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 #if defined(UINT8_MAX) && defined(INT16_MAX) && defined(INT32_MAX)
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
218 #define I16_MAX INT16_MAX
219 #define I16_MIN INT16_MIN
220 #define U16_MAX UINT16_MAX
221 #define U16_MIN UINT16_MIN
223 #define I32_MAX INT32_MAX
224 #define I32_MIN INT32_MIN
225 #ifndef UINT32_MAX_BROKEN /* e.g. HP-UX with gcc messes this up */
226 # define U32_MAX UINT32_MAX
228 # define U32_MAX 4294967295U
230 #define U32_MIN UINT32_MIN
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. */
236 #define U8_MAX PERL_UCHAR_MAX
237 #define U8_MIN PERL_UCHAR_MIN
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
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
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
258 /* log(2) is pretty close to 0.30103, just in case anyone is grepping for it */
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 */
263 #define Ctl(ch) ((ch) & 037)
266 =head1 SV-Body Allocation
268 =for apidoc Ama|SV*|newSVpvs|const char* s
269 Like C<newSVpvn>, but takes a literal string instead of a string/length pair.
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
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.
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
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
287 =for apidoc Am|void|sv_catpvs|SV* sv|const char* s
288 Like C<sv_catpvn>, but takes a literal string instead of a string/length pair.
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
294 =for apidoc Am|void|sv_setpvs|SV* sv|const char* s
295 Like C<sv_setpvn>, but takes a literal string instead of a string/length pair.
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
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
305 =head1 Memory Management
307 =for apidoc Ama|char*|savepvs|const char* s
308 Like C<savepvn>, but takes a literal string instead of a string/length pair.
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.
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.
319 =head1 Hash Manipulation Functions
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.
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.
328 =head1 Lexer interface
330 =for apidoc Amx|void|lex_stuff_pvs|const char *pv|U32 flags
332 Like L</lex_stuff_pvn>, but takes a literal string instead of a
338 /* concatenating with "" ensures that only literal strings are accepted as argument */
339 #define STR_WITH_LEN(s) ("" s ""), (sizeof(s)-1)
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.
346 /* STR_WITH_LEN() shortcuts */
347 #define newSVpvs(str) Perl_newSVpvn(aTHX_ STR_WITH_LEN(str))
348 #define newSVpvs_flags(str,flags) \
349 Perl_newSVpvn_flags(aTHX_ STR_WITH_LEN(str), flags)
350 #define newSVpvs_share(str) Perl_newSVpvn_share(aTHX_ STR_WITH_LEN(str), 0)
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)
359 #define sv_setpvs(sv, str) Perl_sv_setpvn(aTHX_ sv, STR_WITH_LEN(str))
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))
363 #define savepvs(str) Perl_savepvn(aTHX_ STR_WITH_LEN(str))
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)
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))
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))
382 #define lex_stuff_pvs(pv,flags) Perl_lex_stuff_pvn(aTHX_ STR_WITH_LEN(pv), flags)
384 #define get_cvs(str, flags) \
385 Perl_get_cvn_flags(aTHX_ STR_WITH_LEN(str), (flags))
388 =head1 Miscellaneous Functions
390 =for apidoc Am|bool|strNE|char* s1|char* s2
391 Test two strings to see if they are different. Returns true or
394 =for apidoc Am|bool|strEQ|char* s1|char* s2
395 Test two strings to see if they are equal. Returns true or false.
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.
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.
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.
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.
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>).
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
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))
436 # define memNE(s1,s2,l) (memcmp(s1,s2,l))
437 # define memEQ(s1,s2,l) (!memcmp(s1,s2,l))
439 # define memNE(s1,s2,l) (bcmp(s1,s2,l))
440 # define memEQ(s1,s2,l) (!bcmp(s1,s2,l))
443 #define memEQs(s1, l, s2) \
444 (sizeof(s2)-1 == l && memEQ(s1, ("" s2 ""), (sizeof(s2)-1)))
445 #define memNEs(s1, l, s2) !memEQs(s1, l, s2)
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.
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".
460 #ifdef HAS_SETLOCALE /* XXX Is there a better test for this? */
468 =head1 Character classes
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
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.
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.
489 =for apidoc Am|bool|isALPHA|char ch
490 Returns a boolean indicating whether the specified character is an
491 alphabetic character in the platform's native character set.
492 See the L<top of this section|/Character classes> for an explanation of variants
493 C<isALPHA_A> and C<isALPHA_L1>.
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()>.
501 =for apidoc Am|bool|isDIGIT|char ch
502 Returns a boolean indicating whether the specified character is a
503 digit in the platform's native character set.
504 Variants C<isDIGIT_A> and C<isDIGIT_L1> are identical to C<isDIGIT>.
506 =for apidoc Am|bool|isLOWER|char ch
507 Returns a boolean indicating whether the specified character is a
508 lowercase character in the platform's native character set.
509 See the L<top of this section|/Character classes> for an explanation of variants
510 C<isLOWER_A> and C<isLOWER_L1>.
512 =for apidoc Am|bool|isOCTAL|char ch
513 Returns a boolean indicating whether the specified character is an
514 octal digit, [0-7] in the platform's native character set.
515 Variants C<isOCTAL_A> and C<isOCTAL_L1> are identical to C<isOCTAL>.
517 =for apidoc Am|bool|isSPACE|char ch
518 Returns a boolean indicating whether the specified character is a
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>.
524 =for apidoc Am|bool|isUPPER|char ch
525 Returns a boolean indicating whether the specified character is an
526 uppercase character in the platform's native character set.
527 See the L<top of this section|/Character classes> for an explanation of variants
528 C<isUPPER_A> and C<isUPPER_L1>.
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.
537 See the L<top of this section|/Character classes> for an explanation of variants
538 C<isWORDCHAR_A> and C<isWORDCHAR_L1>.
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()>.
545 =head1 Character case changing
547 =for apidoc Am|char|toUPPER|char ch
548 Converts the specified character to uppercase in the platform's native
549 character set, if possible; otherwise returns the input character itself.
551 =for apidoc Am|char|toLOWER|char ch
552 Converts the specified character to lowercase in the platform's native
553 character set, if possible; otherwise returns the input character itself.
557 Note that these macros are repeated in Devel::PPPort, so should also be
558 patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc
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
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 */
572 # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U64TYPE)(c) & 0xFF) == (U64TYPE)(c)))
574 # define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || (((U32)(c) & 0xFF) == (U32)(c)))
577 #define isASCII(c) (FITS_IN_8_BITS(c) ? NATIVE_TO_UNI((U8) c) <= 127 : 0)
578 #define isASCII_A(c) isASCII(c)
580 /* ASCII range only */
581 #ifdef H_PERL /* If have access to perl.h, lookup in its table */
582 /* Bits for PL_charclass[] */
583 # define _CC_ALNUMC_A (1<<0)
584 # define _CC_ALNUMC_L1 (1<<1)
585 # define _CC_ALPHA_A (1<<2)
586 # define _CC_ALPHA_L1 (1<<3)
587 # define _CC_BLANK_A (1<<4)
588 # define _CC_BLANK_L1 (1<<5)
589 # define _CC_CHARNAME_CONT (1<<6)
590 # define _CC_CNTRL_A (1<<7)
591 # define _CC_CNTRL_L1 (1<<8)
592 # define _CC_DIGIT_A (1<<9)
593 # define _CC_GRAPH_A (1<<10)
594 # define _CC_GRAPH_L1 (1<<11)
595 # define _CC_IDFIRST_A (1<<12)
596 # define _CC_IDFIRST_L1 (1<<13)
597 # define _CC_LOWER_A (1<<14)
598 # define _CC_LOWER_L1 (1<<15)
599 # define _CC_OCTAL_A (1<<16)
600 # define _CC_PRINT_A (1<<17)
601 # define _CC_PRINT_L1 (1<<18)
602 # define _CC_PSXSPC_A (1<<19)
603 # define _CC_PSXSPC_L1 (1<<20)
604 # define _CC_PUNCT_A (1<<21)
605 # define _CC_PUNCT_L1 (1<<22)
606 # define _CC_SPACE_A (1<<23)
607 # define _CC_SPACE_L1 (1<<24)
608 # define _CC_UPPER_A (1<<25)
609 # define _CC_UPPER_L1 (1<<26)
610 # define _CC_WORDCHAR_A (1<<27)
611 # define _CC_WORDCHAR_L1 (1<<28)
612 # define _CC_XDIGIT_A (1<<29)
613 # define _CC_NONLATIN1_FOLD (1<<30)
619 EXTCONST U32 PL_charclass[] = {
620 # include "l1_char_class_tab.h"
623 # else /* ! DOINIT */
624 EXTCONST U32 PL_charclass[];
627 # define isALNUMC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_A))
628 # define isALPHA_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_A))
629 # define isBLANK_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_A))
630 # define isCNTRL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_A))
631 # define isDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_DIGIT_A))
632 # define isGRAPH_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_A))
633 # define isIDFIRST_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_A))
634 # define isLOWER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_A))
635 # define isOCTAL_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_OCTAL_A))
636 # define isPRINT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_A))
637 # define isPSXSPC_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_A))
638 # define isPUNCT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_A))
639 # define isSPACE_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_A))
640 # define isUPPER_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_A))
641 # define isWORDCHAR_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_A))
642 # define isXDIGIT_A(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_XDIGIT_A))
643 /* Either participates in a fold with a character above 255, or is a
645 # 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))
646 #else /* No perl.h. */
647 # define isOCTAL_A(c) ((c) >= '0' && (c) <= '9')
649 # define isALNUMC_A(c) (isASCII(c) && isALNUMC(c))
650 # define isALPHA_A(c) (isASCII(c) && isALPHA(c))
651 # define isBLANK_A(c) (isASCII(c) && isBLANK(c))
652 # define isCNTRL_A(c) (isASCII(c) && isCNTRL(c))
653 # define isDIGIT_A(c) (isASCII(c) && isDIGIT(c))
654 # define isGRAPH_A(c) (isASCII(c) && isGRAPH(c))
655 # define isIDFIRST_A(c) (isASCII(c) && isIDFIRST(c))
656 # define isLOWER_A(c) (isASCII(c) && isLOWER(c))
657 # define isPRINT_A(c) (isASCII(c) && isPRINT(c))
658 # define isPSXSPC_A(c) (isASCII(c) && isPSXSPC(c))
659 # define isPUNCT_A(c) (isASCII(c) && isPUNCT(c))
660 # define isSPACE_A(c) (isASCII(c) && isSPACE(c))
661 # define isUPPER_A(c) (isASCII(c) && isUPPER(c))
662 # define isWORDCHAR_A(c) (isASCII(c) && isWORDCHAR(c))
663 # define isXDIGIT_A(c) (isASCII(c) && isXDIGIT(c))
664 # else /* ASCII platform, no perl.h */
665 # define isALNUMC_A(c) (isALPHA_A(c) || isDIGIT_A(c))
666 # define isALPHA_A(c) (isUPPER_A(c) || isLOWER_A(c))
667 # define isBLANK_A(c) ((c) == ' ' || (c) == '\t')
668 # define isCNTRL_A(c) (FITS_IN_8_BITS(c) ? ((U8) (c) < ' ' || (c) == 127) : 0)
669 # define isDIGIT_A(c) ((c) >= '0' && (c) <= '9')
670 # define isGRAPH_A(c) (isWORDCHAR_A(c) || isPUNCT_A(c))
671 # define isIDFIRST_A(c) (isALPHA_A(c) || (c) == '_')
672 # define isLOWER_A(c) ((c) >= 'a' && (c) <= 'z')
673 # define isPRINT_A(c) (((c) >= 32 && (c) < 127))
674 # define isPSXSPC_A(c) (isSPACE_A(c) || (c) == '\v')
675 # define isPUNCT_A(c) (((c) >= 33 && (c) <= 47) || ((c) >= 58 && (c) <= 64) || ((c) >= 91 && (c) <= 96) || ((c) >= 123 && (c) <= 126))
676 # define isSPACE_A(c) ((c) == ' ' || (c) == '\t' || (c) == '\n' || (c) =='\r' || (c) == '\f')
677 # define isUPPER_A(c) ((c) >= 'A' && (c) <= 'Z')
678 # define isWORDCHAR_A(c) (isALPHA_A(c) || isDIGIT_A(c) || (c) == '_')
679 # define isXDIGIT_A(c) (isDIGIT_A(c) || ((c) >= 'a' && (c) <= 'f') || ((c) >= 'A' && (c) <= 'F'))
681 #endif /* ASCII range definitions */
683 /* Latin1 definitions */
685 # define isALNUMC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALNUMC_L1))
686 # define isALPHA_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_ALPHA_L1))
687 # define isBLANK_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_BLANK_L1))
688 /* continuation character for legal NAME in \N{NAME} */
689 # define isCHARNAME_CONT(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CHARNAME_CONT))
690 # define isCNTRL_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_CNTRL_L1))
691 # define isGRAPH_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_GRAPH_L1))
692 # define isIDFIRST_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_IDFIRST_L1))
693 # define isLOWER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_LOWER_L1))
694 # define isPRINT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PRINT_L1))
695 # define isPSXSPC_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PSXSPC_L1))
696 # define isPUNCT_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_PUNCT_L1))
697 # define isSPACE_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_SPACE_L1))
698 # define isUPPER_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_UPPER_L1))
699 # define isWORDCHAR_L1(c) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & _CC_WORDCHAR_L1))
700 #else /* No access to perl.h. Only a few provided here, just in case needed
701 * for backwards compatibility */
702 /* ALPHAU includes Unicode semantics for latin1 characters. It has an extra
703 * >= AA test to speed up ASCII-only tests at the expense of the others */
704 # define isALPHA_L1(c) (isALPHA(c) || (NATIVE_TO_UNI((U8) c) >= 0xAA \
705 && ((NATIVE_TO_UNI((U8) c) >= 0xC0 \
706 && NATIVE_TO_UNI((U8) c) != 0xD7 && NATIVE_TO_UNI((U8) c) != 0xF7) \
707 || NATIVE_TO_UNI((U8) c) == 0xAA \
708 || NATIVE_TO_UNI((U8) c) == 0xB5 \
709 || NATIVE_TO_UNI((U8) c) == 0xBA)))
710 # define isCHARNAME_CONT(c) (isALNUM_L1(c) || (c) == ' ' || (c) == '-' || (c) == '(' || (c) == ')' || (c) == ':' || NATIVE_TO_UNI((U8) c) == 0xA0)
713 /* Macros for backwards compatibility and for completeness when the ASCII and
714 * Latin1 values are identical */
715 #define isALNUM(c) isWORDCHAR(c)
716 #define isALNUMU(c) isWORDCHAR_L1(c)
717 #define isALPHAU(c) isALPHA_L1(c)
718 #define isDIGIT_L1(c) isDIGIT_A(c)
719 #define isOCTAL(c) isOCTAL_A(c)
720 #define isOCTAL_L1(c) isOCTAL_A(c)
721 #define isXDIGIT_L1(c) isXDIGIT_A(c)
723 /* Macros that differ between EBCDIC and ASCII. Where C89 defines a function,
724 * that is used in the EBCDIC form, because in EBCDIC we do not do locales:
725 * therefore can use native functions. For those where C89 doesn't define a
726 * function, use our function, assuming that the EBCDIC code page is isomorphic
727 * with Latin1, which the three currently recognized by Perl are. Some libc's
728 * have an isblank(), but it's not guaranteed. */
730 # define isALNUMC(c) isalnum(c)
731 # define isALPHA(c) isalpha(c)
732 # define isBLANK(c) ((c) == ' ' || (c) == '\t' || NATIVE_TO_UNI(c) == 0xA0)
733 # define isCNTRL(c) iscntrl(c)
734 # define isDIGIT(c) isdigit(c)
735 # define isGRAPH(c) isgraph(c)
736 # define isIDFIRST(c) (isALPHA(c) || (c) == '_')
737 # define isLOWER(c) islower(c)
738 # define isPRINT(c) isprint(c)
739 # define isPSXSPC(c) isspace(c)
740 # define isPUNCT(c) ispunct(c)
741 # define isSPACE(c) (isPSXSPC(c) && (c) != '\v')
742 # define isUPPER(c) isupper(c)
743 # define isXDIGIT(c) isxdigit(c)
744 # define isWORDCHAR(c) (isalnum(c) || (c) == '_')
745 # define toLOWER(c) tolower(c)
746 # define toUPPER(c) toupper(c)
747 #else /* Not EBCDIC: ASCII-only matching */
748 # define isALNUMC(c) isALNUMC_A(c)
749 # define isALPHA(c) isALPHA_A(c)
750 # define isBLANK(c) isBLANK_A(c)
751 # define isCNTRL(c) isCNTRL_A(c)
752 # define isDIGIT(c) isDIGIT_A(c)
753 # define isGRAPH(c) isGRAPH_A(c)
754 # define isIDFIRST(c) isIDFIRST_A(c)
755 # define isLOWER(c) isLOWER_A(c)
756 # define isPRINT(c) isPRINT_A(c)
757 # define isPSXSPC(c) isPSXSPC_A(c)
758 # define isPUNCT(c) isPUNCT_A(c)
759 # define isSPACE(c) isSPACE_A(c)
760 # define isUPPER(c) isUPPER_A(c)
761 # define isWORDCHAR(c) isWORDCHAR_A(c)
762 # define isXDIGIT(c) isXDIGIT_A(c)
764 /* ASCII casing. These could also be written as
765 #define toLOWER(c) (isASCII(c) ? toLOWER_LATIN1(c) : (c))
766 #define toUPPER(c) (isASCII(c) ? toUPPER_LATIN1_MOD(c) : (c))
767 which uses table lookup and mask instead of subtraction. (This would
768 work because the _MOD does not apply in the ASCII range) */
769 # define toLOWER(c) (isUPPER(c) ? (c) + ('a' - 'A') : (c))
770 # define toUPPER(c) (isLOWER(c) ? (c) - ('a' - 'A') : (c))
774 /* Use table lookup for speed; return error character for input
776 #define toLOWER_LATIN1(c) (FITS_IN_8_BITS(c) \
777 ? UNI_TO_NATIVE(PL_latin1_lc[ \
778 NATIVE_TO_UNI( (U8) (c)) ]) \
779 : UNICODE_REPLACEMENT)
780 /* Modified uc. Is correct uc except for three non-ascii chars which are
781 * all mapped to one of them, and these need special handling; error
782 * character for input out-of-range */
783 #define toUPPER_LATIN1_MOD(c) (FITS_IN_8_BITS(c) \
784 ? UNI_TO_NATIVE(PL_mod_latin1_uc[ \
785 NATIVE_TO_UNI( (U8) (c)) ]) \
786 : UNICODE_REPLACEMENT)
788 #ifdef USE_NEXT_CTYPE
790 # define isALNUM_LC(c) \
791 (NXIsAlNum((unsigned int)(c)) || (char)(c) == '_')
792 # define isIDFIRST_LC(c) \
793 (NXIsAlpha((unsigned int)(c)) || (char)(c) == '_')
794 # define isALPHA_LC(c) NXIsAlpha((unsigned int)(c))
795 # define isSPACE_LC(c) NXIsSpace((unsigned int)(c))
796 # define isDIGIT_LC(c) NXIsDigit((unsigned int)(c))
797 # define isUPPER_LC(c) NXIsUpper((unsigned int)(c))
798 # define isLOWER_LC(c) NXIsLower((unsigned int)(c))
799 # define isALNUMC_LC(c) NXIsAlNum((unsigned int)(c))
800 # define isCNTRL_LC(c) NXIsCntrl((unsigned int)(c))
801 # define isGRAPH_LC(c) NXIsGraph((unsigned int)(c))
802 # define isPRINT_LC(c) NXIsPrint((unsigned int)(c))
803 # define isPUNCT_LC(c) NXIsPunct((unsigned int)(c))
804 # define toUPPER_LC(c) NXToUpper((unsigned int)(c))
805 # define toLOWER_LC(c) NXToLower((unsigned int)(c))
807 #else /* !USE_NEXT_CTYPE */
809 # if defined(CTYPE256) || (!defined(isascii) && !defined(HAS_ISASCII))
811 # define isALNUM_LC(c) (isalnum((unsigned char)(c)) || (char)(c) == '_')
812 # define isIDFIRST_LC(c) (isalpha((unsigned char)(c)) || (char)(c) == '_')
813 # define isALPHA_LC(c) isalpha((unsigned char)(c))
814 # define isSPACE_LC(c) isspace((unsigned char)(c))
815 # define isDIGIT_LC(c) isdigit((unsigned char)(c))
816 # define isUPPER_LC(c) isupper((unsigned char)(c))
817 # define isLOWER_LC(c) islower((unsigned char)(c))
818 # define isALNUMC_LC(c) isalnum((unsigned char)(c))
819 # define isCNTRL_LC(c) iscntrl((unsigned char)(c))
820 # define isGRAPH_LC(c) isgraph((unsigned char)(c))
821 # define isPRINT_LC(c) isprint((unsigned char)(c))
822 # define isPUNCT_LC(c) ispunct((unsigned char)(c))
823 # define toUPPER_LC(c) toupper((unsigned char)(c))
824 # define toLOWER_LC(c) tolower((unsigned char)(c))
828 # define isALNUM_LC(c) (isascii(c) && (isalnum(c) || (c) == '_'))
829 # define isIDFIRST_LC(c) (isascii(c) && (isalpha(c) || (c) == '_'))
830 # define isALPHA_LC(c) (isascii(c) && isalpha(c))
831 # define isSPACE_LC(c) (isascii(c) && isspace(c))
832 # define isDIGIT_LC(c) (isascii(c) && isdigit(c))
833 # define isUPPER_LC(c) (isascii(c) && isupper(c))
834 # define isLOWER_LC(c) (isascii(c) && islower(c))
835 # define isALNUMC_LC(c) (isascii(c) && isalnum(c))
836 # define isCNTRL_LC(c) (isascii(c) && iscntrl(c))
837 # define isGRAPH_LC(c) (isascii(c) && isgraph(c))
838 # define isPRINT_LC(c) (isascii(c) && isprint(c))
839 # define isPUNCT_LC(c) (isascii(c) && ispunct(c))
840 # define toUPPER_LC(c) toupper(c)
841 # define toLOWER_LC(c) tolower(c)
844 #endif /* USE_NEXT_CTYPE */
846 #define isPSXSPC_LC(c) (isSPACE_LC(c) || (c) == '\v')
847 #define isBLANK_LC(c) isBLANK(c) /* could be wrong */
849 #define isALNUM_uni(c) is_uni_alnum(c)
850 #define isIDFIRST_uni(c) is_uni_idfirst(c)
851 #define isALPHA_uni(c) is_uni_alpha(c)
852 #define isSPACE_uni(c) is_uni_space(c)
853 #define isDIGIT_uni(c) is_uni_digit(c)
854 #define isUPPER_uni(c) is_uni_upper(c)
855 #define isLOWER_uni(c) is_uni_lower(c)
856 #define isASCII_uni(c) is_uni_ascii(c)
857 #define isCNTRL_uni(c) is_uni_cntrl(c)
858 #define isGRAPH_uni(c) is_uni_graph(c)
859 #define isPRINT_uni(c) is_uni_print(c)
860 #define isPUNCT_uni(c) is_uni_punct(c)
861 #define isXDIGIT_uni(c) is_uni_xdigit(c)
862 #define toUPPER_uni(c,s,l) to_uni_upper(c,s,l)
863 #define toTITLE_uni(c,s,l) to_uni_title(c,s,l)
864 #define toLOWER_uni(c,s,l) to_uni_lower(c,s,l)
865 #define toFOLD_uni(c,s,l) to_uni_fold(c,s,l)
867 #define isPSXSPC_uni(c) (isSPACE_uni(c) ||(c) == '\f')
868 #define isBLANK_uni(c) isBLANK(c) /* could be wrong */
870 #define isALNUM_LC_uvchr(c) (c < 256 ? isALNUM_LC(c) : is_uni_alnum_lc(c))
871 #define isIDFIRST_LC_uvchr(c) (c < 256 ? isIDFIRST_LC(c) : is_uni_idfirst_lc(c))
872 #define isALPHA_LC_uvchr(c) (c < 256 ? isALPHA_LC(c) : is_uni_alpha_lc(c))
873 #define isSPACE_LC_uvchr(c) (c < 256 ? isSPACE_LC(c) : is_uni_space_lc(c))
874 #define isDIGIT_LC_uvchr(c) (c < 256 ? isDIGIT_LC(c) : is_uni_digit_lc(c))
875 #define isUPPER_LC_uvchr(c) (c < 256 ? isUPPER_LC(c) : is_uni_upper_lc(c))
876 #define isLOWER_LC_uvchr(c) (c < 256 ? isLOWER_LC(c) : is_uni_lower_lc(c))
877 #define isCNTRL_LC_uvchr(c) (c < 256 ? isCNTRL_LC(c) : is_uni_cntrl_lc(c))
878 #define isGRAPH_LC_uvchr(c) (c < 256 ? isGRAPH_LC(c) : is_uni_graph_lc(c))
879 #define isPRINT_LC_uvchr(c) (c < 256 ? isPRINT_LC(c) : is_uni_print_lc(c))
880 #define isPUNCT_LC_uvchr(c) (c < 256 ? isPUNCT_LC(c) : is_uni_punct_lc(c))
882 #define isPSXSPC_LC_uni(c) (isSPACE_LC_uni(c) ||(c) == '\f')
883 #define isBLANK_LC_uni(c) isBLANK(c) /* could be wrong */
885 #define isALNUM_utf8(p) is_utf8_alnum(p)
886 /* To prevent S_scan_word in toke.c from hanging, we have to make sure that
887 * IDFIRST is an alnum. See
888 * http://rt.perl.org/rt3/Ticket/Display.html?id=74022
889 * for more detail than you ever wanted to know about. This used to be not the
890 * XID version, but we decided to go with the more modern Unicode definition */
891 #define isIDFIRST_utf8(p) (is_utf8_xidfirst(p) && is_utf8_alnum(p))
892 #define isIDCONT_utf8(p) is_utf8_xidcont(p)
893 #define isALPHA_utf8(p) is_utf8_alpha(p)
894 #define isSPACE_utf8(p) is_utf8_space(p)
895 #define isDIGIT_utf8(p) is_utf8_digit(p)
896 #define isUPPER_utf8(p) is_utf8_upper(p)
897 #define isLOWER_utf8(p) is_utf8_lower(p)
898 #define isASCII_utf8(p) is_utf8_ascii(p)
899 #define isCNTRL_utf8(p) is_utf8_cntrl(p)
900 #define isGRAPH_utf8(p) is_utf8_graph(p)
901 #define isPRINT_utf8(p) is_utf8_print(p)
902 #define isPUNCT_utf8(p) is_utf8_punct(p)
903 #define isXDIGIT_utf8(p) is_utf8_xdigit(p)
904 #define toUPPER_utf8(p,s,l) to_utf8_upper(p,s,l)
905 #define toTITLE_utf8(p,s,l) to_utf8_title(p,s,l)
906 #define toLOWER_utf8(p,s,l) to_utf8_lower(p,s,l)
908 #define isPSXSPC_utf8(c) (isSPACE_utf8(c) ||(c) == '\f')
909 #define isBLANK_utf8(c) isBLANK(c) /* could be wrong */
911 #define isALNUM_LC_utf8(p) isALNUM_LC_uvchr(utf8_to_uvchr(p, 0))
912 #define isIDFIRST_LC_utf8(p) isIDFIRST_LC_uvchr(utf8_to_uvchr(p, 0))
913 #define isALPHA_LC_utf8(p) isALPHA_LC_uvchr(utf8_to_uvchr(p, 0))
914 #define isSPACE_LC_utf8(p) isSPACE_LC_uvchr(utf8_to_uvchr(p, 0))
915 #define isDIGIT_LC_utf8(p) isDIGIT_LC_uvchr(utf8_to_uvchr(p, 0))
916 #define isUPPER_LC_utf8(p) isUPPER_LC_uvchr(utf8_to_uvchr(p, 0))
917 #define isLOWER_LC_utf8(p) isLOWER_LC_uvchr(utf8_to_uvchr(p, 0))
918 #define isALNUMC_LC_utf8(p) isALNUMC_LC_uvchr(utf8_to_uvchr(p, 0))
919 #define isCNTRL_LC_utf8(p) isCNTRL_LC_uvchr(utf8_to_uvchr(p, 0))
920 #define isGRAPH_LC_utf8(p) isGRAPH_LC_uvchr(utf8_to_uvchr(p, 0))
921 #define isPRINT_LC_utf8(p) isPRINT_LC_uvchr(utf8_to_uvchr(p, 0))
922 #define isPUNCT_LC_utf8(p) isPUNCT_LC_uvchr(utf8_to_uvchr(p, 0))
924 #define isPSXSPC_LC_utf8(c) (isSPACE_LC_utf8(c) ||(c) == '\f')
925 #define isBLANK_LC_utf8(c) isBLANK(c) /* could be wrong */
927 /* This conversion works both ways, strangely enough. On EBCDIC platforms,
928 * CTRL-@ is 0, CTRL-A is 1, etc, just like on ASCII */
929 # define toCTRL(c) (toUPPER(NATIVE_TO_UNI(c)) ^ 64)
931 /* Line numbers are unsigned, 32 bits. */
933 #define NOLINE ((line_t) 4294967295UL)
935 /* Helpful alias for version prescan */
936 #define is_LAX_VERSION(a,b) \
937 (a != Perl_prescan_version(aTHX_ a, FALSE, b, NULL, NULL, NULL, NULL))
939 #define is_STRICT_VERSION(a,b) \
940 (a != Perl_prescan_version(aTHX_ a, TRUE, b, NULL, NULL, NULL, NULL))
942 #define BADVERSION(a,b,c) \
949 =head1 Memory Management
951 =for apidoc Am|void|Newx|void* ptr|int nitems|type
952 The XSUB-writer's interface to the C C<malloc> function.
954 In 5.9.3, Newx() and friends replace the older New() API, and drops
955 the first parameter, I<x>, a debug aid which allowed callers to identify
956 themselves. This aid has been superseded by a new build option,
957 PERL_MEM_LOG (see L<perlhacktips/PERL_MEM_LOG>). The older API is still
958 there for use in XS modules supporting older perls.
960 =for apidoc Am|void|Newxc|void* ptr|int nitems|type|cast
961 The XSUB-writer's interface to the C C<malloc> function, with
962 cast. See also C<Newx>.
964 =for apidoc Am|void|Newxz|void* ptr|int nitems|type
965 The XSUB-writer's interface to the C C<malloc> function. The allocated
966 memory is zeroed with C<memzero>. See also C<Newx>.
968 =for apidoc Am|void|Renew|void* ptr|int nitems|type
969 The XSUB-writer's interface to the C C<realloc> function.
971 =for apidoc Am|void|Renewc|void* ptr|int nitems|type|cast
972 The XSUB-writer's interface to the C C<realloc> function, with
975 =for apidoc Am|void|Safefree|void* ptr
976 The XSUB-writer's interface to the C C<free> function.
978 =for apidoc Am|void|Move|void* src|void* dest|int nitems|type
979 The XSUB-writer's interface to the C C<memmove> function. The C<src> is the
980 source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is
981 the type. Can do overlapping moves. See also C<Copy>.
983 =for apidoc Am|void *|MoveD|void* src|void* dest|int nitems|type
984 Like C<Move> but returns dest. Useful for encouraging compilers to tail-call
987 =for apidoc Am|void|Copy|void* src|void* dest|int nitems|type
988 The XSUB-writer's interface to the C C<memcpy> function. The C<src> is the
989 source, C<dest> is the destination, C<nitems> is the number of items, and C<type> is
990 the type. May fail on overlapping copies. See also C<Move>.
992 =for apidoc Am|void *|CopyD|void* src|void* dest|int nitems|type
994 Like C<Copy> but returns dest. Useful for encouraging compilers to tail-call
997 =for apidoc Am|void|Zero|void* dest|int nitems|type
999 The XSUB-writer's interface to the C C<memzero> function. The C<dest> is the
1000 destination, C<nitems> is the number of items, and C<type> is the type.
1002 =for apidoc Am|void *|ZeroD|void* dest|int nitems|type
1004 Like C<Zero> but returns dest. Useful for encouraging compilers to tail-call
1007 =for apidoc Am|void|StructCopy|type src|type dest|type
1008 This is an architecture-independent macro to copy one structure to another.
1010 =for apidoc Am|void|PoisonWith|void* dest|int nitems|type|U8 byte
1012 Fill up memory with a byte pattern (a byte repeated over and over
1013 again) that hopefully catches attempts to access uninitialized memory.
1015 =for apidoc Am|void|PoisonNew|void* dest|int nitems|type
1017 PoisonWith(0xAB) for catching access to allocated but uninitialized memory.
1019 =for apidoc Am|void|PoisonFree|void* dest|int nitems|type
1021 PoisonWith(0xEF) for catching access to freed memory.
1023 =for apidoc Am|void|Poison|void* dest|int nitems|type
1025 PoisonWith(0xEF) for catching access to freed memory.
1029 /* Maintained for backwards-compatibility only. Use newSV() instead. */
1031 #define NEWSV(x,len) newSV(len)
1034 #define MEM_SIZE_MAX ((MEM_SIZE)~0)
1036 /* The +0.0 in MEM_WRAP_CHECK_ is an attempt to foil
1037 * overly eager compilers that will bleat about e.g.
1038 * (U16)n > (size_t)~0/sizeof(U16) always being false. */
1039 #ifdef PERL_MALLOC_WRAP
1040 #define MEM_WRAP_CHECK(n,t) MEM_WRAP_CHECK_1(n,t,PL_memory_wrap)
1041 #define MEM_WRAP_CHECK_1(n,t,a) \
1042 (void)(sizeof(t) > 1 && ((MEM_SIZE)(n)+0.0) > MEM_SIZE_MAX/sizeof(t) && (Perl_croak_nocontext("%s",(a)),0))
1043 #define MEM_WRAP_CHECK_(n,t) MEM_WRAP_CHECK(n,t),
1045 #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)))
1049 #define MEM_WRAP_CHECK(n,t)
1050 #define MEM_WRAP_CHECK_1(n,t,a)
1051 #define MEM_WRAP_CHECK_2(n,t,a,b)
1052 #define MEM_WRAP_CHECK_(n,t)
1054 #define PERL_STRLEN_ROUNDUP(n) (((n-1+PERL_STRLEN_ROUNDUP_QUANTUM)&~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM-1)))
1060 * If PERL_MEM_LOG is defined, all Newx()s, Renew()s, and Safefree()s
1061 * go through functions, which are handy for debugging breakpoints, but
1062 * which more importantly get the immediate calling environment (file and
1063 * line number, and C function name if available) passed in. This info can
1064 * then be used for logging the calls, for which one gets a sample
1065 * implementation unless -DPERL_MEM_LOG_NOIMPL is also defined.
1068 * - not all memory allocs get logged, only those
1069 * that go through Newx() and derivatives (while all
1070 * Safefrees do get logged)
1071 * - __FILE__ and __LINE__ do not work everywhere
1072 * - __func__ or __FUNCTION__ even less so
1073 * - I think more goes on after the perlio frees but
1074 * the thing is that STDERR gets closed (as do all
1075 * the file descriptors)
1076 * - no deeper calling stack than the caller of the Newx()
1077 * or the kind, but do I look like a C reflection/introspection
1079 * - the function prototypes for the logging functions
1080 * probably should maybe be somewhere else than handy.h
1081 * - one could consider inlining (macrofying) the logging
1082 * for speed, but I am too lazy
1083 * - one could imagine recording the allocations in a hash,
1084 * (keyed by the allocation address?), and maintain that
1085 * through reallocs and frees, but how to do that without
1086 * any News() happening...?
1087 * - lots of -Ddefines to get useful/controllable output
1088 * - lots of ENV reads
1091 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);
1093 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);
1095 PERL_EXPORT_C Malloc_t Perl_mem_log_free(Malloc_t oldalloc, const char *filename, const int linenumber, const char *funcname);
1098 # ifndef PERL_MEM_LOG_NOIMPL
1107 # if defined(PERL_IN_SV_C) /* those are only used in sv.c */
1108 void Perl_mem_log_new_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname);
1109 void Perl_mem_log_del_sv(const SV *sv, const char *filename, const int linenumber, const char *funcname);
1116 #define MEM_LOG_ALLOC(n,t,a) Perl_mem_log_alloc(n,sizeof(t),STRINGIFY(t),a,__FILE__,__LINE__,FUNCTION__)
1117 #define MEM_LOG_REALLOC(n,t,v,a) Perl_mem_log_realloc(n,sizeof(t),STRINGIFY(t),v,a,__FILE__,__LINE__,FUNCTION__)
1118 #define MEM_LOG_FREE(a) Perl_mem_log_free(a,__FILE__,__LINE__,FUNCTION__)
1121 #ifndef MEM_LOG_ALLOC
1122 #define MEM_LOG_ALLOC(n,t,a) (a)
1124 #ifndef MEM_LOG_REALLOC
1125 #define MEM_LOG_REALLOC(n,t,v,a) (a)
1127 #ifndef MEM_LOG_FREE
1128 #define MEM_LOG_FREE(a) (a)
1131 #define Newx(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t))))))
1132 #define Newxc(v,n,t,c) (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t))))))
1133 #define Newxz(v,n,t) (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safecalloc((n),sizeof(t)))))
1136 /* pre 5.9.x compatibility */
1137 #define New(x,v,n,t) Newx(v,n,t)
1138 #define Newc(x,v,n,t,c) Newxc(v,n,t,c)
1139 #define Newz(x,v,n,t) Newxz(v,n,t)
1142 #define Renew(v,n,t) \
1143 (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t))))))
1144 #define Renewc(v,n,t,c) \
1145 (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t))))))
1148 #define Safefree(d) \
1149 ((d) ? (void)(safefree(MEM_LOG_FREE((Malloc_t)(d))), Poison(&(d), 1, Malloc_t)) : (void) 0)
1151 #define Safefree(d) safefree(MEM_LOG_FREE((Malloc_t)(d)))
1154 #define Move(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memmove((char*)(d),(const char*)(s), (n) * sizeof(t)))
1155 #define Copy(s,d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memcpy((char*)(d),(const char*)(s), (n) * sizeof(t)))
1156 #define Zero(d,n,t) (MEM_WRAP_CHECK_(n,t) (void)memzero((char*)(d), (n) * sizeof(t)))
1158 #define MoveD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memmove((char*)(d),(const char*)(s), (n) * sizeof(t)))
1159 #define CopyD(s,d,n,t) (MEM_WRAP_CHECK_(n,t) memcpy((char*)(d),(const char*)(s), (n) * sizeof(t)))
1161 #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t)))
1163 /* Using bzero(), which returns void. */
1164 #define ZeroD(d,n,t) (MEM_WRAP_CHECK_(n,t) memzero((char*)(d), (n) * sizeof(t)),d)
1167 #define PoisonWith(d,n,t,b) (MEM_WRAP_CHECK_(n,t) (void)memset((char*)(d), (U8)(b), (n) * sizeof(t)))
1168 #define PoisonNew(d,n,t) PoisonWith(d,n,t,0xAB)
1169 #define PoisonFree(d,n,t) PoisonWith(d,n,t,0xEF)
1170 #define Poison(d,n,t) PoisonFree(d,n,t)
1172 #ifdef USE_STRUCT_COPY
1173 #define StructCopy(s,d,t) (*((t*)(d)) = *((t*)(s)))
1175 #define StructCopy(s,d,t) Copy(s,d,1,t)
1178 #define C_ARRAY_LENGTH(a) (sizeof(a)/sizeof((a)[0]))
1179 #define C_ARRAY_END(a) (a) + (sizeof(a)/sizeof((a)[0]))
1183 # define Perl_va_copy(s, d) va_copy(d, s)
1185 # if defined(__va_copy)
1186 # define Perl_va_copy(s, d) __va_copy(d, s)
1188 # define Perl_va_copy(s, d) Copy(s, d, 1, va_list)
1193 /* convenience debug macros */
1195 #define pTHX_FORMAT "Perl interpreter: 0x%p"
1196 #define pTHX__FORMAT ", Perl interpreter: 0x%p"
1197 #define pTHX_VALUE_ (void *)my_perl,
1198 #define pTHX_VALUE (void *)my_perl
1199 #define pTHX__VALUE_ ,(void *)my_perl,
1200 #define pTHX__VALUE ,(void *)my_perl
1203 #define pTHX__FORMAT
1206 #define pTHX__VALUE_
1208 #endif /* USE_ITHREADS */
1210 /* Perl_deprecate was not part of the public API, and did not have a deprecate()
1211 shortcut macro defined without -DPERL_CORE. Neither codesearch.google.com nor
1212 CPAN::Unpack show any users outside the core. */
1214 # define deprecate(s) Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), "Use of " s " is deprecated")
1219 * c-indentation-style: bsd
1221 * indent-tabs-mode: t
1224 * ex: set ts=8 sts=4 sw=4 noet: