) | |
a2a2844f JH |
1000 | #define toUPPER_utf8(p,s,l) to_utf8_upper(p,s,l) |
1001 | #define toTITLE_utf8(p,s,l) to_utf8_title(p,s,l) | |
1002 | #define toLOWER_utf8(p,s,l) to_utf8_lower(p,s,l) | |
a0ed51b3 | 1003 | |
c8362b00 KW |
1004 | /* Posix and regular space differ only in U+000B, which is in ASCII (and hence |
1005 | * Latin1 */ | |
1006 | #define isPSXSPC_utf8(p) ((isASCII(*(p))) \ | |
1007 | ? isPSXSPC_A(*(p)) \ | |
7d7a6efc KW |
1008 | : (UTF8_IS_DOWNGRADEABLE_START(*(p)) \ |
1009 | ? isPSXSPC_L1(TWO_BYTE_UTF8_TO_UNI(*(p), \ | |
1010 | *((p)+1)))\ | |
1011 | : isSPACE_utf8(p))) | |
4b88fb76 KW |
1012 | #define isALNUM_LC_utf8(p) isALNUM_LC_uvchr(valid_utf8_to_uvchr(p, 0)) |
1013 | #define isIDFIRST_LC_utf8(p) isIDFIRST_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1014 | #define isALPHA_LC_utf8(p) isALPHA_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
bdd8600f | 1015 | #define isBLANK_LC_utf8(p) isBLANK_LC_uvchr(valid_utf8_to_uvchr(p, 0)) |
4b88fb76 KW |
1016 | #define isSPACE_LC_utf8(p) isSPACE_LC_uvchr(valid_utf8_to_uvchr(p, 0)) |
1017 | #define isDIGIT_LC_utf8(p) isDIGIT_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1018 | #define isUPPER_LC_utf8(p) isUPPER_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1019 | #define isLOWER_LC_utf8(p) isLOWER_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1020 | #define isALNUMC_LC_utf8(p) isALNUMC_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1021 | #define isCNTRL_LC_utf8(p) isCNTRL_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1022 | #define isGRAPH_LC_utf8(p) isGRAPH_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1023 | #define isPRINT_LC_utf8(p) isPRINT_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
1024 | #define isPUNCT_LC_utf8(p) isPUNCT_LC_uvchr(valid_utf8_to_uvchr(p, 0)) | |
a0ed51b3 | 1025 | |
aaa51d5e | 1026 | #define isPSXSPC_LC_utf8(c) (isSPACE_LC_utf8(c) ||(c) == '\f') |
aaa51d5e | 1027 | |
6ae9f32a KW |
1028 | /* This conversion works both ways, strangely enough. On EBCDIC platforms, |
1029 | * CTRL-@ is 0, CTRL-A is 1, etc, just like on ASCII */ | |
1030 | # define toCTRL(c) (toUPPER(NATIVE_TO_UNI(c)) ^ 64) | |
bbce6d69 | 1031 | |
dea28490 JJ |
1032 | /* Line numbers are unsigned, 32 bits. */ |
1033 | typedef U32 line_t; | |
dea28490 | 1034 | #define NOLINE ((line_t) 4294967295UL) |
378cc40b | 1035 | |
91152fc1 DG |
1036 | /* Helpful alias for version prescan */ |
1037 | #define is_LAX_VERSION(a,b) \ | |
1038 | (a != Perl_prescan_version(aTHX_ a, FALSE, b, NULL, NULL, NULL, NULL)) | |
1039 | ||
1040 | #define is_STRICT_VERSION(a,b) \ | |
1041 | (a != Perl_prescan_version(aTHX_ a, TRUE, b, NULL, NULL, NULL, NULL)) | |
1042 | ||
1043 | #define BADVERSION(a,b,c) \ | |
1044 | if (b) { \ | |
1045 | *b = c; \ | |
1046 | } \ | |
1047 | return a; | |
8c52afec | 1048 | |
8e84507e | 1049 | /* |
ccfc67b7 JH |
1050 | =head1 Memory Management |
1051 | ||
a02a5408 | 1052 | =for apidoc Am|void|Newx|void* ptr|int nitems|type |
954c1994 GS |
1053 | The XSUB-writer's interface to the C C<malloc> function. |
1054 | ||
c5008215 JC |
1055 | In 5.9.3, Newx() and friends replace the older New() API, and drops |
1056 | the first parameter, I<x>, a debug aid which allowed callers to identify | |
37b8b4c9 | 1057 | themselves. This aid has been superseded by a new build option, |
d10b4965 | 1058 | PERL_MEM_LOG (see L<perlhacktips/PERL_MEM_LOG>). The older API is still |
c5008215 JC |
1059 | there for use in XS modules supporting older perls. |
1060 | ||
a02a5408 | 1061 | =for apidoc Am|void|Newxc|void* ptr|int nitems|type|cast |
954c1994 | 1062 | The XSUB-writer's interface to the C C<malloc> function, with |
c5008215 | 1063 | cast. See also C<Newx>. |
954c1994 | 1064 | |
a02a5408 | 1065 | =for apidoc Am|void|Newxz|void* ptr|int nitems|type |
954c1994 | 1066 | The XSUB-writer's interface to the C C<malloc> function. The allocated |
c5008215 | 1067 | memory is zeroed with C<memzero>. See also C<Newx>. |
a02a5408 | 1068 | |
954c1994 GS |
1069 | =for apidoc Am|void|Renew|void* ptr|int nitems|type |
1070 | The XSUB-writer's interface to the C C<realloc> function. | |
1071 | ||
1072 | =for apidoc Am|void|Renewc|void* ptr|int nitems|type|cast | |
1073 | The XSUB-writer's interface to the C C<realloc> function, with | |
1074 | cast. | |
1075 | ||
49b8b560 | 1076 | =for apidoc Am|void|Safefree|void* ptr |
954c1994 GS |
1077 | The XSUB-writer's interface to the C C<free> function. |
1078 | ||
1079 | =for apidoc Am|void|Move|void* src|void* dest|int nitems|type | |
1080 | The XSUB-writer's interface to the C C<memmove> function. The C<src> is the | |
1081 | source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is | |
1082 | the type. Can do overlapping moves. See also C<Copy>. | |
1083 | ||
e90e2364 NC |
1084 | =for apidoc Am|void *|MoveD|void* src|void* dest|int nitems|type |
1085 | Like C<Move> but returns dest. Useful for encouraging compilers to tail-call | |
1086 | optimise. | |
1087 | ||
954c1994 GS |
1088 | =for apidoc Am|void|Copy|void* src|void* dest|int nitems|type |
1089 | The XSUB-writer's interface to the C C<memcpy> function. The C<src> is the | |
1090 | source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is | |
1091 | the type. May fail on overlapping copies. See also C<Move>. | |
1092 | ||
e90e2364 NC |
1093 | =for apidoc Am|void *|CopyD|void* src|void* dest|int nitems|type |
1094 | ||
1095 | Like C<Copy> but returns dest. Useful for encouraging compilers to tail-call | |
1096 | optimise. | |
1097 | ||
954c1994 GS |
1098 | =for apidoc Am|void|Zero|void* dest|int nitems|type |
1099 | ||
1100 | The XSUB-writer's interface to the C C<memzero> function. The C<dest> is the | |
1101 | destination, C<nitems> is the number of items, and C<type> is the type. | |
1102 | ||
e90e2364 NC |
1103 | =for apidoc Am|void *|ZeroD|void* dest|int nitems|type |
1104 | ||
1105 | Like C<Zero> but returns dest. Useful for encouraging compilers to tail-call | |
1106 | optimise. | |
1107 | ||
da5d8dbb | 1108 | =for apidoc Am|void|StructCopy|type *src|type *dest|type |
4375e838 | 1109 | This is an architecture-independent macro to copy one structure to another. |
954c1994 | 1110 | |
7e337ee0 JH |
1111 | =for apidoc Am|void|PoisonWith|void* dest|int nitems|type|U8 byte |
1112 | ||
1113 | Fill up memory with a byte pattern (a byte repeated over and over | |
1114 | again) that hopefully catches attempts to access uninitialized memory. | |
1115 | ||
1116 | =for apidoc Am|void|PoisonNew|void* dest|int nitems|type | |
1117 | ||
1118 | PoisonWith(0xAB) for catching access to allocated but uninitialized memory. | |
1119 | ||
1c12ffb4 | 1120 | =for apidoc Am|void|PoisonFree|void* dest|int nitems|type |
7e337ee0 JH |
1121 | |
1122 | PoisonWith(0xEF) for catching access to freed memory. | |
1123 | ||
9965345d JH |
1124 | =for apidoc Am|void|Poison|void* dest|int nitems|type |
1125 | ||
7e337ee0 | 1126 | PoisonWith(0xEF) for catching access to freed memory. |
9965345d JH |
1127 | |
1128 | =cut */ | |
954c1994 | 1129 | |
561b68a9 SH |
1130 | /* Maintained for backwards-compatibility only. Use newSV() instead. */ |
1131 | #ifndef PERL_CORE | |
ff06c60c | 1132 | #define NEWSV(x,len) newSV(len) |
561b68a9 | 1133 | #endif |
ff06c60c | 1134 | |
19a94d75 JH |
1135 | #define MEM_SIZE_MAX ((MEM_SIZE)~0) |
1136 | ||
2720f793 JH |
1137 | /* The +0.0 in MEM_WRAP_CHECK_ is an attempt to foil |
1138 | * overly eager compilers that will bleat about e.g. | |
1139 | * (U16)n > (size_t)~0/sizeof(U16) always being false. */ | |
27d5b266 | 1140 | #ifdef PERL_MALLOC_WRAP |
fabdb6c0 | 1141 | #define MEM_WRAP_CHECK(n,t) MEM_WRAP_CHECK_1(n,t,PL_memory_wrap) |
27d5b266 | 1142 | #define MEM_WRAP_CHECK_1(n,t,a) \ |
f1f66076 | 1143 | (void)(sizeof(t) > 1 && ((MEM_SIZE)(n)+0.0) > MEM_SIZE_MAX/sizeof(t) && (Perl_croak_nocontext("%s",(a)),0)) |
8b44ba4c | 1144 | #define MEM_WRAP_CHECK_(n,t) MEM_WRAP_CHECK(n,t), |
27d5b266 | 1145 | |
f1f66076 | 1146 | #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 | 1147 | |
27d5b266 JH |
1148 | #else |
1149 | ||
410319be NC |
1150 | #define MEM_WRAP_CHECK(n,t) |
1151 | #define MEM_WRAP_CHECK_1(n,t,a) | |
1152 | #define MEM_WRAP_CHECK_2(n,t,a,b) | |
8b44ba4c NC |
1153 | #define MEM_WRAP_CHECK_(n,t) |
1154 | ||
6fff79ce | 1155 | #define PERL_STRLEN_ROUNDUP(n) (((n-1+PERL_STRLEN_ROUNDUP_QUANTUM)&~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM-1))) |
27d5b266 | 1156 | |
1936d2a7 | 1157 | #endif |
8b44ba4c | 1158 | |
fe4f188c | 1159 | #ifdef PERL_MEM_LOG |
46c6c7e2 | 1160 | /* |
9f653bb5 | 1161 | * If PERL_MEM_LOG is defined, all Newx()s, Renew()s, and Safefree()s |
46c6c7e2 JH |
1162 | * go through functions, which are handy for debugging breakpoints, but |
1163 | * which more importantly get the immediate calling environment (file and | |
e352bcff JH |
1164 | * line number, and C function name if available) passed in. This info can |
1165 | * then be used for logging the calls, for which one gets a sample | |
73d1d973 | 1166 | * implementation unless -DPERL_MEM_LOG_NOIMPL is also defined. |
3609ea0d | 1167 | * |
46c6c7e2 | 1168 | * Known problems: |
94e892a6 | 1169 | * - not all memory allocs get logged, only those |
46c6c7e2 | 1170 | * that go through Newx() and derivatives (while all |
94e892a6 | 1171 | * Safefrees do get logged) |
46c6c7e2 JH |
1172 | * - __FILE__ and __LINE__ do not work everywhere |
1173 | * - __func__ or __FUNCTION__ even less so | |
1174 | * - I think more goes on after the perlio frees but | |
1175 | * the thing is that STDERR gets closed (as do all | |
1176 | * the file descriptors) | |
1177 | * - no deeper calling stack than the caller of the Newx() | |
1178 | * or the kind, but do I look like a C reflection/introspection | |
1179 | * utility to you? | |
1180 | * - the function prototypes for the logging functions | |
1181 | * probably should maybe be somewhere else than handy.h | |
1182 | * - one could consider inlining (macrofying) the logging | |
1183 | * for speed, but I am too lazy | |
1184 | * - one could imagine recording the allocations in a hash, | |
1185 | * (keyed by the allocation address?), and maintain that | |
1186 | * through reallocs and frees, but how to do that without | |
1187 | * any News() happening...? | |
73d1d973 | 1188 | * - lots of -Ddefines to get useful/controllable output |
b953482e | 1189 | * - lots of ENV reads |
46c6c7e2 JH |
1190 | */ |
1191 | ||
12754f92 | 1192 | 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 | 1193 | |
12754f92 | 1194 | 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 | 1195 | |
12754f92 | 1196 | PERL_EXPORT_C Malloc_t Perl_mem_log_free(Malloc_t oldalloc, const char *filename, const int linenumber, const char *funcname); |
46c6c7e2 | 1197 | |
0b0ab801 | 1198 | # ifdef PERL_CORE |
73d1d973 | 1199 | # ifndef PERL_MEM_LOG_NOIMPL |
0b0ab801 MHM |
1200 | enum mem_log_type { |
1201 | MLT_ALLOC, | |
1202 | MLT_REALLOC, | |
d7a2c63c MHM |
1203 | MLT_FREE, |
1204 | MLT_NEW_SV, | |
1205 | MLT_DEL_SV | |
0b0ab801 MHM |
1206 | }; |
1207 | # endif | |
12754f92 | 1208 | # if defined(PERL_IN_SV_C) /* those are only used in sv.c */ |
d7a2c63c MHM |
1209 | void Perl_mem_log_new_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname); |
1210 | void Perl_mem_log_del_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname); | |
12754f92 | 1211 | # endif |
0b0ab801 MHM |
1212 | # endif |
1213 | ||
fe4f188c JH |
1214 | #endif |
1215 | ||
1216 | #ifdef PERL_MEM_LOG | |
d1401ee9 MHM |
1217 | #define MEM_LOG_ALLOC(n,t,a) Perl_mem_log_alloc(n,sizeof(t),STRINGIFY(t),a,__FILE__,__LINE__,FUNCTION__) |
1218 | #define MEM_LOG_REALLOC(n,t,v,a) Perl_mem_log_realloc(n,sizeof(t),STRINGIFY(t),v,a,__FILE__,__LINE__,FUNCTION__) | |
46c6c7e2 | 1219 | #define MEM_LOG_FREE(a) Perl_mem_log_free(a,__FILE__,__LINE__,FUNCTION__) |
fe4f188c JH |
1220 | #endif |
1221 | ||
1222 | #ifndef MEM_LOG_ALLOC | |
1223 | #define MEM_LOG_ALLOC(n,t,a) (a) | |
1224 | #endif | |
1225 | #ifndef MEM_LOG_REALLOC | |
1226 | #define MEM_LOG_REALLOC(n,t,v,a) (a) | |
1227 | #endif | |
1228 | #ifndef MEM_LOG_FREE | |
1229 | #define MEM_LOG_FREE(a) (a) | |
1230 | #endif | |
1231 | ||
d1401ee9 MHM |
1232 | #define Newx(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t)))))) |
1233 | #define Newxc(v,n,t,c) (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t)))))) | |
1234 | #define Newxz(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safecalloc((n),sizeof(t))))) | |
a6f6820f NC |
1235 | |
1236 | #ifndef PERL_CORE | |
a02a5408 JC |
1237 | /* pre 5.9.x compatibility */ |
1238 | #define New(x,v,n,t) Newx(v,n,t) | |
1239 | #define Newc(x,v,n,t,c) Newxc(v,n,t,c) | |
4541904d | 1240 | #define Newz(x,v,n,t) Newxz(v,n,t) |
a6f6820f | 1241 | #endif |
a02a5408 | 1242 | |
ff68c719 | 1243 | #define Renew(v,n,t) \ |
d1401ee9 | 1244 | (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t)))))) |
ff68c719 | 1245 | #define Renewc(v,n,t,c) \ |
d1401ee9 | 1246 | (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t)))))) |
94010e71 NC |
1247 | |
1248 | #ifdef PERL_POISON | |
1249 | #define Safefree(d) \ | |
06c0cc96 | 1250 | ((d) ? (void)(safefree(MEM_LOG_FREE((Malloc_t)(d))), Poison(&(d), 1, Malloc_t)) : (void) 0) |
94010e71 | 1251 | #else |
fe4f188c | 1252 | #define Safefree(d) safefree(MEM_LOG_FREE((Malloc_t)(d))) |
94010e71 | 1253 | #endif |
55497cff | 1254 | |
8b44ba4c NC |
1255 | #define Move(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memmove((char*)(d),(const char*)(s), (n) * sizeof(t))) |
1256 | #define Copy(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memcpy((char*)(d),(const char*)(s), (n) * sizeof(t))) | |
1257 | #define Zero(d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memzero((char*)(d), (n) * sizeof(t))) | |
55497cff | 1258 | |
8b44ba4c NC |
1259 | #define MoveD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memmove((char*)(d),(const char*)(s), (n) * sizeof(t))) |
1260 | #define CopyD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memcpy((char*)(d),(const char*)(s), (n) * sizeof(t))) | |
e90e2364 | 1261 | #ifdef HAS_MEMSET |
8b44ba4c | 1262 | #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t))) |
e90e2364 | 1263 | #else |
8b44ba4c NC |
1264 | /* Using bzero(), which returns void. */ |
1265 | #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t)),d) | |
e90e2364 NC |
1266 | #endif |
1267 | ||
7e337ee0 JH |
1268 | #define PoisonWith(d,n,t,b) (MEM_WRAP_CHECK_(n,t) (void)memset((char*)(d), (U8)(b), (n) * sizeof(t))) |
1269 | #define PoisonNew(d,n,t) PoisonWith(d,n,t,0xAB) | |
1270 | #define PoisonFree(d,n,t) PoisonWith(d,n,t,0xEF) | |
1271 | #define Poison(d,n,t) PoisonFree(d,n,t) | |
27d5b266 | 1272 | |
2304df62 | 1273 | #ifdef USE_STRUCT_COPY |
ff68c719 | 1274 | #define StructCopy(s,d,t) (*((t*)(d)) = *((t*)(s))) |
bee1dbe2 LW |
1275 | #else |
1276 | #define StructCopy(s,d,t) Copy(s,d,1,t) | |
1277 | #endif | |
2cc61e15 | 1278 | |
622913ab | 1279 | #define C_ARRAY_LENGTH(a) (sizeof(a)/sizeof((a)[0])) |
cd431fde | 1280 | #define C_ARRAY_END(a) (a) + (sizeof(a)/sizeof((a)[0])) |
622913ab | 1281 | |
2cc61e15 DD |
1282 | #ifdef NEED_VA_COPY |
1283 | # ifdef va_copy | |
1284 | # define Perl_va_copy(s, d) va_copy(d, s) | |
2cc61e15 | 1285 | # else |
a1866d1b JH |
1286 | # if defined(__va_copy) |
1287 | # define Perl_va_copy(s, d) __va_copy(d, s) | |
1288 | # else | |
1289 | # define Perl_va_copy(s, d) Copy(s, d, 1, va_list) | |
1290 | # endif | |
2cc61e15 DD |
1291 | # endif |
1292 | #endif | |
1293 | ||
472d47bc SB |
1294 | /* convenience debug macros */ |
1295 | #ifdef USE_ITHREADS | |
1296 | #define pTHX_FORMAT "Perl interpreter: 0x%p" | |
1297 | #define pTHX__FORMAT ", Perl interpreter: 0x%p" | |
f54cb97a AL |
1298 | #define pTHX_VALUE_ (void *)my_perl, |
1299 | #define pTHX_VALUE (void *)my_perl | |
1300 | #define pTHX__VALUE_ ,(void *)my_perl, | |
1301 | #define pTHX__VALUE ,(void *)my_perl | |
472d47bc | 1302 | #else |
3609ea0d | 1303 | #define pTHX_FORMAT |
472d47bc | 1304 | #define pTHX__FORMAT |
3609ea0d | 1305 | #define pTHX_VALUE_ |
472d47bc | 1306 | #define pTHX_VALUE |
3609ea0d | 1307 | #define pTHX__VALUE_ |
472d47bc SB |
1308 | #define pTHX__VALUE |
1309 | #endif /* USE_ITHREADS */ | |
3609ea0d | 1310 | |
2acdbac1 NC |
1311 | /* Perl_deprecate was not part of the public API, and did not have a deprecate() |
1312 | shortcut macro defined without -DPERL_CORE. Neither codesearch.google.com nor | |
1313 | CPAN::Unpack show any users outside the core. */ | |
1314 | #ifdef PERL_CORE | |
d1d15184 | 1315 | # define deprecate(s) Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), "Use of " s " is deprecated") |
2acdbac1 NC |
1316 | #endif |
1317 | ||
e9a8c099 MHM |
1318 | /* |
1319 | * Local variables: | |
1320 | * c-indentation-style: bsd | |
1321 | * c-basic-offset: 4 | |
14d04a33 | 1322 | * indent-tabs-mode: nil |
e9a8c099 MHM |
1323 | * End: |
1324 | * | |
14d04a33 | 1325 | * ex: set ts=8 sts=4 sw=4 et: |
e9a8c099 | 1326 | */ |