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
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.
11 #if !defined(__STDC__)
23 # define Null(type) ((type)NULL)
28 =for apidoc AmU||Nullch
29 Null character pointer. (No longer available when C<PERL_CORE> is defined.)
31 =for apidoc AmU||Nullsv
32 Null SV pointer. (No longer available when C<PERL_CORE> is defined.)
37 # define Nullch Null(char*)
38 # define Nullfp Null(PerlIO*)
39 # define Nullsv Null(SV*)
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;
56 * AV *av1 = (AV*)sv; <== BAD: the const has been silently cast away
57 * AV *av2 = MUTABLE_AV(sv); <== GOOD: it may warn
60 #if defined(__GNUC__) && !defined(PERL_GCC_BRACE_GROUPS_FORBIDDEN)
61 # define MUTABLE_PTR(p) ({ void *_p = (p); _p; })
63 # define MUTABLE_PTR(p) ((void *) (p))
66 #define MUTABLE_AV(p) ((AV *)MUTABLE_PTR(p))
67 #define MUTABLE_CV(p) ((CV *)MUTABLE_PTR(p))
68 #define MUTABLE_GV(p) ((GV *)MUTABLE_PTR(p))
69 #define MUTABLE_HV(p) ((HV *)MUTABLE_PTR(p))
70 #define MUTABLE_IO(p) ((IO *)MUTABLE_PTR(p))
71 #define MUTABLE_SV(p) ((SV *)MUTABLE_PTR(p))
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
77 /* bool is built-in for g++-2.6.3 and later, which might be used
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
80 also defines _G_HAVE_BOOL for both gcc and g++, but only g++
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
85 #ifdef __GNUG__ /* GNU g++ has bool built-in */
91 /* The NeXT dynamic loader headers will not build with the bool macro
92 So declare them now to clear confusion.
94 #if defined(NeXT) || defined(__NeXT__)
97 typedef enum bool { FALSE = 0, TRUE = 1 } bool;
101 # endif /* !HAS_BOOL */
102 #endif /* NeXT || __NeXT__ */
105 # if defined(UTS) || defined(VMS)
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
117 #define cBOOL(cbool) ((bool)!!(cbool))
119 /* Try to figure out __func__ or __FUNCTION__ equivalent, if any.
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. */
123 #if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) || (defined(__SUNPRO_C)) /* C99 or close enough. */
124 # define FUNCTION__ __func__
126 # if (defined(_MSC_VER) && _MSC_VER < 1300) || /* Pre-MSVC 7.0 has neither __func__ nor __FUNCTION and no good workarounds, either. */ \
127 (defined(__DECC_VER)) /* Tru64 or VMS, and strict C89 being used, but not modern enough cc (in Tur64, -c99 not known, only -std1). */
128 # define FUNCTION__ ""
130 # define FUNCTION__ __FUNCTION__ /* Common extension. */
134 /* XXX A note on the perl source internal type system. The
135 original intent was that I32 be *exactly* 32 bits.
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
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.
148 Similarly, there is no guarantee that I16 and U16 have exactly 16
151 For dealing with issues that may arise from various 32/64-bit
152 systems, we will ask Configure to check out
154 SHORTSIZE == sizeof(short)
155 INTSIZE == sizeof(int)
156 LONGSIZE == sizeof(long)
157 LONGLONGSIZE == sizeof(long long) (if HAS_LONG_LONG)
158 PTRSIZE == sizeof(void *)
159 DOUBLESIZE == sizeof(double)
160 LONG_DOUBLESIZE == sizeof(long double) (if HAS_LONG_DOUBLE).
164 #ifdef I_INTTYPES /* e.g. Linux has int64_t without <inttypes.h> */
165 # include <inttypes.h>
166 # ifdef INT32_MIN_BROKEN
168 # define INT32_MIN (-2147483647-1)
170 # ifdef INT64_MIN_BROKEN
172 # define INT64_MIN (-9223372036854775807LL-1)
187 #endif /* PERL_CORE */
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)
195 # if LONGSIZE == 8 && QUADKIND == QUAD_IS_LONG
196 # define INT64_C(c) CAT2(c,L)
197 # define UINT64_C(c) CAT2(c,UL)
199 # if defined(_WIN64) && defined(_MSC_VER)
200 # define INT64_C(c) CAT2(c,I64)
201 # define UINT64_C(c) CAT2(c,UI64)
203 # define INT64_C(c) ((I64TYPE)(c))
204 # define UINT64_C(c) ((U64TYPE)(c))
211 /* HMB H.Merijn Brand - a placeholder for preparing Configure patches:
213 * USE_DTRACE HAS_PSEUDOFORK HAS_TIMEGM LOCALTIME_R_NEEDS_TZSET
214 * GMTIME_MAX GMTIME_MIN LOCALTIME_MAX LOCALTIME_MIN
215 * HAS_CTIME64 HAS_LOCALTIME64 HAS_GMTIME64 HAS_DIFFTIME64
216 * HAS_MKTIME64 HAS_ASCTIME64 HAS_GETADDRINFO HAS_GETNAMEINFO
217 * HAS_INETNTOP HAS_INETPTON CHARBITS HAS_PRCTL
218 * HAS_SOCKADDR_SA_LEN HAS_SIN6_SCOPE_ID
219 * Not (yet) used at top level, but mention them for metaconfig
222 /* Mention I8SIZE, U8SIZE, I16SIZE, U16SIZE, I32SIZE, U32SIZE,
223 I64SIZE, and U64SIZE here so that metaconfig pulls them in. */
225 #if defined(UINT8_MAX) && defined(INT16_MAX) && defined(INT32_MAX)
227 /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type.
228 Please search CHAR_MAX in perl.h for further details. */
229 #define U8_MAX UINT8_MAX
230 #define U8_MIN UINT8_MIN
232 #define I16_MAX INT16_MAX
233 #define I16_MIN INT16_MIN
234 #define U16_MAX UINT16_MAX
235 #define U16_MIN UINT16_MIN
237 #define I32_MAX INT32_MAX
238 #define I32_MIN INT32_MIN
239 #ifndef UINT32_MAX_BROKEN /* e.g. HP-UX with gcc messes this up */
240 # define U32_MAX UINT32_MAX
242 # define U32_MAX 4294967295U
244 #define U32_MIN UINT32_MIN
248 /* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type.
249 Please search CHAR_MAX in perl.h for further details. */
250 #define U8_MAX PERL_UCHAR_MAX
251 #define U8_MIN PERL_UCHAR_MIN
253 #define I16_MAX PERL_SHORT_MAX
254 #define I16_MIN PERL_SHORT_MIN
255 #define U16_MAX PERL_USHORT_MAX
256 #define U16_MIN PERL_USHORT_MIN
259 # define I32_MAX PERL_INT_MAX
260 # define I32_MIN PERL_INT_MIN
261 # define U32_MAX PERL_UINT_MAX
262 # define U32_MIN PERL_UINT_MIN
264 # define I32_MAX PERL_LONG_MAX
265 # define I32_MIN PERL_LONG_MIN
266 # define U32_MAX PERL_ULONG_MAX
267 # define U32_MIN PERL_ULONG_MIN
272 /* log(2) is pretty close to 0.30103, just in case anyone is grepping for it */
273 #define BIT_DIGITS(N) (((N)*146)/485 + 1) /* log2(10) =~ 146/485 */
274 #define TYPE_DIGITS(T) BIT_DIGITS(sizeof(T) * 8)
275 #define TYPE_CHARS(T) (TYPE_DIGITS(T) + 2) /* sign, NUL */
277 #define Ctl(ch) ((ch) & 037)
280 =head1 SV-Body Allocation
282 =for apidoc Ama|SV*|newSVpvs|const char* s
283 Like C<newSVpvn>, but takes a literal string instead of a string/length pair.
285 =for apidoc Ama|SV*|newSVpvs_flags|const char* s|U32 flags
286 Like C<newSVpvn_flags>, but takes a literal string instead of a string/length
289 =for apidoc Ama|SV*|newSVpvs_share|const char* s
290 Like C<newSVpvn_share>, but takes a literal string instead of a string/length
291 pair and omits the hash parameter.
293 =for apidoc Am|void|sv_catpvs_flags|SV* sv|const char* s|I32 flags
294 Like C<sv_catpvn_flags>, but takes a literal string instead of a
297 =for apidoc Am|void|sv_catpvs_nomg|SV* sv|const char* s
298 Like C<sv_catpvn_nomg>, but takes a literal string instead of a
301 =for apidoc Am|void|sv_catpvs|SV* sv|const char* s
302 Like C<sv_catpvn>, but takes a literal string instead of a string/length pair.
304 =for apidoc Am|void|sv_catpvs_mg|SV* sv|const char* s
305 Like C<sv_catpvn_mg>, but takes a literal string instead of a
308 =for apidoc Am|void|sv_setpvs|SV* sv|const char* s
309 Like C<sv_setpvn>, but takes a literal string instead of a string/length pair.
311 =for apidoc Am|void|sv_setpvs_mg|SV* sv|const char* s
312 Like C<sv_setpvn_mg>, but takes a literal string instead of a
315 =for apidoc Am|SV *|sv_setref_pvs|const char* s
316 Like C<sv_setref_pvn>, but takes a literal string instead of a
319 =head1 Memory Management
321 =for apidoc Ama|char*|savepvs|const char* s
322 Like C<savepvn>, but takes a literal string instead of a string/length pair.
324 =for apidoc Ama|char*|savesharedpvs|const char* s
325 A version of C<savepvs()> which allocates the duplicate string in memory
326 which is shared between threads.
330 =for apidoc Am|HV*|gv_stashpvs|const char* name|I32 create
331 Like C<gv_stashpvn>, but takes a literal string instead of a string/length pair.
333 =head1 Hash Manipulation Functions
335 =for apidoc Am|SV**|hv_fetchs|HV* tb|const char* key|I32 lval
336 Like C<hv_fetch>, but takes a literal string instead of a string/length pair.
338 =for apidoc Am|SV**|hv_stores|HV* tb|const char* key|NULLOK SV* val
339 Like C<hv_store>, but takes a literal string instead of a string/length pair
340 and omits the hash parameter.
342 =head1 Lexer interface
344 =for apidoc Amx|void|lex_stuff_pvs|const char *pv|U32 flags
346 Like L</lex_stuff_pvn>, but takes a literal string instead of a
352 /* concatenating with "" ensures that only literal strings are accepted as argument */
353 #define STR_WITH_LEN(s) ("" s ""), (sizeof(s)-1)
355 /* note that STR_WITH_LEN() can't be used as argument to macros or functions that
356 * under some configurations might be macros, which means that it requires the full
357 * Perl_xxx(aTHX_ ...) form for any API calls where it's used.
360 /* STR_WITH_LEN() shortcuts */
361 #define newSVpvs(str) Perl_newSVpvn(aTHX_ STR_WITH_LEN(str))
362 #define newSVpvs_flags(str,flags) \
363 Perl_newSVpvn_flags(aTHX_ STR_WITH_LEN(str), flags)
364 #define newSVpvs_share(str) Perl_newSVpvn_share(aTHX_ STR_WITH_LEN(str), 0)
365 #define sv_catpvs_flags(sv, str, flags) \
366 Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), flags)
367 #define sv_catpvs_nomg(sv, str) \
368 Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), 0)
369 #define sv_catpvs(sv, str) \
370 Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC)
371 #define sv_catpvs_mg(sv, str) \
372 Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC|SV_SMAGIC)
373 #define sv_setpvs(sv, str) Perl_sv_setpvn(aTHX_ sv, STR_WITH_LEN(str))
374 #define sv_setpvs_mg(sv, str) Perl_sv_setpvn_mg(aTHX_ sv, STR_WITH_LEN(str))
375 #define sv_setref_pvs(rv, classname, str) \
376 Perl_sv_setref_pvn(aTHX_ rv, classname, STR_WITH_LEN(str))
377 #define savepvs(str) Perl_savepvn(aTHX_ STR_WITH_LEN(str))
378 #define savesharedpvs(str) Perl_savesharedpvn(aTHX_ STR_WITH_LEN(str))
379 #define gv_stashpvs(str, create) \
380 Perl_gv_stashpvn(aTHX_ STR_WITH_LEN(str), create)
381 #define gv_fetchpvs(namebeg, add, sv_type) \
382 Perl_gv_fetchpvn_flags(aTHX_ STR_WITH_LEN(namebeg), add, sv_type)
383 #define gv_fetchpvn(namebeg, len, add, sv_type) \
384 Perl_gv_fetchpvn_flags(aTHX_ namebeg, len, add, sv_type)
385 #define sv_catxmlpvs(dsv, str, utf8) \
386 Perl_sv_catxmlpvn(aTHX_ dsv, STR_WITH_LEN(str), utf8)
387 #define hv_fetchs(hv,key,lval) \
388 ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \
389 (lval) ? (HV_FETCH_JUST_SV | HV_FETCH_LVALUE) \
390 : HV_FETCH_JUST_SV, NULL, 0))
392 #define hv_stores(hv,key,val) \
393 ((SV **)Perl_hv_common(aTHX_ (hv), NULL, STR_WITH_LEN(key), 0, \
394 (HV_FETCH_ISSTORE|HV_FETCH_JUST_SV), (val), 0))
396 #define lex_stuff_pvs(pv,flags) Perl_lex_stuff_pvn(aTHX_ STR_WITH_LEN(pv), flags)
398 #define get_cvs(str, flags) \
399 Perl_get_cvn_flags(aTHX_ STR_WITH_LEN(str), (flags))
402 =head1 Miscellaneous Functions
404 =for apidoc Am|bool|strNE|char* s1|char* s2
405 Test two strings to see if they are different. Returns true or
408 =for apidoc Am|bool|strEQ|char* s1|char* s2
409 Test two strings to see if they are equal. Returns true or false.
411 =for apidoc Am|bool|strLT|char* s1|char* s2
412 Test two strings to see if the first, C<s1>, is less than the second,
413 C<s2>. Returns true or false.
415 =for apidoc Am|bool|strLE|char* s1|char* s2
416 Test two strings to see if the first, C<s1>, is less than or equal to the
417 second, C<s2>. Returns true or false.
419 =for apidoc Am|bool|strGT|char* s1|char* s2
420 Test two strings to see if the first, C<s1>, is greater than the second,
421 C<s2>. Returns true or false.
423 =for apidoc Am|bool|strGE|char* s1|char* s2
424 Test two strings to see if the first, C<s1>, is greater than or equal to
425 the second, C<s2>. Returns true or false.
427 =for apidoc Am|bool|strnNE|char* s1|char* s2|STRLEN len
428 Test two strings to see if they are different. The C<len> parameter
429 indicates the number of bytes to compare. Returns true or false. (A
430 wrapper for C<strncmp>).
432 =for apidoc Am|bool|strnEQ|char* s1|char* s2|STRLEN len
433 Test two strings to see if they are equal. The C<len> parameter indicates
434 the number of bytes to compare. Returns true or false. (A wrapper for
440 #define strNE(s1,s2) (strcmp(s1,s2))
441 #define strEQ(s1,s2) (!strcmp(s1,s2))
442 #define strLT(s1,s2) (strcmp(s1,s2) < 0)
443 #define strLE(s1,s2) (strcmp(s1,s2) <= 0)
444 #define strGT(s1,s2) (strcmp(s1,s2) > 0)
445 #define strGE(s1,s2) (strcmp(s1,s2) >= 0)
446 #define strnNE(s1,s2,l) (strncmp(s1,s2,l))
447 #define strnEQ(s1,s2,l) (!strncmp(s1,s2,l))
450 # define memNE(s1,s2,l) (memcmp(s1,s2,l))
451 # define memEQ(s1,s2,l) (!memcmp(s1,s2,l))
453 # define memNE(s1,s2,l) (bcmp(s1,s2,l))
454 # define memEQ(s1,s2,l) (!bcmp(s1,s2,l))
457 #define memEQs(s1, l, s2) \
458 (sizeof(s2)-1 == l && memEQ(s1, ("" s2 ""), (sizeof(s2)-1)))
459 #define memNEs(s1, l, s2) !memEQs(s1, l, s2)
464 * Unfortunately, the introduction of locales means that we
465 * can't trust isupper(), etc. to tell the truth. And when
466 * it comes to /\w+/ with tainting enabled, we *must* be able
467 * to trust our character classes.
469 * Therefore, the default tests in the text of Perl will be
470 * independent of locale. Any code that wants to depend on
471 * the current locale will use the tests that begin with "lc".
474 #ifdef HAS_SETLOCALE /* XXX Is there a better test for this? */
482 =head1 Character classes
483 There are three variants for all the functions in this section. The base ones
484 operate using the character set of the platform Perl is running on. The ones
485 with an C<_A> suffix operate on the ASCII character set, and the ones with an
486 C<_L1> suffix operate on the full Latin1 character set. All are unaffected by
489 For ASCII platforms, the base function with no suffix and the one with the
490 C<_A> suffix are identical. The function with the C<_L1> suffix imposes the
491 Latin-1 character set onto the platform. That is, the code points that are
492 ASCII are unaffected, since ASCII is a subset of Latin-1. But the non-ASCII
493 code points are treated as if they are Latin-1 characters. For example,
494 C<isSPACE_L1()> will return true when called with the code point 0xA0, which is
495 the Latin-1 NO-BREAK SPACE.
497 For EBCDIC platforms, the base function with no suffix and the one with the
498 C<_L1> suffix should be identical, since, as of this writing, the EBCDIC code
499 pages that Perl knows about all are equivalent to Latin-1. The function that
500 ends in an C<_A> suffix will not return true unless the specified character also
501 has an ASCII equivalent.
503 =for apidoc Am|bool|isALPHA|char ch
504 Returns a boolean indicating whether the specified character is an
505 alphabetic character in the platform's native character set.
506 See the L<top of this section|/Character classes> for an explanation of variants
507 C<isALPHA_A> and C<isALPHA_L1>.
509 =for apidoc Am|bool|isASCII|char ch
510 Returns a boolean indicating whether the specified character is one of the 128
511 characters in the ASCII character set. On non-ASCII platforms, it is if this
512 character corresponds to an ASCII character. Variants C<isASCII_A()> and
513 C<isASCII_L1()> are identical to C<isASCII()>.
515 =for apidoc Am|bool|isDIGIT|char ch
516 Returns a boolean indicating whether the specified character is a
517 digit in the platform's native character set.
518 Variants C<isDIGIT_A> and C<isDIGIT_L1> are identical to C<isDIGIT>.
520 =for apidoc Am|bool|isLOWER|char ch
521 Returns a boolean indicating whether the specified character is a
522 lowercase character in the platform's native character set.
523 See the L<top of this section|/Character classes> for an explanation of variants
524 C<isLOWER_A> and C<isLOWER_L1>.
526 =for apidoc Am|bool|isOCTAL|char ch
527 Returns a boolean indicating whether the specified character is an
528 octal digit, [0-7] in the platform's native character set.
529 Variants C<isOCTAL_A> and C<isOCTAL_L1> are identical to C<isOCTAL>.
531 =for apidoc Am|bool|isSPACE|char ch
532 Returns a boolean indicating whether the specified character is a
533 whitespace character in the platform's native character set. This is the same
534 as what C<\s> matches in a regular expression.
535 See the L<top of this section|/Character classes> for an explanation of variants
536 C<isSPACE_A> and C<isSPACE_L1>.
538 =for apidoc Am|bool|isUPPER|char ch
539 Returns a boolean indicating whether the specified character is an
540 uppercase character in the platform's native character set.
541 See the L<top of this section|/Character classes> for an explanation of variants
542 C<isUPPER_A> and C<isUPPER_L1>.
544 =for apidoc Am|bool|isWORDCHAR|char ch
545 Returns a boolean indicating whether the specified character is a
546 character that is any of: alphabetic, numeric, or an underscore. This is the
547 same as what C<\w> matches in a regular expression.
548 C<isALNUM()> is a synonym provided for backward compatibility. Note that it
549 does not have the standard C language meaning of alphanumeric, since it matches
550 an underscore and the standard meaning does not.
551 See the L<top of this section|/Character classes> for an explanation of variants
552 C<isWORDCHAR_A> and C<isWORDCHAR_L1>.
554 =for apidoc Am|bool|isXDIGIT|char ch
555 Returns a boolean indicating whether the specified character is a hexadecimal
556 digit, [0-9A-Fa-f]. Variants C<isXDIGIT_A()> and C<isXDIGIT_L1()> are
557 identical to C<isXDIGIT()>.
559 =head1 Character case changing
561 =for apidoc Am|char|toUPPER|char ch
562 Converts the specified character to uppercase in the platform's native
563 character set, if possible; otherwise returns the input character itself.
565 =for apidoc Am|char|toLOWER|char ch
566 Converts the specified character to lowercase in the platform's native
567 character set, if possible; otherwise returns the input character itself.
571 Note that these macros are repeated in Devel::PPPort, so should also be
572 patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc
576 /* FITS_IN_8_BITS(c) returns true if c occupies no more than 8 bits. It is
577 * designed to be hopefully bomb-proof, making sure that no bits of
578 * information are lost even on a 64-bit machine, but to get the compiler to
579 * optimize it out if possible. This is because Configure makes sure that the
580 * machine has an 8-bit byte, so if c is stored in a byte, the sizeof()
581 * guarantees that this evaluates to a constant true at compile time. The use
582 * of the mask instead of '< 256' keeps gcc from complaining that it is alway
583 * true, when c's storage class is a byte. Use U64TYPE because U64 is known
584 * only in the perl core, and this macro can be called from outside that */
586 # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U64TYPE)(c) & 0xFF) == (U64TYPE)(c)))
588 # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U32)(c) & 0xFF) == (U32)(c)))
591 #define isASCII(c) (FITS_IN_8_BITS(c) ? NATIVE_TO_UNI((U8) c) <= 127 : 0)
592 #define isASCII_A(c) isASCII(c)
594 /* ASCII range only */
595 #ifdef H_PERL /* If have access to perl.h, lookup in its table */
596 # include "l1_char_class_tab.h"
597 # define isALNUMC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_A))
598 # define isALPHA_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_A))
599 # define isBLANK_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_A))
600 # define isCNTRL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_A))
601 # define isDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_DIGIT_A))
602 # define isGRAPH_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_A))
603 # define isIDFIRST_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_A))
604 # define isLOWER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_A))
605 # define isOCTAL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_OCTAL_A))
606 # define isPRINT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_A))
607 # define isPSXSPC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_A))
608 # define isPUNCT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_A))
609 # define isSPACE_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_A))
610 # define isUPPER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_A))
611 # define isWORDCHAR_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_A))
612 # define isXDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_XDIGIT_A))
613 # 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))
614 #else /* No perl.h. */
615 # define isOCTAL_A(c) ((c) >= '0' && (c) <= '9')
617 # define isALNUMC_A(c) (isASCII(c) && isALNUMC(c))
618 # define isALPHA_A(c) (isASCII(c) && isALPHA(c))
619 # define isBLANK_A(c) (isASCII(c) && isBLANK(c))
620 # define isCNTRL_A(c) (isASCII(c) && isCNTRL(c))
621 # define isDIGIT_A(c) (isASCII(c) && isDIGIT(c))
622 # define isGRAPH_A(c) (isASCII(c) && isGRAPH(c))
623 # define isIDFIRST_A(c) (isASCII(c) && isIDFIRST(c))
624 # define isLOWER_A(c) (isASCII(c) && isLOWER(c))
625 # define isPRINT_A(c) (isASCII(c) && isPRINT(c))
626 # define isPSXSPC_A(c) (isASCII(c) && isPSXSPC(c))
627 # define isPUNCT_A(c) (isASCII(c) && isPUNCT(c))
628 # define isSPACE_A(c) (isASCII(c) && isSPACE(c))
629 # define isUPPER_A(c) (isASCII(c) && isUPPER(c))
630 # define isWORDCHAR_A(c) (isASCII(c) && isWORDCHAR(c))
631 # define isXDIGIT_A(c) (isASCII(c) && isXDIGIT(c))
632 # else /* ASCII platform, no perl.h */
633 # define isALNUMC_A(c) (isALPHA_A(c) || isDIGIT_A(c))
634 # define isALPHA_A(c) (isUPPER_A(c) || isLOWER_A(c))
635 # define isBLANK_A(c) ((c) == ' ' || (c) == '\t')
636 # define isCNTRL_A(c) (FITS_IN_8_BITS(c) ? ((U8) (c) < ' ' || (c) == 127) : 0)
637 # define isDIGIT_A(c) ((c) >= '0' && (c) <= '9')
638 # define isGRAPH_A(c) (isWORDCHAR_A(c) || isPUNCT_A(c))
639 # define isIDFIRST_A(c) (isALPHA_A(c) || (c) == '_')
640 # define isLOWER_A(c) ((c) >= 'a' && (c) <= 'z')
641 # define isPRINT_A(c) (((c) >= 32 && (c) < 127))
642 # define isPSXSPC_A(c) (isSPACE_A(c) || (c) == '\v')
643 # define isPUNCT_A(c) (((c) >= 33 && (c) <= 47) || ((c) >= 58 && (c) <= 64) || ((c) >= 91 && (c) <= 96) || ((c) >= 123 && (c) <= 126))
644 # define isSPACE_A(c) ((c) == ' ' || (c) == '\t' || (c) == '\n' || (c) =='\r' || (c) == '\f')
645 # define isUPPER_A(c) ((c) >= 'A' && (c) <= 'Z')
646 # define isWORDCHAR_A(c) (isALPHA_A(c) || isDIGIT_A(c) || (c) == '_')
647 # define isXDIGIT_A(c) (isDIGIT_A(c) || ((c) >= 'a' && (c) <= 'f') || ((c) >= 'A' && (c) <= 'F'))
649 #endif /* ASCII range definitions */
651 /* Latin1 definitions */
653 # define isALNUMC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_L1))
654 # define isALPHA_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_L1))
655 # define isBLANK_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_L1))
656 /* continuation character for legal NAME in \N{NAME} */
657 # define isCHARNAME_CONT(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CHARNAME_CONT))
658 # define isCNTRL_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_L1))
659 # define isGRAPH_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_L1))
660 # define isIDFIRST_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_L1))
661 # define isLOWER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_L1))
662 # define isPRINT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_L1))
663 # define isPSXSPC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_L1))
664 # define isPUNCT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_L1))
665 # define isSPACE_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_L1))
666 # define isUPPER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_L1))
667 # define isWORDCHAR_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_L1))
668 #else /* No access to perl.h. Only a few provided here, just in case needed
669 * for backwards compatibility */
670 /* ALPHAU includes Unicode semantics for latin1 characters. It has an extra
671 * >= AA test to speed up ASCII-only tests at the expense of the others */
672 # define isALPHA_L1(c) (isALPHA(c) || (NATIVE_TO_UNI((U8) c) >= 0xAA \
673 && ((NATIVE_TO_UNI((U8) c) >= 0xC0 \
674 && NATIVE_TO_UNI((U8) c) != 0xD7 && NATIVE_TO_UNI((U8) c) != 0xF7) \
675 || NATIVE_TO_UNI((U8) c) == 0xAA \
676 || NATIVE_TO_UNI((U8) c) == 0xB5 \
677 || NATIVE_TO_UNI((U8) c) == 0xBA)))
678 # define isCHARNAME_CONT(c) (isALNUM_L1(c) || (c) == ' ' || (c) == '-' || (c) == '(' || (c) == ')' || (c) == ':' || NATIVE_TO_UNI((U8) c) == 0xA0)
681 /* Macros for backwards compatibility and for completeness when the ASCII and
682 * Latin1 values are identical */
683 #define isALNUM(c) isWORDCHAR(c)
684 #define isALNUMU(c) isWORDCHAR_L1(c)
685 #define isALPHAU(c) isALPHA_L1(c)
686 #define isDIGIT_L1(c) isDIGIT_A(c)
687 #define isOCTAL(c) isOCTAL_A(c)
688 #define isOCTAL_L1(c) isOCTAL_A(c)
689 #define isXDIGIT_L1(c) isXDIGIT_A(c)
691 /* Macros that differ between EBCDIC and ASCII. Where C89 defines a function,
692 * that is used in the EBCDIC form, because in EBCDIC we do not do locales:
693 * therefore can use native functions. For those where C89 doesn't define a
694 * function, use our function, assuming that the EBCDIC code page is isomorphic
695 * with Latin1, which the three currently recognized by Perl are. Some libc's
696 * have an isblank(), but it's not guaranteed. */
698 # define isALNUMC(c) isalnum(c)
699 # define isALPHA(c) isalpha(c)
700 # define isBLANK(c) ((c) == ' ' || (c) == '\t' || NATIVE_TO_UNI(c) == 0xA0)
701 # define isCNTRL(c) iscntrl(c)
702 # define isDIGIT(c) isdigit(c)
703 # define isGRAPH(c) isgraph(c)
704 # define isIDFIRST(c) (isALPHA(c) || (c) == '_')
705 # define isLOWER(c) islower(c)
706 # define isPRINT(c) isprint(c)
707 # define isPSXSPC(c) isspace(c)
708 # define isPUNCT(c) ispunct(c)
709 # define isSPACE(c) (isPSXSPC(c) && (c) != '\v')
710 # define isUPPER(c) isupper(c)
711 # define isXDIGIT(c) isxdigit(c)
712 # define isWORDCHAR(c) (isalnum(c) || (c) == '_')
713 # define toLOWER(c) tolower(c)
714 # define toUPPER(c) toupper(c)
715 #else /* Not EBCDIC: ASCII-only matching */
716 # define isALNUMC(c) isALNUMC_A(c)
717 # define isALPHA(c) isALPHA_A(c)
718 # define isBLANK(c) isBLANK_A(c)
719 # define isCNTRL(c) isCNTRL_A(c)
720 # define isDIGIT(c) isDIGIT_A(c)
721 # define isGRAPH(c) isGRAPH_A(c)
722 # define isIDFIRST(c) isIDFIRST_A(c)
723 # define isLOWER(c) isLOWER_A(c)
724 # define isPRINT(c) isPRINT_A(c)
725 # define isPSXSPC(c) isPSXSPC_A(c)
726 # define isPUNCT(c) isPUNCT_A(c)
727 # define isSPACE(c) isSPACE_A(c)
728 # define isUPPER(c) isUPPER_A(c)
729 # define isWORDCHAR(c) isWORDCHAR_A(c)
730 # define isXDIGIT(c) isXDIGIT_A(c)
732 /* ASCII casing. These could also be written as
733 #define toLOWER(c) (isASCII(c) ? toLOWER_LATIN1(c) : (c))
734 #define toUPPER(c) (isASCII(c) ? toUPPER_LATIN1_MOD(c) : (c))
735 which uses table lookup and mask instead of subtraction. (This would
736 work because the _MOD does not apply in the ASCII range) */
737 # define toLOWER(c) (isUPPER(c) ? (c) + ('a' - 'A') : (c))
738 # define toUPPER(c) (isLOWER(c) ? (c) - ('a' - 'A') : (c))
742 /* Use table lookup for speed; return error character for input
744 #define toLOWER_LATIN1(c) (FITS_IN_8_BITS(c) \
745 ? UNI_TO_NATIVE(PL_latin1_lc[ \
746 NATIVE_TO_UNI( (U8) (c)) ]) \
747 : UNICODE_REPLACEMENT)
748 /* Modified uc. Is correct uc except for three non-ascii chars which are
749 * all mapped to one of them, and these need special handling; error
750 * character for input out-of-range */
751 #define toUPPER_LATIN1_MOD(c) (FITS_IN_8_BITS(c) \
752 ? UNI_TO_NATIVE(PL_mod_latin1_uc[ \
753 NATIVE_TO_UNI( (U8) (c)) ]) \
754 : UNICODE_REPLACEMENT)
756 #ifdef USE_NEXT_CTYPE
758 # define isALNUM_LC(c) \
759 (NXIsAlNum((unsigned int)(c)) || (char)(c) == '_')
760 # define isIDFIRST_LC(c) \
761 (NXIsAlpha((unsigned int)(c)) || (char)(c) == '_')
762 # define isALPHA_LC(c) NXIsAlpha((unsigned int)(c))
763 # define isSPACE_LC(c) NXIsSpace((unsigned int)(c))
764 # define isDIGIT_LC(c) NXIsDigit((unsigned int)(c))
765 # define isUPPER_LC(c) NXIsUpper((unsigned int)(c))
766 # define isLOWER_LC(c) NXIsLower((unsigned int)(c))
767 # define isALNUMC_LC(c) NXIsAlNum((unsigned int)(c))
768 # define isCNTRL_LC(c) NXIsCntrl((unsigned int)(c))
769 # define isGRAPH_LC(c) NXIsGraph((unsigned int)(c))
770 # define isPRINT_LC(c) NXIsPrint((unsigned int)(c))
771 # define isPUNCT_LC(c) NXIsPunct((unsigned int)(c))
772 # define toUPPER_LC(c) NXToUpper((unsigned int)(c))
773 # define toLOWER_LC(c) NXToLower((unsigned int)(c))
775 #else /* !USE_NEXT_CTYPE */
777 # if defined(CTYPE256) || (!defined(isascii) && !defined(HAS_ISASCII))
779 # define isALNUM_LC(c) (isalnum((unsigned char)(c)) || (char)(c) == '_')
780 # define isIDFIRST_LC(c) (isalpha((unsigned char)(c)) || (char)(c) == '_')
781 # define isALPHA_LC(c) isalpha((unsigned char)(c))
782 # define isSPACE_LC(c) isspace((unsigned char)(c))
783 # define isDIGIT_LC(c) isdigit((unsigned char)(c))
784 # define isUPPER_LC(c) isupper((unsigned char)(c))
785 # define isLOWER_LC(c) islower((unsigned char)(c))
786 # define isALNUMC_LC(c) isalnum((unsigned char)(c))
787 # define isCNTRL_LC(c) iscntrl((unsigned char)(c))
788 # define isGRAPH_LC(c) isgraph((unsigned char)(c))
789 # define isPRINT_LC(c) isprint((unsigned char)(c))
790 # define isPUNCT_LC(c) ispunct((unsigned char)(c))
791 # define toUPPER_LC(c) toupper((unsigned char)(c))
792 # define toLOWER_LC(c) tolower((unsigned char)(c))
796 # define isALNUM_LC(c) (isascii(c) && (isalnum(c) || (c) == '_'))
797 # define isIDFIRST_LC(c) (isascii(c) && (isalpha(c) || (c) == '_'))
798 # define isALPHA_LC(c) (isascii(c) && isalpha(c))
799 # define isSPACE_LC(c) (isascii(c) && isspace(c))
800 # define isDIGIT_LC(c) (isascii(c) && isdigit(c))
801 # define isUPPER_LC(c) (isascii(c) && isupper(c))
802 # define isLOWER_LC(c) (isascii(c) && islower(c))
803 # define isALNUMC_LC(c) (isascii(c) && isalnum(c))
804 # define isCNTRL_LC(c) (isascii(c) && iscntrl(c))
805 # define isGRAPH_LC(c) (isascii(c) && isgraph(c))
806 # define isPRINT_LC(c) (isascii(c) && isprint(c))
807 # define isPUNCT_LC(c) (isascii(c) && ispunct(c))
808 # define toUPPER_LC(c) toupper(c)
809 # define toLOWER_LC(c) tolower(c)
812 #endif /* USE_NEXT_CTYPE */
814 #define isPSXSPC_LC(c) (isSPACE_LC(c) || (c) == '\v')
815 #define isBLANK_LC(c) isBLANK(c) /* could be wrong */
817 #define isALNUM_uni(c) is_uni_alnum(c)
818 #define isIDFIRST_uni(c) is_uni_idfirst(c)
819 #define isALPHA_uni(c) is_uni_alpha(c)
820 #define isSPACE_uni(c) is_uni_space(c)
821 #define isDIGIT_uni(c) is_uni_digit(c)
822 #define isUPPER_uni(c) is_uni_upper(c)
823 #define isLOWER_uni(c) is_uni_lower(c)
824 #define isASCII_uni(c) is_uni_ascii(c)
825 #define isCNTRL_uni(c) is_uni_cntrl(c)
826 #define isGRAPH_uni(c) is_uni_graph(c)
827 #define isPRINT_uni(c) is_uni_print(c)
828 #define isPUNCT_uni(c) is_uni_punct(c)
829 #define isXDIGIT_uni(c) is_uni_xdigit(c)
830 #define toUPPER_uni(c,s,l) to_uni_upper(c,s,l)
831 #define toTITLE_uni(c,s,l) to_uni_title(c,s,l)
832 #define toLOWER_uni(c,s,l) to_uni_lower(c,s,l)
833 #define toFOLD_uni(c,s,l) to_uni_fold(c,s,l)
835 #define isPSXSPC_uni(c) (isSPACE_uni(c) ||(c) == '\f')
836 #define isBLANK_uni(c) isBLANK(c) /* could be wrong */
838 #define isALNUM_LC_uvchr(c) (c < 256 ? isALNUM_LC(c) : is_uni_alnum_lc(c))
839 #define isIDFIRST_LC_uvchr(c) (c < 256 ? isIDFIRST_LC(c) : is_uni_idfirst_lc(c))
840 #define isALPHA_LC_uvchr(c) (c < 256 ? isALPHA_LC(c) : is_uni_alpha_lc(c))
841 #define isSPACE_LC_uvchr(c) (c < 256 ? isSPACE_LC(c) : is_uni_space_lc(c))
842 #define isDIGIT_LC_uvchr(c) (c < 256 ? isDIGIT_LC(c) : is_uni_digit_lc(c))
843 #define isUPPER_LC_uvchr(c) (c < 256 ? isUPPER_LC(c) : is_uni_upper_lc(c))
844 #define isLOWER_LC_uvchr(c) (c < 256 ? isLOWER_LC(c) : is_uni_lower_lc(c))
845 #define isCNTRL_LC_uvchr(c) (c < 256 ? isCNTRL_LC(c) : is_uni_cntrl_lc(c))
846 #define isGRAPH_LC_uvchr(c) (c < 256 ? isGRAPH_LC(c) : is_uni_graph_lc(c))
847 #define isPRINT_LC_uvchr(c) (c < 256 ? isPRINT_LC(c) : is_uni_print_lc(c))
848 #define isPUNCT_LC_uvchr(c) (c < 256 ? isPUNCT_LC(c) : is_uni_punct_lc(c))
850 #define isPSXSPC_LC_uni(c) (isSPACE_LC_uni(c) ||(c) == '\f')
851 #define isBLANK_LC_uni(c) isBLANK(c) /* could be wrong */
853 #define isALNUM_utf8(p) is_utf8_alnum(p)
854 /* The ID_Start of Unicode was originally quite limiting: it assumed an
855 * L-class character (meaning that you could not have, say, a CJK charac-
856 * ter). So, instead, perl has for a long time allowed ID_Continue but
858 * We still preserve that for backward compatibility. But we also make sure
859 * that it is alphanumeric, so S_scan_word in toke.c will not hang. See
860 * http://rt.perl.org/rt3/Ticket/Display.html?id=74022
861 * for more detail than you ever wanted to know about. */
862 #define isIDFIRST_utf8(p) \
863 (is_utf8_idcont(p) && !is_utf8_digit(p) && is_utf8_alnum(p))
864 #define isALPHA_utf8(p) is_utf8_alpha(p)
865 #define isSPACE_utf8(p) is_utf8_space(p)
866 #define isDIGIT_utf8(p) is_utf8_digit(p)
867 #define isUPPER_utf8(p) is_utf8_upper(p)
868 #define isLOWER_utf8(p) is_utf8_lower(p)
869 #define isASCII_utf8(p) is_utf8_ascii(p)
870 #define isCNTRL_utf8(p) is_utf8_cntrl(p)
871 #define isGRAPH_utf8(p) is_utf8_graph(p)
872 #define isPRINT_utf8(p) is_utf8_print(p)
873 #define isPUNCT_utf8(p) is_utf8_punct(p)
874 #define isXDIGIT_utf8(p) is_utf8_xdigit(p)
875 #define toUPPER_utf8(p,s,l) to_utf8_upper(p,s,l)
876 #define toTITLE_utf8(p,s,l) to_utf8_title(p,s,l)
877 #define toLOWER_utf8(p,s,l) to_utf8_lower(p,s,l)
879 #define isPSXSPC_utf8(c) (isSPACE_utf8(c) ||(c) == '\f')
880 #define isBLANK_utf8(c) isBLANK(c) /* could be wrong */
882 #define isALNUM_LC_utf8(p) isALNUM_LC_uvchr(utf8_to_uvchr(p, 0))
883 #define isIDFIRST_LC_utf8(p) isIDFIRST_LC_uvchr(utf8_to_uvchr(p, 0))
884 #define isALPHA_LC_utf8(p) isALPHA_LC_uvchr(utf8_to_uvchr(p, 0))
885 #define isSPACE_LC_utf8(p) isSPACE_LC_uvchr(utf8_to_uvchr(p, 0))
886 #define isDIGIT_LC_utf8(p) isDIGIT_LC_uvchr(utf8_to_uvchr(p, 0))
887 #define isUPPER_LC_utf8(p) isUPPER_LC_uvchr(utf8_to_uvchr(p, 0))
888 #define isLOWER_LC_utf8(p) isLOWER_LC_uvchr(utf8_to_uvchr(p, 0))
889 #define isALNUMC_LC_utf8(p) isALNUMC_LC_uvchr(utf8_to_uvchr(p, 0))
890 #define isCNTRL_LC_utf8(p) isCNTRL_LC_uvchr(utf8_to_uvchr(p, 0))
891 #define isGRAPH_LC_utf8(p) isGRAPH_LC_uvchr(utf8_to_uvchr(p, 0))
892 #define isPRINT_LC_utf8(p) isPRINT_LC_uvchr(utf8_to_uvchr(p, 0))
893 #define isPUNCT_LC_utf8(p) isPUNCT_LC_uvchr(utf8_to_uvchr(p, 0))
895 #define isPSXSPC_LC_utf8(c) (isSPACE_LC_utf8(c) ||(c) == '\f')
896 #define isBLANK_LC_utf8(c) isBLANK(c) /* could be wrong */
898 /* This conversion works both ways, strangely enough. On EBCDIC platforms,
899 * CTRL-@ is 0, CTRL-A is 1, etc, just like on ASCII */
900 # define toCTRL(c) (toUPPER(NATIVE_TO_UNI(c)) ^ 64)
902 /* Line numbers are unsigned, 32 bits. */
904 #define NOLINE ((line_t) 4294967295UL)
906 /* Helpful alias for version prescan */
907 #define is_LAX_VERSION(a,b) \
908 (a != Perl_prescan_version(aTHX_ a, FALSE, b, NULL, NULL, NULL, NULL))
910 #define is_STRICT_VERSION(a,b) \
911 (a != Perl_prescan_version(aTHX_ a, TRUE, b, NULL, NULL, NULL, NULL))
913 #define BADVERSION(a,b,c) \
920 =head1 Memory Management
922 =for apidoc Am|void|Newx|void* ptr|int nitems|type
923 The XSUB-writer's interface to the C C<malloc> function.
925 In 5.9.3, Newx() and friends replace the older New() API, and drops
926 the first parameter, I<x>, a debug aid which allowed callers to identify
927 themselves. This aid has been superseded by a new build option,
928 PERL_MEM_LOG (see L<perlhack/PERL_MEM_LOG>). The older API is still
929 there for use in XS modules supporting older perls.
931 =for apidoc Am|void|Newxc|void* ptr|int nitems|type|cast
932 The XSUB-writer's interface to the C C<malloc> function, with
933 cast. See also C<Newx>.
935 =for apidoc Am|void|Newxz|void* ptr|int nitems|type
936 The XSUB-writer's interface to the C C<malloc> function. The allocated
937 memory is zeroed with C<memzero>. See also C<Newx>.
939 =for apidoc Am|void|Renew|void* ptr|int nitems|type
940 The XSUB-writer's interface to the C C<realloc> function.
942 =for apidoc Am|void|Renewc|void* ptr|int nitems|type|cast
943 The XSUB-writer's interface to the C C<realloc> function, with
946 =for apidoc Am|void|Safefree|void* ptr
947 The XSUB-writer's interface to the C C<free> function.
949 =for apidoc Am|void|Move|void* src|void* dest|int nitems|type
950 The XSUB-writer's interface to the C C<memmove> function. The C<src> is the
951 source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is
952 the type. Can do overlapping moves. See also C<Copy>.
954 =for apidoc Am|void *|MoveD|void* src|void* dest|int nitems|type
955 Like C<Move> but returns dest. Useful for encouraging compilers to tail-call
958 =for apidoc Am|void|Copy|void* src|void* dest|int nitems|type
959 The XSUB-writer's interface to the C C<memcpy> function. The C<src> is the
960 source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is
961 the type. May fail on overlapping copies. See also C<Move>.
963 =for apidoc Am|void *|CopyD|void* src|void* dest|int nitems|type
965 Like C<Copy> but returns dest. Useful for encouraging compilers to tail-call
968 =for apidoc Am|void|Zero|void* dest|int nitems|type
970 The XSUB-writer's interface to the C C<memzero> function. The C<dest> is the
971 destination, C<nitems> is the number of items, and C<type> is the type.
973 =for apidoc Am|void *|ZeroD|void* dest|int nitems|type
975 Like C<Zero> but returns dest. Useful for encouraging compilers to tail-call
978 =for apidoc Am|void|StructCopy|type src|type dest|type
979 This is an architecture-independent macro to copy one structure to another.
981 =for apidoc Am|void|PoisonWith|void* dest|int nitems|type|U8 byte
983 Fill up memory with a byte pattern (a byte repeated over and over
984 again) that hopefully catches attempts to access uninitialized memory.
986 =for apidoc Am|void|PoisonNew|void* dest|int nitems|type
988 PoisonWith(0xAB) for catching access to allocated but uninitialized memory.
990 =for apidoc Am|void|PoisonFree|void* dest|int nitems|type
992 PoisonWith(0xEF) for catching access to freed memory.
994 =for apidoc Am|void|Poison|void* dest|int nitems|type
996 PoisonWith(0xEF) for catching access to freed memory.
1000 /* Maintained for backwards-compatibility only. Use newSV() instead. */
1002 #define NEWSV(x,len) newSV(len)
1005 #define MEM_SIZE_MAX ((MEM_SIZE)~0)
1007 /* The +0.0 in MEM_WRAP_CHECK_ is an attempt to foil
1008 * overly eager compilers that will bleat about e.g.
1009 * (U16)n > (size_t)~0/sizeof(U16) always being false. */
1010 #ifdef PERL_MALLOC_WRAP
1011 #define MEM_WRAP_CHECK(n,t) MEM_WRAP_CHECK_1(n,t,PL_memory_wrap)
1012 #define MEM_WRAP_CHECK_1(n,t,a) \
1013 (void)(sizeof(t) > 1 && ((MEM_SIZE)(n)+0.0) > MEM_SIZE_MAX/sizeof(t) && (Perl_croak_nocontext("%s",(a)),0))
1014 #define MEM_WRAP_CHECK_(n,t) MEM_WRAP_CHECK(n,t),
1016 #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)))
1020 #define MEM_WRAP_CHECK(n,t)
1021 #define MEM_WRAP_CHECK_1(n,t,a)
1022 #define MEM_WRAP_CHECK_2(n,t,a,b)
1023 #define MEM_WRAP_CHECK_(n,t)
1025 #define PERL_STRLEN_ROUNDUP(n) (((n-1+PERL_STRLEN_ROUNDUP_QUANTUM)&~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM-1)))
1031 * If PERL_MEM_LOG is defined, all Newx()s, Renew()s, and Safefree()s
1032 * go through functions, which are handy for debugging breakpoints, but
1033 * which more importantly get the immediate calling environment (file and
1034 * line number, and C function name if available) passed in. This info can
1035 * then be used for logging the calls, for which one gets a sample
1036 * implementation unless -DPERL_MEM_LOG_NOIMPL is also defined.
1039 * - not all memory allocs get logged, only those
1040 * that go through Newx() and derivatives (while all
1041 * Safefrees do get logged)
1042 * - __FILE__ and __LINE__ do not work everywhere
1043 * - __func__ or __FUNCTION__ even less so
1044 * - I think more goes on after the perlio frees but
1045 * the thing is that STDERR gets closed (as do all
1046 * the file descriptors)
1047 * - no deeper calling stack than the caller of the Newx()
1048 * or the kind, but do I look like a C reflection/introspection
1050 * - the function prototypes for the logging functions
1051 * probably should maybe be somewhere else than handy.h
1052 * - one could consider inlining (macrofying) the logging
1053 * for speed, but I am too lazy
1054 * - one could imagine recording the allocations in a hash,
1055 * (keyed by the allocation address?), and maintain that
1056 * through reallocs and frees, but how to do that without
1057 * any News() happening...?
1058 * - lots of -Ddefines to get useful/controllable output
1059 * - lots of ENV reads
1062 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);
1064 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);
1066 PERL_EXPORT_C Malloc_t Perl_mem_log_free(Malloc_t oldalloc, const char *filename, const int linenumber, const char *funcname);
1069 # ifndef PERL_MEM_LOG_NOIMPL
1078 # if defined(PERL_IN_SV_C) /* those are only used in sv.c */
1079 void Perl_mem_log_new_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname);
1080 void Perl_mem_log_del_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname);
1087 #define MEM_LOG_ALLOC(n,t,a) Perl_mem_log_alloc(n,sizeof(t),STRINGIFY(t),a,__FILE__,__LINE__,FUNCTION__)
1088 #define MEM_LOG_REALLOC(n,t,v,a) Perl_mem_log_realloc(n,sizeof(t),STRINGIFY(t),v,a,__FILE__,__LINE__,FUNCTION__)
1089 #define MEM_LOG_FREE(a) Perl_mem_log_free(a,__FILE__,__LINE__,FUNCTION__)
1092 #ifndef MEM_LOG_ALLOC
1093 #define MEM_LOG_ALLOC(n,t,a) (a)
1095 #ifndef MEM_LOG_REALLOC
1096 #define MEM_LOG_REALLOC(n,t,v,a) (a)
1098 #ifndef MEM_LOG_FREE
1099 #define MEM_LOG_FREE(a) (a)
1102 #define Newx(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t))))))
1103 #define Newxc(v,n,t,c) (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t))))))
1104 #define Newxz(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safecalloc((n),sizeof(t)))))
1107 /* pre 5.9.x compatibility */
1108 #define New(x,v,n,t) Newx(v,n,t)
1109 #define Newc(x,v,n,t,c) Newxc(v,n,t,c)
1110 #define Newz(x,v,n,t) Newxz(v,n,t)
1113 #define Renew(v,n,t) \
1114 (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t))))))
1115 #define Renewc(v,n,t,c) \
1116 (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t))))))
1119 #define Safefree(d) \
1120 ((d) ? (void)(safefree(MEM_LOG_FREE((Malloc_t)(d))), Poison(&(d), 1, Malloc_t)) : (void) 0)
1122 #define Safefree(d) safefree(MEM_LOG_FREE((Malloc_t)(d)))
1125 #define Move(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memmove((char*)(d),(const char*)(s), (n) * sizeof(t)))
1126 #define Copy(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memcpy((char*)(d),(const char*)(s), (n) * sizeof(t)))
1127 #define Zero(d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memzero((char*)(d), (n) * sizeof(t)))
1129 #define MoveD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memmove((char*)(d),(const char*)(s), (n) * sizeof(t)))
1130 #define CopyD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memcpy((char*)(d),(const char*)(s), (n) * sizeof(t)))
1132 #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t)))
1134 /* Using bzero(), which returns void. */
1135 #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t)),d)
1138 #define PoisonWith(d,n,t,b) (MEM_WRAP_CHECK_(n,t) (void)memset((char*)(d), (U8)(b), (n) * sizeof(t)))
1139 #define PoisonNew(d,n,t) PoisonWith(d,n,t,0xAB)
1140 #define PoisonFree(d,n,t) PoisonWith(d,n,t,0xEF)
1141 #define Poison(d,n,t) PoisonFree(d,n,t)
1143 #ifdef USE_STRUCT_COPY
1144 #define StructCopy(s,d,t) (*((t*)(d)) = *((t*)(s)))
1146 #define StructCopy(s,d,t) Copy(s,d,1,t)
1149 #define C_ARRAY_LENGTH(a) (sizeof(a)/sizeof((a)[0]))
1150 #define C_ARRAY_END(a) (a) + (sizeof(a)/sizeof((a)[0]))
1154 # define Perl_va_copy(s, d) va_copy(d, s)
1156 # if defined(__va_copy)
1157 # define Perl_va_copy(s, d) __va_copy(d, s)
1159 # define Perl_va_copy(s, d) Copy(s, d, 1, va_list)
1164 /* convenience debug macros */
1166 #define pTHX_FORMAT "Perl interpreter: 0x%p"
1167 #define pTHX__FORMAT ", Perl interpreter: 0x%p"
1168 #define pTHX_VALUE_ (void *)my_perl,
1169 #define pTHX_VALUE (void *)my_perl
1170 #define pTHX__VALUE_ ,(void *)my_perl,
1171 #define pTHX__VALUE ,(void *)my_perl
1174 #define pTHX__FORMAT
1177 #define pTHX__VALUE_
1179 #endif /* USE_ITHREADS */
1181 /* Perl_deprecate was not part of the public API, and did not have a deprecate()
1182 shortcut macro defined without -DPERL_CORE. Neither codesearch.google.com nor
1183 CPAN::Unpack show any users outside the core. */
1185 # define deprecate(s) Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), "Use of " s " is deprecated")
1190 * c-indentation-style: bsd
1192 * indent-tabs-mode: t
1195 * ex: set ts=8 sts=4 sw=4 noet: