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 | |
d8668976 | 211 | #if defined(UINT8_MAX) && defined(INT16_MAX) && defined(INT32_MAX) |
5ff3f7a4 | 212 | |
5ff3f7a4 GS |
213 | /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type. |
214 | Please search CHAR_MAX in perl.h for further details. */ | |
215 | #define U8_MAX UINT8_MAX | |
216 | #define U8_MIN UINT8_MIN | |
217 | ||
5ff3f7a4 GS |
218 | #define I16_MAX INT16_MAX |
219 | #define I16_MIN INT16_MIN | |
220 | #define U16_MAX UINT16_MAX | |
221 | #define U16_MIN UINT16_MIN | |
222 | ||
5ff3f7a4 GS |
223 | #define I32_MAX INT32_MAX |
224 | #define I32_MIN INT32_MIN | |
0e983133 GS |
225 | #ifndef UINT32_MAX_BROKEN /* e.g. HP-UX with gcc messes this up */ |
226 | # define U32_MAX UINT32_MAX | |
227 | #else | |
228 | # define U32_MAX 4294967295U | |
229 | #endif | |
5ff3f7a4 GS |
230 | #define U32_MIN UINT32_MIN |
231 | ||
232 | #else | |
233 | ||
5c9fa16e KA |
234 | /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type. |
235 | Please search CHAR_MAX in perl.h for further details. */ | |
27d4fb96 | 236 | #define U8_MAX PERL_UCHAR_MAX |
237 | #define U8_MIN PERL_UCHAR_MIN | |
79072805 | 238 | |
27d4fb96 | 239 | #define I16_MAX PERL_SHORT_MAX |
240 | #define I16_MIN PERL_SHORT_MIN | |
241 | #define U16_MAX PERL_USHORT_MAX | |
242 | #define U16_MIN PERL_USHORT_MIN | |
79072805 | 243 | |
c4f23d77 | 244 | #if LONGSIZE > 4 |
27d4fb96 | 245 | # define I32_MAX PERL_INT_MAX |
246 | # define I32_MIN PERL_INT_MIN | |
247 | # define U32_MAX PERL_UINT_MAX | |
248 | # define U32_MIN PERL_UINT_MIN | |
79072805 | 249 | #else |
27d4fb96 | 250 | # define I32_MAX PERL_LONG_MAX |
251 | # define I32_MIN PERL_LONG_MIN | |
252 | # define U32_MAX PERL_ULONG_MAX | |
253 | # define U32_MIN PERL_ULONG_MIN | |
79072805 LW |
254 | #endif |
255 | ||
5ff3f7a4 GS |
256 | #endif |
257 | ||
58a9a5d5 | 258 | /* log(2) is pretty close to 0.30103, just in case anyone is grepping for it */ |
fc36a67e | 259 | #define BIT_DIGITS(N) (((N)*146)/485 + 1) /* log2(10) =~ 146/485 */ |
260 | #define TYPE_DIGITS(T) BIT_DIGITS(sizeof(T) * 8) | |
261 | #define TYPE_CHARS(T) (TYPE_DIGITS(T) + 2) /* sign, NUL */ | |
262 | ||
ff68c719 | 263 | #define Ctl(ch) ((ch) & 037) |
8d063cd8 | 264 | |
3fe05580 MHM |
265 | /* |
266 | =head1 SV-Body Allocation | |
267 | ||
268 | =for apidoc Ama|SV*|newSVpvs|const char* s | |
269 | Like C<newSVpvn>, but takes a literal string instead of a string/length pair. | |
270 | ||
84bafc02 NC |
271 | =for apidoc Ama|SV*|newSVpvs_flags|const char* s|U32 flags |
272 | Like C<newSVpvn_flags>, but takes a literal string instead of a string/length | |
273 | pair. | |
274 | ||
3fe05580 MHM |
275 | =for apidoc Ama|SV*|newSVpvs_share|const char* s |
276 | Like C<newSVpvn_share>, but takes a literal string instead of a string/length | |
277 | pair and omits the hash parameter. | |
278 | ||
9dcc53ea Z |
279 | =for apidoc Am|void|sv_catpvs_flags|SV* sv|const char* s|I32 flags |
280 | Like C<sv_catpvn_flags>, but takes a literal string instead of a | |
281 | string/length pair. | |
282 | ||
283 | =for apidoc Am|void|sv_catpvs_nomg|SV* sv|const char* s | |
284 | Like C<sv_catpvn_nomg>, but takes a literal string instead of a | |
285 | string/length pair. | |
286 | ||
7b4e7469 | 287 | =for apidoc Am|void|sv_catpvs|SV* sv|const char* s |
3fe05580 MHM |
288 | Like C<sv_catpvn>, but takes a literal string instead of a string/length pair. |
289 | ||
9dcc53ea Z |
290 | =for apidoc Am|void|sv_catpvs_mg|SV* sv|const char* s |
291 | Like C<sv_catpvn_mg>, but takes a literal string instead of a | |
292 | string/length pair. | |
293 | ||
7b4e7469 | 294 | =for apidoc Am|void|sv_setpvs|SV* sv|const char* s |
3fe05580 MHM |
295 | Like C<sv_setpvn>, but takes a literal string instead of a string/length pair. |
296 | ||
9dcc53ea Z |
297 | =for apidoc Am|void|sv_setpvs_mg|SV* sv|const char* s |
298 | Like C<sv_setpvn_mg>, but takes a literal string instead of a | |
299 | string/length pair. | |
300 | ||
301 | =for apidoc Am|SV *|sv_setref_pvs|const char* s | |
302 | Like C<sv_setref_pvn>, but takes a literal string instead of a | |
303 | string/length pair. | |
304 | ||
3fe05580 MHM |
305 | =head1 Memory Management |
306 | ||
307 | =for apidoc Ama|char*|savepvs|const char* s | |
308 | Like C<savepvn>, but takes a literal string instead of a string/length pair. | |
309 | ||
9dcc53ea Z |
310 | =for apidoc Ama|char*|savesharedpvs|const char* s |
311 | A version of C<savepvs()> which allocates the duplicate string in memory | |
312 | which is shared between threads. | |
313 | ||
3fe05580 MHM |
314 | =head1 GV Functions |
315 | ||
316 | =for apidoc Am|HV*|gv_stashpvs|const char* name|I32 create | |
317 | Like C<gv_stashpvn>, but takes a literal string instead of a string/length pair. | |
318 | ||
319 | =head1 Hash Manipulation Functions | |
320 | ||
321 | =for apidoc Am|SV**|hv_fetchs|HV* tb|const char* key|I32 lval | |
322 | Like C<hv_fetch>, but takes a literal string instead of a string/length pair. | |
323 | ||
324 | =for apidoc Am|SV**|hv_stores|HV* tb|const char* key|NULLOK SV* val | |
325 | Like C<hv_store>, but takes a literal string instead of a string/length pair | |
326 | and omits the hash parameter. | |
327 | ||
510966aa Z |
328 | =head1 Lexer interface |
329 | ||
330 | =for apidoc Amx|void|lex_stuff_pvs|const char *pv|U32 flags | |
331 | ||
332 | Like L</lex_stuff_pvn>, but takes a literal string instead of a | |
333 | string/length pair. | |
334 | ||
3fe05580 MHM |
335 | =cut |
336 | */ | |
337 | ||
65177ae0 | 338 | /* concatenating with "" ensures that only literal strings are accepted as argument */ |
26bfd924 | 339 | #define STR_WITH_LEN(s) ("" s ""), (sizeof(s)-1) |
2efa8cc7 | 340 | |
ba3a79e7 GA |
341 | /* note that STR_WITH_LEN() can't be used as argument to macros or functions that |
342 | * under some configurations might be macros, which means that it requires the full | |
343 | * Perl_xxx(aTHX_ ...) form for any API calls where it's used. | |
344 | */ | |
345 | ||
346 | /* STR_WITH_LEN() shortcuts */ | |
347 | #define newSVpvs(str) Perl_newSVpvn(aTHX_ STR_WITH_LEN(str)) | |
84bafc02 NC |
348 | #define newSVpvs_flags(str,flags) \ |
349 | Perl_newSVpvn_flags(aTHX_ STR_WITH_LEN(str), flags) | |
ba3a79e7 | 350 | #define newSVpvs_share(str) Perl_newSVpvn_share(aTHX_ STR_WITH_LEN(str), 0) |
9dcc53ea Z |
351 | #define sv_catpvs_flags(sv, str, flags) \ |
352 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), flags) | |
353 | #define sv_catpvs_nomg(sv, str) \ | |
354 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), 0) | |
355 | #define sv_catpvs(sv, str) \ | |
356 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC) | |
357 | #define sv_catpvs_mg(sv, str) \ | |
358 | Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC|SV_SMAGIC) | |
3fe05580 | 359 | #define sv_setpvs(sv, str) Perl_sv_setpvn(aTHX_ sv, STR_WITH_LEN(str)) |
9dcc53ea Z |
360 | #define sv_setpvs_mg(sv, str) Perl_sv_setpvn_mg(aTHX_ sv, STR_WITH_LEN(str)) |
361 | #define sv_setref_pvs(rv, classname, str) \ | |
362 | Perl_sv_setref_pvn(aTHX_ rv, classname, STR_WITH_LEN(str)) | |
ba3a79e7 | 363 | #define savepvs(str) Perl_savepvn(aTHX_ STR_WITH_LEN(str)) |
9dcc53ea Z |
364 | #define savesharedpvs(str) Perl_savesharedpvn(aTHX_ STR_WITH_LEN(str)) |
365 | #define gv_stashpvs(str, create) \ | |
366 | Perl_gv_stashpvn(aTHX_ STR_WITH_LEN(str), create) | |
367 | #define gv_fetchpvs(namebeg, add, sv_type) \ | |
368 | Perl_gv_fetchpvn_flags(aTHX_ STR_WITH_LEN(namebeg), add, sv_type) | |
369 | #define gv_fetchpvn(namebeg, len, add, sv_type) \ | |
370 | Perl_gv_fetchpvn_flags(aTHX_ namebeg, len, add, sv_type) | |
371 | #define sv_catxmlpvs(dsv, str, utf8) \ | |
372 | Perl_sv_catxmlpvn(aTHX_ dsv, STR_WITH_LEN(str), utf8) | |
a038e571 NC |
373 | #define hv_fetchs(hv,key,lval) \ |
374 | ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \ | |
375 | (lval) ? (HV_FETCH_JUST_SV | HV_FETCH_LVALUE) \ | |
376 | : HV_FETCH_JUST_SV, NULL, 0)) | |
377 | ||
378 | #define hv_stores(hv,key,val) \ | |
379 | ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \ | |
380 | (HV_FETCH_ISSTORE|HV_FETCH_JUST_SV), (val), 0)) | |
ba3a79e7 | 381 | |
510966aa Z |
382 | #define lex_stuff_pvs(pv,flags) Perl_lex_stuff_pvn(aTHX_ STR_WITH_LEN(pv), flags) |
383 | ||
b96d8cd9 NC |
384 | #define get_cvs(str, flags) \ |
385 | Perl_get_cvn_flags(aTHX_ STR_WITH_LEN(str), (flags)) | |
5c1737d1 | 386 | |
954c1994 | 387 | /* |
ccfc67b7 JH |
388 | =head1 Miscellaneous Functions |
389 | ||
954c1994 GS |
390 | =for apidoc Am|bool|strNE|char* s1|char* s2 |
391 | Test two strings to see if they are different. Returns true or | |
392 | false. | |
393 | ||
394 | =for apidoc Am|bool|strEQ|char* s1|char* s2 | |
395 | Test two strings to see if they are equal. Returns true or false. | |
396 | ||
397 | =for apidoc Am|bool|strLT|char* s1|char* s2 | |
398 | Test two strings to see if the first, C<s1>, is less than the second, | |
399 | C<s2>. Returns true or false. | |
400 | ||
401 | =for apidoc Am|bool|strLE|char* s1|char* s2 | |
402 | Test two strings to see if the first, C<s1>, is less than or equal to the | |
403 | second, C<s2>. Returns true or false. | |
404 | ||
405 | =for apidoc Am|bool|strGT|char* s1|char* s2 | |
406 | Test two strings to see if the first, C<s1>, is greater than the second, | |
407 | C<s2>. Returns true or false. | |
408 | ||
409 | =for apidoc Am|bool|strGE|char* s1|char* s2 | |
410 | Test two strings to see if the first, C<s1>, is greater than or equal to | |
411 | the second, C<s2>. Returns true or false. | |
412 | ||
413 | =for apidoc Am|bool|strnNE|char* s1|char* s2|STRLEN len | |
414 | Test two strings to see if they are different. The C<len> parameter | |
415 | indicates the number of bytes to compare. Returns true or false. (A | |
416 | wrapper for C<strncmp>). | |
417 | ||
418 | =for apidoc Am|bool|strnEQ|char* s1|char* s2|STRLEN len | |
419 | Test two strings to see if they are equal. The C<len> parameter indicates | |
420 | the number of bytes to compare. Returns true or false. (A wrapper for | |
421 | C<strncmp>). | |
422 | ||
423 | =cut | |
424 | */ | |
425 | ||
8d063cd8 LW |
426 | #define strNE(s1,s2) (strcmp(s1,s2)) |
427 | #define strEQ(s1,s2) (!strcmp(s1,s2)) | |
428 | #define strLT(s1,s2) (strcmp(s1,s2) < 0) | |
429 | #define strLE(s1,s2) (strcmp(s1,s2) <= 0) | |
430 | #define strGT(s1,s2) (strcmp(s1,s2) > 0) | |
431 | #define strGE(s1,s2) (strcmp(s1,s2) >= 0) | |
432 | #define strnNE(s1,s2,l) (strncmp(s1,s2,l)) | |
433 | #define strnEQ(s1,s2,l) (!strncmp(s1,s2,l)) | |
378cc40b | 434 | |
36477c24 | 435 | #ifdef HAS_MEMCMP |
436 | # define memNE(s1,s2,l) (memcmp(s1,s2,l)) | |
437 | # define memEQ(s1,s2,l) (!memcmp(s1,s2,l)) | |
438 | #else | |
439 | # define memNE(s1,s2,l) (bcmp(s1,s2,l)) | |
440 | # define memEQ(s1,s2,l) (!bcmp(s1,s2,l)) | |
441 | #endif | |
442 | ||
568a785a | 443 | #define memEQs(s1, l, s2) \ |
8e6497b6 | 444 | (sizeof(s2)-1 == l && memEQ(s1, ("" s2 ""), (sizeof(s2)-1))) |
568a785a NC |
445 | #define memNEs(s1, l, s2) !memEQs(s1, l, s2) |
446 | ||
bbce6d69 | 447 | /* |
448 | * Character classes. | |
449 | * | |
450 | * Unfortunately, the introduction of locales means that we | |
451 | * can't trust isupper(), etc. to tell the truth. And when | |
452 | * it comes to /\w+/ with tainting enabled, we *must* be able | |
453 | * to trust our character classes. | |
454 | * | |
455 | * Therefore, the default tests in the text of Perl will be | |
456 | * independent of locale. Any code that wants to depend on | |
457 | * the current locale will use the tests that begin with "lc". | |
458 | */ | |
459 | ||
2304df62 AD |
460 | #ifdef HAS_SETLOCALE /* XXX Is there a better test for this? */ |
461 | # ifndef CTYPE256 | |
462 | # define CTYPE256 | |
463 | # endif | |
464 | #endif | |
465 | ||
954c1994 | 466 | /* |
ccfc67b7 JH |
467 | |
468 | =head1 Character classes | |
8a58bdcf KW |
469 | There are three variants for all the functions in this section. The base ones |
470 | operate using the character set of the platform Perl is running on. The ones | |
471 | with an C<_A> suffix operate on the ASCII character set, and the ones with an | |
472 | C<_L1> suffix operate on the full Latin1 character set. All are unaffected by | |
473 | locale | |
474 | ||
475 | For ASCII platforms, the base function with no suffix and the one with the | |
476 | C<_A> suffix are identical. The function with the C<_L1> suffix imposes the | |
477 | Latin-1 character set onto the platform. That is, the code points that are | |
478 | ASCII are unaffected, since ASCII is a subset of Latin-1. But the non-ASCII | |
479 | code points are treated as if they are Latin-1 characters. For example, | |
480 | C<isSPACE_L1()> will return true when called with the code point 0xA0, which is | |
481 | the Latin-1 NO-BREAK SPACE. | |
482 | ||
483 | For EBCDIC platforms, the base function with no suffix and the one with the | |
484 | C<_L1> suffix should be identical, since, as of this writing, the EBCDIC code | |
485 | pages that Perl knows about all are equivalent to Latin-1. The function that | |
486 | ends in an C<_A> suffix will not return true unless the specified character also | |
487 | has an ASCII equivalent. | |
ccfc67b7 | 488 | |
954c1994 | 489 | =for apidoc Am|bool|isALPHA|char ch |
2787a470 | 490 | Returns a boolean indicating whether the specified character is an |
06dc89af | 491 | alphabetic character in the platform's native character set. |
8a58bdcf KW |
492 | See the L<top of this section|/Character classes> for an explanation of variants |
493 | C<isALPHA_A> and C<isALPHA_L1>. | |
494 | ||
495 | =for apidoc Am|bool|isASCII|char ch | |
496 | Returns a boolean indicating whether the specified character is one of the 128 | |
497 | characters in the ASCII character set. On non-ASCII platforms, it is if this | |
498 | character corresponds to an ASCII character. Variants C<isASCII_A()> and | |
499 | C<isASCII_L1()> are identical to C<isASCII()>. | |
954c1994 | 500 | |
954c1994 | 501 | =for apidoc Am|bool|isDIGIT|char ch |
2787a470 | 502 | Returns a boolean indicating whether the specified character is a |
06dc89af | 503 | digit in the platform's native character set. |
8a58bdcf | 504 | Variants C<isDIGIT_A> and C<isDIGIT_L1> are identical to C<isDIGIT>. |
954c1994 | 505 | |
0c82b6df | 506 | =for apidoc Am|bool|isLOWER|char ch |
2787a470 | 507 | Returns a boolean indicating whether the specified character is a |
0c82b6df | 508 | lowercase character in the platform's native character set. |
8a58bdcf KW |
509 | See the L<top of this section|/Character classes> for an explanation of variants |
510 | C<isLOWER_A> and C<isLOWER_L1>. | |
0c82b6df | 511 | |
c99e91e9 | 512 | =for apidoc Am|bool|isOCTAL|char ch |
2787a470 | 513 | Returns a boolean indicating whether the specified character is an |
06dc89af | 514 | octal digit, [0-7] in the platform's native character set. |
8a58bdcf | 515 | Variants C<isOCTAL_A> and C<isOCTAL_L1> are identical to C<isOCTAL>. |
c99e91e9 | 516 | |
0c82b6df | 517 | =for apidoc Am|bool|isSPACE|char ch |
2787a470 | 518 | Returns a boolean indicating whether the specified character is a |
8a58bdcf KW |
519 | whitespace character in the platform's native character set. This is the same |
520 | as what C<\s> matches in a regular expression. | |
521 | See the L<top of this section|/Character classes> for an explanation of variants | |
522 | C<isSPACE_A> and C<isSPACE_L1>. | |
0c82b6df | 523 | |
954c1994 | 524 | =for apidoc Am|bool|isUPPER|char ch |
2787a470 | 525 | Returns a boolean indicating whether the specified character is an |
06dc89af | 526 | uppercase character in the platform's native character set. |
8a58bdcf KW |
527 | See the L<top of this section|/Character classes> for an explanation of variants |
528 | C<isUPPER_A> and C<isUPPER_L1>. | |
954c1994 | 529 | |
8eea39dd KW |
530 | =for apidoc Am|bool|isWORDCHAR|char ch |
531 | Returns a boolean indicating whether the specified character is a | |
532 | character that is any of: alphabetic, numeric, or an underscore. This is the | |
533 | same as what C<\w> matches in a regular expression. | |
534 | C<isALNUM()> is a synonym provided for backward compatibility. Note that it | |
535 | does not have the standard C language meaning of alphanumeric, since it matches | |
536 | an underscore and the standard meaning does not. | |
8a58bdcf KW |
537 | See the L<top of this section|/Character classes> for an explanation of variants |
538 | C<isWORDCHAR_A> and C<isWORDCHAR_L1>. | |
539 | ||
540 | =for apidoc Am|bool|isXDIGIT|char ch | |
541 | Returns a boolean indicating whether the specified character is a hexadecimal | |
542 | digit, [0-9A-Fa-f]. Variants C<isXDIGIT_A()> and C<isXDIGIT_L1()> are | |
543 | identical to C<isXDIGIT()>. | |
8eea39dd | 544 | |
e7c1e6c1 KW |
545 | =head1 Character case changing |
546 | ||
954c1994 | 547 | =for apidoc Am|char|toUPPER|char ch |
06dc89af KW |
548 | Converts the specified character to uppercase in the platform's native |
549 | character set, if possible; otherwise returns the input character itself. | |
954c1994 GS |
550 | |
551 | =for apidoc Am|char|toLOWER|char ch | |
06dc89af KW |
552 | Converts the specified character to lowercase in the platform's native |
553 | character set, if possible; otherwise returns the input character itself. | |
954c1994 GS |
554 | |
555 | =cut | |
353c9b6f | 556 | |
8a58bdcf | 557 | Note that these macros are repeated in Devel::PPPort, so should also be |
62fa66b6 KW |
558 | patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc |
559 | ||
954c1994 GS |
560 | */ |
561 | ||
cf301eb7 KW |
562 | /* FITS_IN_8_BITS(c) returns true if c occupies no more than 8 bits. It is |
563 | * designed to be hopefully bomb-proof, making sure that no bits of | |
564 | * information are lost even on a 64-bit machine, but to get the compiler to | |
565 | * optimize it out if possible. This is because Configure makes sure that the | |
566 | * machine has an 8-bit byte, so if c is stored in a byte, the sizeof() | |
567 | * guarantees that this evaluates to a constant true at compile time. The use | |
568 | * of the mask instead of '< 256' keeps gcc from complaining that it is alway | |
d0dcc402 KW |
569 | * true, when c's storage class is a byte. Use U64TYPE because U64 is known |
570 | * only in the perl core, and this macro can be called from outside that */ | |
cf301eb7 | 571 | #ifdef HAS_QUAD |
d0dcc402 | 572 | # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U64TYPE)(c) & 0xFF) == (U64TYPE)(c))) |
cf301eb7 KW |
573 | #else |
574 | # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U32)(c) & 0xFF) == (U32)(c))) | |
575 | #endif | |
576 | ||
ec79cfbb KW |
577 | #define isASCII(c) (FITS_IN_8_BITS(c) ? NATIVE_TO_UNI((U8) c) <= 127 : 0) |
578 | #define isASCII_A(c) isASCII(c) | |
6515cb76 | 579 | |
ec79cfbb | 580 | /* ASCII range only */ |
9fb1bf9d KW |
581 | #ifdef H_PERL /* If have access to perl.h, lookup in its table */ |
582 | # include "l1_char_class_tab.h" | |
583 | # define isALNUMC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_A)) | |
584 | # define isALPHA_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_A)) | |
585 | # define isBLANK_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_A)) | |
586 | # define isCNTRL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_A)) | |
587 | # define isDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_DIGIT_A)) | |
588 | # define isGRAPH_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_A)) | |
589 | # define isIDFIRST_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_A)) | |
590 | # define isLOWER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_A)) | |
591 | # define isOCTAL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_OCTAL_A)) | |
592 | # define isPRINT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_A)) | |
593 | # define isPSXSPC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_A)) | |
594 | # define isPUNCT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_A)) | |
595 | # define isSPACE_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_A)) | |
596 | # define isUPPER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_A)) | |
597 | # define isWORDCHAR_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_A)) | |
598 | # define isXDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_XDIGIT_A)) | |
e2e75538 | 599 | # 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)) |
9fb1bf9d KW |
600 | #else /* No perl.h. */ |
601 | # define isOCTAL_A(c) ((c) >= '0' && (c) <= '9') | |
602 | # ifdef EBCDIC | |
603 | # define isALNUMC_A(c) (isASCII(c) && isALNUMC(c)) | |
604 | # define isALPHA_A(c) (isASCII(c) && isALPHA(c)) | |
605 | # define isBLANK_A(c) (isASCII(c) && isBLANK(c)) | |
606 | # define isCNTRL_A(c) (isASCII(c) && isCNTRL(c)) | |
607 | # define isDIGIT_A(c) (isASCII(c) && isDIGIT(c)) | |
608 | # define isGRAPH_A(c) (isASCII(c) && isGRAPH(c)) | |
609 | # define isIDFIRST_A(c) (isASCII(c) && isIDFIRST(c)) | |
610 | # define isLOWER_A(c) (isASCII(c) && isLOWER(c)) | |
611 | # define isPRINT_A(c) (isASCII(c) && isPRINT(c)) | |
612 | # define isPSXSPC_A(c) (isASCII(c) && isPSXSPC(c)) | |
613 | # define isPUNCT_A(c) (isASCII(c) && isPUNCT(c)) | |
614 | # define isSPACE_A(c) (isASCII(c) && isSPACE(c)) | |
615 | # define isUPPER_A(c) (isASCII(c) && isUPPER(c)) | |
616 | # define isWORDCHAR_A(c) (isASCII(c) && isWORDCHAR(c)) | |
617 | # define isXDIGIT_A(c) (isASCII(c) && isXDIGIT(c)) | |
618 | # else /* ASCII platform, no perl.h */ | |
619 | # define isALNUMC_A(c) (isALPHA_A(c) || isDIGIT_A(c)) | |
620 | # define isALPHA_A(c) (isUPPER_A(c) || isLOWER_A(c)) | |
621 | # define isBLANK_A(c) ((c) == ' ' || (c) == '\t') | |
622 | # define isCNTRL_A(c) (FITS_IN_8_BITS(c) ? ((U8) (c) < ' ' || (c) == 127) : 0) | |
623 | # define isDIGIT_A(c) ((c) >= '0' && (c) <= '9') | |
624 | # define isGRAPH_A(c) (isWORDCHAR_A(c) || isPUNCT_A(c)) | |
625 | # define isIDFIRST_A(c) (isALPHA_A(c) || (c) == '_') | |
626 | # define isLOWER_A(c) ((c) >= 'a' && (c) <= 'z') | |
627 | # define isPRINT_A(c) (((c) >= 32 && (c) < 127)) | |
628 | # define isPSXSPC_A(c) (isSPACE_A(c) || (c) == '\v') | |
629 | # define isPUNCT_A(c) (((c) >= 33 && (c) <= 47) || ((c) >= 58 && (c) <= 64) || ((c) >= 91 && (c) <= 96) || ((c) >= 123 && (c) <= 126)) | |
630 | # define isSPACE_A(c) ((c) == ' ' || (c) == '\t' || (c) == '\n' || (c) =='\r' || (c) == '\f') | |
631 | # define isUPPER_A(c) ((c) >= 'A' && (c) <= 'Z') | |
632 | # define isWORDCHAR_A(c) (isALPHA_A(c) || isDIGIT_A(c) || (c) == '_') | |
633 | # define isXDIGIT_A(c) (isDIGIT_A(c) || ((c) >= 'a' && (c) <= 'f') || ((c) >= 'A' && (c) <= 'F')) | |
634 | # endif | |
635 | #endif /* ASCII range definitions */ | |
ec79cfbb KW |
636 | |
637 | /* Latin1 definitions */ | |
8a58bdcf KW |
638 | #ifdef H_PERL |
639 | # define isALNUMC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_L1)) | |
640 | # define isALPHA_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_L1)) | |
641 | # define isBLANK_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_L1)) | |
642 | /* continuation character for legal NAME in \N{NAME} */ | |
643 | # define isCHARNAME_CONT(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CHARNAME_CONT)) | |
644 | # define isCNTRL_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_L1)) | |
645 | # define isGRAPH_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_L1)) | |
646 | # define isIDFIRST_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_L1)) | |
647 | # define isLOWER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_L1)) | |
648 | # define isPRINT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_L1)) | |
649 | # define isPSXSPC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_L1)) | |
650 | # define isPUNCT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_L1)) | |
651 | # define isSPACE_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_L1)) | |
652 | # define isUPPER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_L1)) | |
653 | # define isWORDCHAR_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_L1)) | |
654 | #else /* No access to perl.h. Only a few provided here, just in case needed | |
655 | * for backwards compatibility */ | |
656 | /* ALPHAU includes Unicode semantics for latin1 characters. It has an extra | |
657 | * >= AA test to speed up ASCII-only tests at the expense of the others */ | |
658 | # define isALPHA_L1(c) (isALPHA(c) || (NATIVE_TO_UNI((U8) c) >= 0xAA \ | |
659 | && ((NATIVE_TO_UNI((U8) c) >= 0xC0 \ | |
660 | && NATIVE_TO_UNI((U8) c) != 0xD7 && NATIVE_TO_UNI((U8) c) != 0xF7) \ | |
661 | || NATIVE_TO_UNI((U8) c) == 0xAA \ | |
662 | || NATIVE_TO_UNI((U8) c) == 0xB5 \ | |
663 | || NATIVE_TO_UNI((U8) c) == 0xBA))) | |
664 | # define isCHARNAME_CONT(c) (isALNUM_L1(c) || (c) == ' ' || (c) == '-' || (c) == '(' || (c) == ')' || (c) == ':' || NATIVE_TO_UNI((U8) c) == 0xA0) | |
665 | #endif | |
666 | ||
667 | /* Macros for backwards compatibility and for completeness when the ASCII and | |
668 | * Latin1 values are identical */ | |
ec79cfbb KW |
669 | #define isALNUM(c) isWORDCHAR(c) |
670 | #define isALNUMU(c) isWORDCHAR_L1(c) | |
8a58bdcf KW |
671 | #define isALPHAU(c) isALPHA_L1(c) |
672 | #define isDIGIT_L1(c) isDIGIT_A(c) | |
673 | #define isOCTAL(c) isOCTAL_A(c) | |
674 | #define isOCTAL_L1(c) isOCTAL_A(c) | |
675 | #define isXDIGIT_L1(c) isXDIGIT_A(c) | |
676 | ||
677 | /* Macros that differ between EBCDIC and ASCII. Where C89 defines a function, | |
678 | * that is used in the EBCDIC form, because in EBCDIC we do not do locales: | |
679 | * therefore can use native functions. For those where C89 doesn't define a | |
680 | * function, use our function, assuming that the EBCDIC code page is isomorphic | |
681 | * with Latin1, which the three currently recognized by Perl are. Some libc's | |
682 | * have an isblank(), but it's not guaranteed. */ | |
9d116dd7 | 683 | #ifdef EBCDIC |
b8c5462f | 684 | # define isALNUMC(c) isalnum(c) |
271b5cb3 | 685 | # define isALPHA(c) isalpha(c) |
0bc70f7f | 686 | # define isBLANK(c) ((c) == ' ' || (c) == '\t' || NATIVE_TO_UNI(c) == 0xA0) |
b8c5462f | 687 | # define isCNTRL(c) iscntrl(c) |
9916188f | 688 | # define isDIGIT(c) isdigit(c) |
b8c5462f | 689 | # define isGRAPH(c) isgraph(c) |
8a58bdcf | 690 | # define isIDFIRST(c) (isALPHA(c) || (c) == '_') |
7d79683e | 691 | # define isLOWER(c) islower(c) |
9d116dd7 | 692 | # define isPRINT(c) isprint(c) |
56e4b611 | 693 | # define isPSXSPC(c) isspace(c) |
b8c5462f | 694 | # define isPUNCT(c) ispunct(c) |
174b5e76 | 695 | # define isSPACE(c) (isPSXSPC(c) && (c) != '\v') |
7d79683e | 696 | # define isUPPER(c) isupper(c) |
b8c5462f | 697 | # define isXDIGIT(c) isxdigit(c) |
8a58bdcf | 698 | # define isWORDCHAR(c) (isalnum(c) || (c) == '_') |
9d116dd7 | 699 | # define toLOWER(c) tolower(c) |
7d79683e | 700 | # define toUPPER(c) toupper(c) |
ec79cfbb KW |
701 | #else /* Not EBCDIC: ASCII-only matching */ |
702 | # define isALNUMC(c) isALNUMC_A(c) | |
271b5cb3 | 703 | # define isALPHA(c) isALPHA_A(c) |
ec79cfbb KW |
704 | # define isBLANK(c) isBLANK_A(c) |
705 | # define isCNTRL(c) isCNTRL_A(c) | |
706 | # define isDIGIT(c) isDIGIT_A(c) | |
707 | # define isGRAPH(c) isGRAPH_A(c) | |
8a58bdcf | 708 | # define isIDFIRST(c) isIDFIRST_A(c) |
ec79cfbb KW |
709 | # define isLOWER(c) isLOWER_A(c) |
710 | # define isPRINT(c) isPRINT_A(c) | |
711 | # define isPSXSPC(c) isPSXSPC_A(c) | |
712 | # define isPUNCT(c) isPUNCT_A(c) | |
713 | # define isSPACE(c) isSPACE_A(c) | |
714 | # define isUPPER(c) isUPPER_A(c) | |
8a58bdcf | 715 | # define isWORDCHAR(c) isWORDCHAR_A(c) |
ec79cfbb | 716 | # define isXDIGIT(c) isXDIGIT_A(c) |
00f254e2 | 717 | |
8a58bdcf KW |
718 | /* ASCII casing. These could also be written as |
719 | #define toLOWER(c) (isASCII(c) ? toLOWER_LATIN1(c) : (c)) | |
720 | #define toUPPER(c) (isASCII(c) ? toUPPER_LATIN1_MOD(c) : (c)) | |
721 | which uses table lookup and mask instead of subtraction. (This would | |
722 | work because the _MOD does not apply in the ASCII range) */ | |
9d116dd7 | 723 | # define toLOWER(c) (isUPPER(c) ? (c) + ('a' - 'A') : (c)) |
7d79683e | 724 | # define toUPPER(c) (isLOWER(c) ? (c) - ('a' - 'A') : (c)) |
9d116dd7 | 725 | #endif |
bbce6d69 | 726 | |
1a0901db KW |
727 | |
728 | /* Use table lookup for speed; return error character for input | |
729 | * out-of-range */ | |
730 | #define toLOWER_LATIN1(c) (FITS_IN_8_BITS(c) \ | |
731 | ? UNI_TO_NATIVE(PL_latin1_lc[ \ | |
732 | NATIVE_TO_UNI( (U8) (c)) ]) \ | |
733 | : UNICODE_REPLACEMENT) | |
734 | /* Modified uc. Is correct uc except for three non-ascii chars which are | |
735 | * all mapped to one of them, and these need special handling; error | |
736 | * character for input out-of-range */ | |
737 | #define toUPPER_LATIN1_MOD(c) (FITS_IN_8_BITS(c) \ | |
738 | ? UNI_TO_NATIVE(PL_mod_latin1_uc[ \ | |
739 | NATIVE_TO_UNI( (U8) (c)) ]) \ | |
740 | : UNICODE_REPLACEMENT) | |
741 | ||
bbce6d69 | 742 | #ifdef USE_NEXT_CTYPE |
743 | ||
744 | # define isALNUM_LC(c) \ | |
37bd1396 | 745 | (NXIsAlNum((unsigned int)(c)) || (char)(c) == '_') |
ff68c719 | 746 | # define isIDFIRST_LC(c) \ |
747 | (NXIsAlpha((unsigned int)(c)) || (char)(c) == '_') | |
748 | # define isALPHA_LC(c) NXIsAlpha((unsigned int)(c)) | |
749 | # define isSPACE_LC(c) NXIsSpace((unsigned int)(c)) | |
750 | # define isDIGIT_LC(c) NXIsDigit((unsigned int)(c)) | |
751 | # define isUPPER_LC(c) NXIsUpper((unsigned int)(c)) | |
752 | # define isLOWER_LC(c) NXIsLower((unsigned int)(c)) | |
37bd1396 | 753 | # define isALNUMC_LC(c) NXIsAlNum((unsigned int)(c)) |
b8c5462f JH |
754 | # define isCNTRL_LC(c) NXIsCntrl((unsigned int)(c)) |
755 | # define isGRAPH_LC(c) NXIsGraph((unsigned int)(c)) | |
ff68c719 | 756 | # define isPRINT_LC(c) NXIsPrint((unsigned int)(c)) |
b8c5462f | 757 | # define isPUNCT_LC(c) NXIsPunct((unsigned int)(c)) |
ff68c719 | 758 | # define toUPPER_LC(c) NXToUpper((unsigned int)(c)) |
759 | # define toLOWER_LC(c) NXToLower((unsigned int)(c)) | |
bbce6d69 | 760 | |
761 | #else /* !USE_NEXT_CTYPE */ | |
b8c5462f | 762 | |
bbce6d69 | 763 | # if defined(CTYPE256) || (!defined(isascii) && !defined(HAS_ISASCII)) |
764 | ||
b8c5462f | 765 | # define isALNUM_LC(c) (isalnum((unsigned char)(c)) || (char)(c) == '_') |
ff68c719 | 766 | # define isIDFIRST_LC(c) (isalpha((unsigned char)(c)) || (char)(c) == '_') |
bbce6d69 | 767 | # define isALPHA_LC(c) isalpha((unsigned char)(c)) |
768 | # define isSPACE_LC(c) isspace((unsigned char)(c)) | |
769 | # define isDIGIT_LC(c) isdigit((unsigned char)(c)) | |
770 | # define isUPPER_LC(c) isupper((unsigned char)(c)) | |
771 | # define isLOWER_LC(c) islower((unsigned char)(c)) | |
b8c5462f JH |
772 | # define isALNUMC_LC(c) isalnum((unsigned char)(c)) |
773 | # define isCNTRL_LC(c) iscntrl((unsigned char)(c)) | |
774 | # define isGRAPH_LC(c) isgraph((unsigned char)(c)) | |
bbce6d69 | 775 | # define isPRINT_LC(c) isprint((unsigned char)(c)) |
b8c5462f | 776 | # define isPUNCT_LC(c) ispunct((unsigned char)(c)) |
bbce6d69 | 777 | # define toUPPER_LC(c) toupper((unsigned char)(c)) |
778 | # define toLOWER_LC(c) tolower((unsigned char)(c)) | |
779 | ||
780 | # else | |
781 | ||
3609ea0d | 782 | # define isALNUM_LC(c) (isascii(c) && (isalnum(c) || (c) == '_')) |
bbce6d69 | 783 | # define isIDFIRST_LC(c) (isascii(c) && (isalpha(c) || (c) == '_')) |
784 | # define isALPHA_LC(c) (isascii(c) && isalpha(c)) | |
785 | # define isSPACE_LC(c) (isascii(c) && isspace(c)) | |
786 | # define isDIGIT_LC(c) (isascii(c) && isdigit(c)) | |
787 | # define isUPPER_LC(c) (isascii(c) && isupper(c)) | |
788 | # define isLOWER_LC(c) (isascii(c) && islower(c)) | |
b8c5462f JH |
789 | # define isALNUMC_LC(c) (isascii(c) && isalnum(c)) |
790 | # define isCNTRL_LC(c) (isascii(c) && iscntrl(c)) | |
791 | # define isGRAPH_LC(c) (isascii(c) && isgraph(c)) | |
bbce6d69 | 792 | # define isPRINT_LC(c) (isascii(c) && isprint(c)) |
b8c5462f | 793 | # define isPUNCT_LC(c) (isascii(c) && ispunct(c)) |
bbce6d69 | 794 | # define toUPPER_LC(c) toupper(c) |
795 | # define toLOWER_LC(c) tolower(c) | |
796 | ||
797 | # endif | |
a0d0e21e | 798 | #endif /* USE_NEXT_CTYPE */ |
55204971 | 799 | |
aaa51d5e JF |
800 | #define isPSXSPC_LC(c) (isSPACE_LC(c) || (c) == '\v') |
801 | #define isBLANK_LC(c) isBLANK(c) /* could be wrong */ | |
802 | ||
a0ed51b3 LW |
803 | #define isALNUM_uni(c) is_uni_alnum(c) |
804 | #define isIDFIRST_uni(c) is_uni_idfirst(c) | |
805 | #define isALPHA_uni(c) is_uni_alpha(c) | |
806 | #define isSPACE_uni(c) is_uni_space(c) | |
807 | #define isDIGIT_uni(c) is_uni_digit(c) | |
808 | #define isUPPER_uni(c) is_uni_upper(c) | |
809 | #define isLOWER_uni(c) is_uni_lower(c) | |
b8c5462f JH |
810 | #define isASCII_uni(c) is_uni_ascii(c) |
811 | #define isCNTRL_uni(c) is_uni_cntrl(c) | |
812 | #define isGRAPH_uni(c) is_uni_graph(c) | |
a0ed51b3 | 813 | #define isPRINT_uni(c) is_uni_print(c) |
b8c5462f JH |
814 | #define isPUNCT_uni(c) is_uni_punct(c) |
815 | #define isXDIGIT_uni(c) is_uni_xdigit(c) | |
a2a2844f JH |
816 | #define toUPPER_uni(c,s,l) to_uni_upper(c,s,l) |
817 | #define toTITLE_uni(c,s,l) to_uni_title(c,s,l) | |
818 | #define toLOWER_uni(c,s,l) to_uni_lower(c,s,l) | |
b8d68ded | 819 | #define toFOLD_uni(c,s,l) to_uni_fold(c,s,l) |
a0ed51b3 | 820 | |
aaa51d5e JF |
821 | #define isPSXSPC_uni(c) (isSPACE_uni(c) ||(c) == '\f') |
822 | #define isBLANK_uni(c) isBLANK(c) /* could be wrong */ | |
823 | ||
9041c2e3 NIS |
824 | #define isALNUM_LC_uvchr(c) (c < 256 ? isALNUM_LC(c) : is_uni_alnum_lc(c)) |
825 | #define isIDFIRST_LC_uvchr(c) (c < 256 ? isIDFIRST_LC(c) : is_uni_idfirst_lc(c)) | |
826 | #define isALPHA_LC_uvchr(c) (c < 256 ? isALPHA_LC(c) : is_uni_alpha_lc(c)) | |
827 | #define isSPACE_LC_uvchr(c) (c < 256 ? isSPACE_LC(c) : is_uni_space_lc(c)) | |
828 | #define isDIGIT_LC_uvchr(c) (c < 256 ? isDIGIT_LC(c) : is_uni_digit_lc(c)) | |
829 | #define isUPPER_LC_uvchr(c) (c < 256 ? isUPPER_LC(c) : is_uni_upper_lc(c)) | |
830 | #define isLOWER_LC_uvchr(c) (c < 256 ? isLOWER_LC(c) : is_uni_lower_lc(c)) | |
9041c2e3 NIS |
831 | #define isCNTRL_LC_uvchr(c) (c < 256 ? isCNTRL_LC(c) : is_uni_cntrl_lc(c)) |
832 | #define isGRAPH_LC_uvchr(c) (c < 256 ? isGRAPH_LC(c) : is_uni_graph_lc(c)) | |
833 | #define isPRINT_LC_uvchr(c) (c < 256 ? isPRINT_LC(c) : is_uni_print_lc(c)) | |
834 | #define isPUNCT_LC_uvchr(c) (c < 256 ? isPUNCT_LC(c) : is_uni_punct_lc(c)) | |
a0ed51b3 | 835 | |
aaa51d5e JF |
836 | #define isPSXSPC_LC_uni(c) (isSPACE_LC_uni(c) ||(c) == '\f') |
837 | #define isBLANK_LC_uni(c) isBLANK(c) /* could be wrong */ | |
838 | ||
a0ed51b3 | 839 | #define isALNUM_utf8(p) is_utf8_alnum(p) |
d7425188 FC |
840 | /* The ID_Start of Unicode was originally quite limiting: it assumed an |
841 | * L-class character (meaning that you could not have, say, a CJK charac- | |
842 | * ter). So, instead, perl has for a long time allowed ID_Continue but | |
843 | * not digits. | |
844 | * We still preserve that for backward compatibility. But we also make sure | |
845 | * that it is alphanumeric, so S_scan_word in toke.c will not hang. See | |
846 | * http://rt.perl.org/rt3/Ticket/Display.html?id=74022 | |
847 | * for more detail than you ever wanted to know about. */ | |
848 | #define isIDFIRST_utf8(p) \ | |
849 | (is_utf8_idcont(p) && !is_utf8_digit(p) && is_utf8_alnum(p)) | |
a0ed51b3 LW |
850 | #define isALPHA_utf8(p) is_utf8_alpha(p) |
851 | #define isSPACE_utf8(p) is_utf8_space(p) | |
852 | #define isDIGIT_utf8(p) is_utf8_digit(p) | |
853 | #define isUPPER_utf8(p) is_utf8_upper(p) | |
854 | #define isLOWER_utf8(p) is_utf8_lower(p) | |
b8c5462f JH |
855 | #define isASCII_utf8(p) is_utf8_ascii(p) |
856 | #define isCNTRL_utf8(p) is_utf8_cntrl(p) | |
857 | #define isGRAPH_utf8(p) is_utf8_graph(p) | |
a0ed51b3 | 858 | #define isPRINT_utf8(p) is_utf8_print(p) |
b8c5462f JH |
859 | #define isPUNCT_utf8(p) is_utf8_punct(p) |
860 | #define isXDIGIT_utf8(p) is_utf8_xdigit(p) | |
a2a2844f JH |
861 | #define toUPPER_utf8(p,s,l) to_utf8_upper(p,s,l) |
862 | #define toTITLE_utf8(p,s,l) to_utf8_title(p,s,l) | |
863 | #define toLOWER_utf8(p,s,l) to_utf8_lower(p,s,l) | |
a0ed51b3 | 864 | |
aaa51d5e JF |
865 | #define isPSXSPC_utf8(c) (isSPACE_utf8(c) ||(c) == '\f') |
866 | #define isBLANK_utf8(c) isBLANK(c) /* could be wrong */ | |
867 | ||
9041c2e3 NIS |
868 | #define isALNUM_LC_utf8(p) isALNUM_LC_uvchr(utf8_to_uvchr(p, 0)) |
869 | #define isIDFIRST_LC_utf8(p) isIDFIRST_LC_uvchr(utf8_to_uvchr(p, 0)) | |
870 | #define isALPHA_LC_utf8(p) isALPHA_LC_uvchr(utf8_to_uvchr(p, 0)) | |
871 | #define isSPACE_LC_utf8(p) isSPACE_LC_uvchr(utf8_to_uvchr(p, 0)) | |
872 | #define isDIGIT_LC_utf8(p) isDIGIT_LC_uvchr(utf8_to_uvchr(p, 0)) | |
873 | #define isUPPER_LC_utf8(p) isUPPER_LC_uvchr(utf8_to_uvchr(p, 0)) | |
874 | #define isLOWER_LC_utf8(p) isLOWER_LC_uvchr(utf8_to_uvchr(p, 0)) | |
875 | #define isALNUMC_LC_utf8(p) isALNUMC_LC_uvchr(utf8_to_uvchr(p, 0)) | |
876 | #define isCNTRL_LC_utf8(p) isCNTRL_LC_uvchr(utf8_to_uvchr(p, 0)) | |
877 | #define isGRAPH_LC_utf8(p) isGRAPH_LC_uvchr(utf8_to_uvchr(p, 0)) | |
878 | #define isPRINT_LC_utf8(p) isPRINT_LC_uvchr(utf8_to_uvchr(p, 0)) | |
879 | #define isPUNCT_LC_utf8(p) isPUNCT_LC_uvchr(utf8_to_uvchr(p, 0)) | |
a0ed51b3 | 880 | |
aaa51d5e JF |
881 | #define isPSXSPC_LC_utf8(c) (isSPACE_LC_utf8(c) ||(c) == '\f') |
882 | #define isBLANK_LC_utf8(c) isBLANK(c) /* could be wrong */ | |
883 | ||
6ae9f32a KW |
884 | /* This conversion works both ways, strangely enough. On EBCDIC platforms, |
885 | * CTRL-@ is 0, CTRL-A is 1, etc, just like on ASCII */ | |
886 | # define toCTRL(c) (toUPPER(NATIVE_TO_UNI(c)) ^ 64) | |
bbce6d69 | 887 | |
dea28490 JJ |
888 | /* Line numbers are unsigned, 32 bits. */ |
889 | typedef U32 line_t; | |
dea28490 | 890 | #define NOLINE ((line_t) 4294967295UL) |
378cc40b | 891 | |
91152fc1 DG |
892 | /* Helpful alias for version prescan */ |
893 | #define is_LAX_VERSION(a,b) \ | |
894 | (a != Perl_prescan_version(aTHX_ a, FALSE, b, NULL, NULL, NULL, NULL)) | |
895 | ||
896 | #define is_STRICT_VERSION(a,b) \ | |
897 | (a != Perl_prescan_version(aTHX_ a, TRUE, b, NULL, NULL, NULL, NULL)) | |
898 | ||
899 | #define BADVERSION(a,b,c) \ | |
900 | if (b) { \ | |
901 | *b = c; \ | |
902 | } \ | |
903 | return a; | |
8c52afec | 904 | |
8e84507e | 905 | /* |
ccfc67b7 JH |
906 | =head1 Memory Management |
907 | ||
a02a5408 | 908 | =for apidoc Am|void|Newx|void* ptr|int nitems|type |
954c1994 GS |
909 | The XSUB-writer's interface to the C C<malloc> function. |
910 | ||
c5008215 JC |
911 | In 5.9.3, Newx() and friends replace the older New() API, and drops |
912 | the first parameter, I<x>, a debug aid which allowed callers to identify | |
37b8b4c9 | 913 | themselves. This aid has been superseded by a new build option, |
c5008215 JC |
914 | PERL_MEM_LOG (see L<perlhack/PERL_MEM_LOG>). The older API is still |
915 | there for use in XS modules supporting older perls. | |
916 | ||
a02a5408 | 917 | =for apidoc Am|void|Newxc|void* ptr|int nitems|type|cast |
954c1994 | 918 | The XSUB-writer's interface to the C C<malloc> function, with |
c5008215 | 919 | cast. See also C<Newx>. |
954c1994 | 920 | |
a02a5408 | 921 | =for apidoc Am|void|Newxz|void* ptr|int nitems|type |
954c1994 | 922 | The XSUB-writer's interface to the C C<malloc> function. The allocated |
c5008215 | 923 | memory is zeroed with C<memzero>. See also C<Newx>. |
a02a5408 | 924 | |
954c1994 GS |
925 | =for apidoc Am|void|Renew|void* ptr|int nitems|type |
926 | The XSUB-writer's interface to the C C<realloc> function. | |
927 | ||
928 | =for apidoc Am|void|Renewc|void* ptr|int nitems|type|cast | |
929 | The XSUB-writer's interface to the C C<realloc> function, with | |
930 | cast. | |
931 | ||
49b8b560 | 932 | =for apidoc Am|void|Safefree|void* ptr |
954c1994 GS |
933 | The XSUB-writer's interface to the C C<free> function. |
934 | ||
935 | =for apidoc Am|void|Move|void* src|void* dest|int nitems|type | |
936 | The XSUB-writer's interface to the C C<memmove> function. The C<src> is the | |
937 | source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is | |
938 | the type. Can do overlapping moves. See also C<Copy>. | |
939 | ||
e90e2364 NC |
940 | =for apidoc Am|void *|MoveD|void* src|void* dest|int nitems|type |
941 | Like C<Move> but returns dest. Useful for encouraging compilers to tail-call | |
942 | optimise. | |
943 | ||
954c1994 GS |
944 | =for apidoc Am|void|Copy|void* src|void* dest|int nitems|type |
945 | The XSUB-writer's interface to the C C<memcpy> function. The C<src> is the | |
946 | source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is | |
947 | the type. May fail on overlapping copies. See also C<Move>. | |
948 | ||
e90e2364 NC |
949 | =for apidoc Am|void *|CopyD|void* src|void* dest|int nitems|type |
950 | ||
951 | Like C<Copy> but returns dest. Useful for encouraging compilers to tail-call | |
952 | optimise. | |
953 | ||
954c1994 GS |
954 | =for apidoc Am|void|Zero|void* dest|int nitems|type |
955 | ||
956 | The XSUB-writer's interface to the C C<memzero> function. The C<dest> is the | |
957 | destination, C<nitems> is the number of items, and C<type> is the type. | |
958 | ||
e90e2364 NC |
959 | =for apidoc Am|void *|ZeroD|void* dest|int nitems|type |
960 | ||
961 | Like C<Zero> but returns dest. Useful for encouraging compilers to tail-call | |
962 | optimise. | |
963 | ||
954c1994 | 964 | =for apidoc Am|void|StructCopy|type src|type dest|type |
4375e838 | 965 | This is an architecture-independent macro to copy one structure to another. |
954c1994 | 966 | |
7e337ee0 JH |
967 | =for apidoc Am|void|PoisonWith|void* dest|int nitems|type|U8 byte |
968 | ||
969 | Fill up memory with a byte pattern (a byte repeated over and over | |
970 | again) that hopefully catches attempts to access uninitialized memory. | |
971 | ||
972 | =for apidoc Am|void|PoisonNew|void* dest|int nitems|type | |
973 | ||
974 | PoisonWith(0xAB) for catching access to allocated but uninitialized memory. | |
975 | ||
1c12ffb4 | 976 | =for apidoc Am|void|PoisonFree|void* dest|int nitems|type |
7e337ee0 JH |
977 | |
978 | PoisonWith(0xEF) for catching access to freed memory. | |
979 | ||
9965345d JH |
980 | =for apidoc Am|void|Poison|void* dest|int nitems|type |
981 | ||
7e337ee0 | 982 | PoisonWith(0xEF) for catching access to freed memory. |
9965345d JH |
983 | |
984 | =cut */ | |
954c1994 | 985 | |
561b68a9 SH |
986 | /* Maintained for backwards-compatibility only. Use newSV() instead. */ |
987 | #ifndef PERL_CORE | |
ff06c60c | 988 | #define NEWSV(x,len) newSV(len) |
561b68a9 | 989 | #endif |
ff06c60c | 990 | |
19a94d75 JH |
991 | #define MEM_SIZE_MAX ((MEM_SIZE)~0) |
992 | ||
2720f793 JH |
993 | /* The +0.0 in MEM_WRAP_CHECK_ is an attempt to foil |
994 | * overly eager compilers that will bleat about e.g. | |
995 | * (U16)n > (size_t)~0/sizeof(U16) always being false. */ | |
27d5b266 | 996 | #ifdef PERL_MALLOC_WRAP |
fabdb6c0 | 997 | #define MEM_WRAP_CHECK(n,t) MEM_WRAP_CHECK_1(n,t,PL_memory_wrap) |
27d5b266 | 998 | #define MEM_WRAP_CHECK_1(n,t,a) \ |
f1f66076 | 999 | (void)(sizeof(t) > 1 && ((MEM_SIZE)(n)+0.0) > MEM_SIZE_MAX/sizeof(t) && (Perl_croak_nocontext("%s",(a)),0)) |
8b44ba4c | 1000 | #define MEM_WRAP_CHECK_(n,t) MEM_WRAP_CHECK(n,t), |
27d5b266 | 1001 | |
f1f66076 | 1002 | #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 | 1003 | |
27d5b266 JH |
1004 | #else |
1005 | ||
410319be NC |
1006 | #define MEM_WRAP_CHECK(n,t) |
1007 | #define MEM_WRAP_CHECK_1(n,t,a) | |
1008 | #define MEM_WRAP_CHECK_2(n,t,a,b) | |
8b44ba4c NC |
1009 | #define MEM_WRAP_CHECK_(n,t) |
1010 | ||
6fff79ce | 1011 | #define PERL_STRLEN_ROUNDUP(n) (((n-1+PERL_STRLEN_ROUNDUP_QUANTUM)&~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM-1))) |
27d5b266 | 1012 | |
1936d2a7 | 1013 | #endif |
8b44ba4c | 1014 | |
fe4f188c | 1015 | #ifdef PERL_MEM_LOG |
46c6c7e2 | 1016 | /* |
9f653bb5 | 1017 | * If PERL_MEM_LOG is defined, all Newx()s, Renew()s, and Safefree()s |
46c6c7e2 JH |
1018 | * go through functions, which are handy for debugging breakpoints, but |
1019 | * which more importantly get the immediate calling environment (file and | |
e352bcff JH |
1020 | * line number, and C function name if available) passed in. This info can |
1021 | * then be used for logging the calls, for which one gets a sample | |
73d1d973 | 1022 | * implementation unless -DPERL_MEM_LOG_NOIMPL is also defined. |
3609ea0d | 1023 | * |
46c6c7e2 | 1024 | * Known problems: |
94e892a6 | 1025 | * - not all memory allocs get logged, only those |
46c6c7e2 | 1026 | * that go through Newx() and derivatives (while all |
94e892a6 | 1027 | * Safefrees do get logged) |
46c6c7e2 JH |
1028 | * - __FILE__ and __LINE__ do not work everywhere |
1029 | * - __func__ or __FUNCTION__ even less so | |
1030 | * - I think more goes on after the perlio frees but | |
1031 | * the thing is that STDERR gets closed (as do all | |
1032 | * the file descriptors) | |
1033 | * - no deeper calling stack than the caller of the Newx() | |
1034 | * or the kind, but do I look like a C reflection/introspection | |
1035 | * utility to you? | |
1036 | * - the function prototypes for the logging functions | |
1037 | * probably should maybe be somewhere else than handy.h | |
1038 | * - one could consider inlining (macrofying) the logging | |
1039 | * for speed, but I am too lazy | |
1040 | * - one could imagine recording the allocations in a hash, | |
1041 | * (keyed by the allocation address?), and maintain that | |
1042 | * through reallocs and frees, but how to do that without | |
1043 | * any News() happening...? | |
73d1d973 | 1044 | * - lots of -Ddefines to get useful/controllable output |
b953482e | 1045 | * - lots of ENV reads |
46c6c7e2 JH |
1046 | */ |
1047 | ||
12754f92 | 1048 | 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 | 1049 | |
12754f92 | 1050 | 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 | 1051 | |
12754f92 | 1052 | PERL_EXPORT_C Malloc_t Perl_mem_log_free(Malloc_t oldalloc, const char *filename, const int linenumber, const char *funcname); |
46c6c7e2 | 1053 | |
0b0ab801 | 1054 | # ifdef PERL_CORE |
73d1d973 | 1055 | # ifndef PERL_MEM_LOG_NOIMPL |
0b0ab801 MHM |
1056 | enum mem_log_type { |
1057 | MLT_ALLOC, | |
1058 | MLT_REALLOC, | |
d7a2c63c MHM |
1059 | MLT_FREE, |
1060 | MLT_NEW_SV, | |
1061 | MLT_DEL_SV | |
0b0ab801 MHM |
1062 | }; |
1063 | # endif | |
12754f92 | 1064 | # if defined(PERL_IN_SV_C) /* those are only used in sv.c */ |
d7a2c63c MHM |
1065 | void Perl_mem_log_new_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname); |
1066 | void Perl_mem_log_del_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname); | |
12754f92 | 1067 | # endif |
0b0ab801 MHM |
1068 | # endif |
1069 | ||
fe4f188c JH |
1070 | #endif |
1071 | ||
1072 | #ifdef PERL_MEM_LOG | |
d1401ee9 MHM |
1073 | #define MEM_LOG_ALLOC(n,t,a) Perl_mem_log_alloc(n,sizeof(t),STRINGIFY(t),a,__FILE__,__LINE__,FUNCTION__) |
1074 | #define MEM_LOG_REALLOC(n,t,v,a) Perl_mem_log_realloc(n,sizeof(t),STRINGIFY(t),v,a,__FILE__,__LINE__,FUNCTION__) | |
46c6c7e2 | 1075 | #define MEM_LOG_FREE(a) Perl_mem_log_free(a,__FILE__,__LINE__,FUNCTION__) |
fe4f188c JH |
1076 | #endif |
1077 | ||
1078 | #ifndef MEM_LOG_ALLOC | |
1079 | #define MEM_LOG_ALLOC(n,t,a) (a) | |
1080 | #endif | |
1081 | #ifndef MEM_LOG_REALLOC | |
1082 | #define MEM_LOG_REALLOC(n,t,v,a) (a) | |
1083 | #endif | |
1084 | #ifndef MEM_LOG_FREE | |
1085 | #define MEM_LOG_FREE(a) (a) | |
1086 | #endif | |
1087 | ||
d1401ee9 MHM |
1088 | #define Newx(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t)))))) |
1089 | #define Newxc(v,n,t,c) (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t)))))) | |
1090 | #define Newxz(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safecalloc((n),sizeof(t))))) | |
a6f6820f NC |
1091 | |
1092 | #ifndef PERL_CORE | |
a02a5408 JC |
1093 | /* pre 5.9.x compatibility */ |
1094 | #define New(x,v,n,t) Newx(v,n,t) | |
1095 | #define Newc(x,v,n,t,c) Newxc(v,n,t,c) | |
4541904d | 1096 | #define Newz(x,v,n,t) Newxz(v,n,t) |
a6f6820f | 1097 | #endif |
a02a5408 | 1098 | |
ff68c719 | 1099 | #define Renew(v,n,t) \ |
d1401ee9 | 1100 | (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t)))))) |
ff68c719 | 1101 | #define Renewc(v,n,t,c) \ |
d1401ee9 | 1102 | (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t)))))) |
94010e71 NC |
1103 | |
1104 | #ifdef PERL_POISON | |
1105 | #define Safefree(d) \ | |
06c0cc96 | 1106 | ((d) ? (void)(safefree(MEM_LOG_FREE((Malloc_t)(d))), Poison(&(d), 1, Malloc_t)) : (void) 0) |
94010e71 | 1107 | #else |
fe4f188c | 1108 | #define Safefree(d) safefree(MEM_LOG_FREE((Malloc_t)(d))) |
94010e71 | 1109 | #endif |
55497cff | 1110 | |
8b44ba4c NC |
1111 | #define Move(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memmove((char*)(d),(const char*)(s), (n) * sizeof(t))) |
1112 | #define Copy(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memcpy((char*)(d),(const char*)(s), (n) * sizeof(t))) | |
1113 | #define Zero(d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memzero((char*)(d), (n) * sizeof(t))) | |
55497cff | 1114 | |
8b44ba4c NC |
1115 | #define MoveD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memmove((char*)(d),(const char*)(s), (n) * sizeof(t))) |
1116 | #define CopyD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memcpy((char*)(d),(const char*)(s), (n) * sizeof(t))) | |
e90e2364 | 1117 | #ifdef HAS_MEMSET |
8b44ba4c | 1118 | #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t))) |
e90e2364 | 1119 | #else |
8b44ba4c NC |
1120 | /* Using bzero(), which returns void. */ |
1121 | #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t)),d) | |
e90e2364 NC |
1122 | #endif |
1123 | ||
7e337ee0 JH |
1124 | #define PoisonWith(d,n,t,b) (MEM_WRAP_CHECK_(n,t) (void)memset((char*)(d), (U8)(b), (n) * sizeof(t))) |
1125 | #define PoisonNew(d,n,t) PoisonWith(d,n,t,0xAB) | |
1126 | #define PoisonFree(d,n,t) PoisonWith(d,n,t,0xEF) | |
1127 | #define Poison(d,n,t) PoisonFree(d,n,t) | |
27d5b266 | 1128 | |
2304df62 | 1129 | #ifdef USE_STRUCT_COPY |
ff68c719 | 1130 | #define StructCopy(s,d,t) (*((t*)(d)) = *((t*)(s))) |
bee1dbe2 LW |
1131 | #else |
1132 | #define StructCopy(s,d,t) Copy(s,d,1,t) | |
1133 | #endif | |
2cc61e15 | 1134 | |
622913ab | 1135 | #define C_ARRAY_LENGTH(a) (sizeof(a)/sizeof((a)[0])) |
cd431fde | 1136 | #define C_ARRAY_END(a) (a) + (sizeof(a)/sizeof((a)[0])) |
622913ab | 1137 | |
2cc61e15 DD |
1138 | #ifdef NEED_VA_COPY |
1139 | # ifdef va_copy | |
1140 | # define Perl_va_copy(s, d) va_copy(d, s) | |
2cc61e15 | 1141 | # else |
a1866d1b JH |
1142 | # if defined(__va_copy) |
1143 | # define Perl_va_copy(s, d) __va_copy(d, s) | |
1144 | # else | |
1145 | # define Perl_va_copy(s, d) Copy(s, d, 1, va_list) | |
1146 | # endif | |
2cc61e15 DD |
1147 | # endif |
1148 | #endif | |
1149 | ||
472d47bc SB |
1150 | /* convenience debug macros */ |
1151 | #ifdef USE_ITHREADS | |
1152 | #define pTHX_FORMAT "Perl interpreter: 0x%p" | |
1153 | #define pTHX__FORMAT ", Perl interpreter: 0x%p" | |
f54cb97a AL |
1154 | #define pTHX_VALUE_ (void *)my_perl, |
1155 | #define pTHX_VALUE (void *)my_perl | |
1156 | #define pTHX__VALUE_ ,(void *)my_perl, | |
1157 | #define pTHX__VALUE ,(void *)my_perl | |
472d47bc | 1158 | #else |
3609ea0d | 1159 | #define pTHX_FORMAT |
472d47bc | 1160 | #define pTHX__FORMAT |
3609ea0d | 1161 | #define pTHX_VALUE_ |
472d47bc | 1162 | #define pTHX_VALUE |
3609ea0d | 1163 | #define pTHX__VALUE_ |
472d47bc SB |
1164 | #define pTHX__VALUE |
1165 | #endif /* USE_ITHREADS */ | |
3609ea0d | 1166 | |
2acdbac1 NC |
1167 | /* Perl_deprecate was not part of the public API, and did not have a deprecate() |
1168 | shortcut macro defined without -DPERL_CORE. Neither codesearch.google.com nor | |
1169 | CPAN::Unpack show any users outside the core. */ | |
1170 | #ifdef PERL_CORE | |
d1d15184 | 1171 | # define deprecate(s) Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), "Use of " s " is deprecated") |
2acdbac1 NC |
1172 | #endif |
1173 | ||
e9a8c099 MHM |
1174 | /* |
1175 | * Local variables: | |
1176 | * c-indentation-style: bsd | |
1177 | * c-basic-offset: 4 | |
1178 | * indent-tabs-mode: t | |
1179 | * End: | |
1180 | * | |
1181 | * ex: set ts=8 sts=4 sw=4 noet: | |
1182 | */ |