X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/0ac55e23f1fba7b1f36cf989f7a4b6e3179a9530..c62c138b082a6c0f855caab461b44983c08f46ca:/handy.h diff --git a/handy.h b/handy.h index 0b63b56..78e5284 100644 --- a/handy.h +++ b/handy.h @@ -8,15 +8,14 @@ * */ +#ifndef HANDY_H /* Guard against nested #inclusion */ +#define HANDY_H + #if !defined(__STDC__) #ifdef NULL #undef NULL #endif -#ifndef I286 # define NULL 0 -#else -# define NULL 0L -#endif #endif #ifndef PERL_CORE @@ -70,7 +69,7 @@ Null SV pointer. (No longer available when C is defined.) #define MUTABLE_IO(p) ((IO *)MUTABLE_PTR(p)) #define MUTABLE_SV(p) ((SV *)MUTABLE_PTR(p)) -#ifdef I_STDBOOL +#if defined(I_STDBOOL) && !defined(PERL_BOOL_AS_CHAR) # include # ifndef HAS_BOOL # define HAS_BOOL 1 @@ -86,9 +85,11 @@ Null SV pointer. (No longer available when C is defined.) Andy Dougherty February 2000 */ #ifdef __GNUG__ /* GNU g++ has bool built-in */ +# ifndef PERL_BOOL_AS_CHAR # ifndef HAS_BOOL # define HAS_BOOL 1 # endif +# endif #endif /* The NeXT dynamic loader headers will not build with the bool macro @@ -105,19 +106,17 @@ Null SV pointer. (No longer available when C is defined.) #endif /* NeXT || __NeXT__ */ #ifndef HAS_BOOL -# if defined(UTS) || defined(VMS) -# define bool int -# else -# define bool char +# ifdef bool +# undef bool # endif +# define bool char # define HAS_BOOL 1 #endif -/* a simple (bool) cast may not do the right thing: if bool is defined - * as char for example, then the cast from int is implementation-defined - * (bool)!!(cbool) in a ternary triggers a bug in xlc on AIX - */ - +/* cast-to-bool. A simple (bool) cast may not do the right thing: if bool is + * defined as char for example, then the cast from int is + * implementation-defined (bool)!!(cbool) in a ternary triggers a bug in xlc on + * AIX */ #define cBOOL(cbool) ((cbool) ? (bool)1 : (bool)0) /* Try to figure out __func__ or __FUNCTION__ equivalent, if any. @@ -127,7 +126,7 @@ Null SV pointer. (No longer available when C is defined.) #if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) || (defined(__SUNPRO_C)) /* C99 or close enough. */ # define FUNCTION__ __func__ #else -# if (defined(_MSC_VER) && _MSC_VER < 1300) || /* Pre-MSVC 7.0 has neither __func__ nor __FUNCTION and no good workarounds, either. */ \ +# if (defined(USING_MSVC6)) || /* MSVC6 has neither __func__ nor __FUNCTION and no good workarounds, either. */ \ (defined(__DECC_VER)) /* Tru64 or VMS, and strict C89 being used, but not modern enough cc (in Tur64, -c99 not known, only -std1). */ # define FUNCTION__ "" # else @@ -195,13 +194,13 @@ typedef U64TYPE U64; # define PeRl_INT64_C(c) CAT2(c,LL) # define PeRl_UINT64_C(c) CAT2(c,ULL) # else -# if LONGSIZE == 8 && QUADKIND == QUAD_IS_LONG -# define PeRl_INT64_C(c) CAT2(c,L) -# define PeRl_UINT64_C(c) CAT2(c,UL) +# if QUADKIND == QUAD_IS___INT64 +# define PeRl_INT64_C(c) CAT2(c,I64) +# define PeRl_UINT64_C(c) CAT2(c,UI64) # else -# if defined(_WIN64) && defined(_MSC_VER) -# define PeRl_INT64_C(c) CAT2(c,I64) -# define PeRl_UINT64_C(c) CAT2(c,UI64) +# if LONGSIZE == 8 && QUADKIND == QUAD_IS_LONG +# define PeRl_INT64_C(c) CAT2(c,L) +# define PeRl_UINT64_C(c) CAT2(c,UL) # else # define PeRl_INT64_C(c) ((I64TYPE)(c)) # define PeRl_UINT64_C(c) ((U64TYPE)(c)) @@ -343,12 +342,13 @@ string/length pair. =cut */ -/* concatenating with "" ensures that only literal strings are accepted as argument */ +/* concatenating with "" ensures that only literal strings are accepted as + * argument */ #define STR_WITH_LEN(s) ("" s ""), (sizeof(s)-1) -/* note that STR_WITH_LEN() can't be used as argument to macros or functions that - * under some configurations might be macros, which means that it requires the full - * Perl_xxx(aTHX_ ...) form for any API calls where it's used. +/* note that STR_WITH_LEN() can't be used as argument to macros or functions + * that under some configurations might be macros, which means that it requires + * the full Perl_xxx(aTHX_ ...) form for any API calls where it's used. */ /* STR_WITH_LEN() shortcuts */ @@ -474,94 +474,359 @@ C). /* =head1 Character classes -There are three variants for all the functions in this section. The base ones -operate using the character set of the platform Perl is running on. The ones -with an C<_A> suffix operate on the ASCII character set, and the ones with an -C<_L1> suffix operate on the full Latin1 character set. All are unaffected by -locale and by C. - -For ASCII platforms, the base function with no suffix and the one with the -C<_A> suffix are identical. The function with the C<_L1> suffix imposes the -Latin-1 character set onto the platform. That is, the code points that are -ASCII are unaffected, since ASCII is a subset of Latin-1. But the non-ASCII -code points are treated as if they are Latin-1 characters. For example, -C will return true when called with the code point 0xA0, which is -the Latin-1 NO-BREAK SPACE. - -For EBCDIC platforms, the base function with no suffix and the one with the -C<_L1> suffix should be identical, since, as of this writing, the EBCDIC code -pages that Perl knows about all are equivalent to Latin-1. The function that -ends in an C<_A> suffix will not return true unless the specified character also -has an ASCII equivalent. +This section is about functions (really macros) that classify characters +into types, such as punctuation versus alphabetic, etc. Most of these are +analogous to regular expression character classes. (See +L.) There are several variants for +each class. (Not all macros have all variants; each item below lists the +ones valid for it.) None are affected by C, and only the ones +with C in the name are affected by the current locale. + +The base function, e.g., C, takes an octet (either a C or a +C) as input and returns a boolean as to whether or not the character +represented by that octet is (or on non-ASCII platforms, corresponds to) an +ASCII character in the named class based on platform, Unicode, and Perl rules. +If the input is a number that doesn't fit in an octet, FALSE is returned. + +Variant C (e.g., C) is identical to the base function +with no suffix C<"_A">. + +Variant C imposes the Latin-1 (or EBCDIC equivlalent) character set +onto the platform. That is, the code points that are ASCII are unaffected, +since ASCII is a subset of Latin-1. But the non-ASCII code points are treated +as if they are Latin-1 characters. For example, C will return +true when called with the code point 0xDF, which is a word character in both +ASCII and EBCDIC (though it represent different characters in each). + +Variant C is like the C variant, but accepts any UV code +point as input. If the code point is larger than 255, Unicode rules are used +to determine if it is in the character class. For example, +C returns TRUE, since 0x100 is LATIN CAPITAL LETTER A +WITH MACRON in Unicode, and is a word character. + +Variant C is like C, but the input is a pointer to a +(known to be well-formed) UTF-8 encoded string (C or C). The +classification of just the first (possibly multi-byte) character in the string +is tested. + +Variant C is like the C and C variants, but uses +the C library function that gives the named classification instead of +hard-coded rules. For example, C returns the result of calling +C. This means that the result is based on the current locale, which +is what C in the name stands for. FALSE is always returned if the input +won't fit into an octet. + +Variant C is like C, but is defined on any UV. It +returns the same as C for input code points less than 256, and +returns the hard-coded, not-affected-by-locale, Unicode results for larger ones. + +Variant C is like C, but the input is a pointer to a +(known to be well-formed) UTF-8 encoded string (C or C). The +classification of just the first (possibly multi-byte) character in the string +is tested. =for apidoc Am|bool|isALPHA|char ch Returns a boolean indicating whether the specified character is an -alphabetic character in the platform's native character set. +alphabetic character, analogous to C. +See the L for an explanation of variants +C, C, C, C, C, +C, and C. + +=for apidoc Am|bool|isALPHANUMERIC|char ch +Returns a boolean indicating whether the specified character is a either an +alphabetic character or decimal digit, analogous to C. See the L for an explanation of variants -C and C. +C, C, C, +C, C, C, and +C. =for apidoc Am|bool|isASCII|char ch Returns a boolean indicating whether the specified character is one of the 128 -characters in the ASCII character set. On non-ASCII platforms, it is if this +characters in the ASCII character set, analogous to C. +On non-ASCII platforms, it returns TRUE iff this character corresponds to an ASCII character. Variants C and C are identical to C. +See the L for an explanation of variants +C, C, C, C, and +C. Note, however, that some platforms do not have the C +library routine C. In these cases, the variants whose names contain +C are the same as the corresponding ones without. + +Also note, that because all ASCII characters are UTF-8 invariant (meaning they +have the exact same representation (always a single byte) whether encoded in +UTF-8 or not), C will give the correct results when called with any +byte in any string encoded or not in UTF-8. And similarly C will +work properly on any string encoded or not in UTF-8. + +=for apidoc Am|bool|isBLANK|char ch +Returns a boolean indicating whether the specified character is a +character considered to be a blank, analogous to C. +See the L for an explanation of variants +C, C, C, C, C, +C, and C. Note, however, that some +platforms do not have the C library routine C. In these cases, the +variants whose names contain C are the same as the corresponding ones +without. + +=for apidoc Am|bool|isCNTRL|char ch +Returns a boolean indicating whether the specified character is a +control character, analogous to C. +See the L for an explanation of variants +C, C, C, C, C, +C, and C +On EBCDIC platforms, you almost always want to use the C variant. =for apidoc Am|bool|isDIGIT|char ch Returns a boolean indicating whether the specified character is a -digit in the platform's native character set. +digit, analogous to C. Variants C and C are identical to C. +See the L for an explanation of variants +C, C, C, C, and +C. + +=for apidoc Am|bool|isGRAPH|char ch +Returns a boolean indicating whether the specified character is a +graphic character, analogous to C. +See the L for an explanation of variants +C, C, C, C, C, +C, and C. =for apidoc Am|bool|isLOWER|char ch Returns a boolean indicating whether the specified character is a -lowercase character in the platform's native character set. +lowercase character, analogous to C. See the L for an explanation of variants -C and C. +C, C, C, C, C, +C, and C. =for apidoc Am|bool|isOCTAL|char ch Returns a boolean indicating whether the specified character is an -octal digit, [0-7] in the platform's native character set. -Variants C and C are identical to C. +octal digit, [0-7]. +The only two variants are C and C; each is identical to +C. + +=for apidoc Am|bool|isPUNCT|char ch +Returns a boolean indicating whether the specified character is a +punctuation character, analogous to C. +Note that the definition of what is punctuation isn't as +straightforward as one might desire. See L for details. +See the L for an explanation of variants +C, C, C, C, C, +C, and C. =for apidoc Am|bool|isSPACE|char ch Returns a boolean indicating whether the specified character is a -whitespace character in the platform's native character set. This is the same -as what C<\s> matches in a regular expression. +whitespace character. This is analogous +to what C matches in a regular expression. Starting in Perl 5.18 +(experimentally), this also matches what C does. +("Experimentally" means that this change may be backed out in 5.20 or 5.22 if +field experience indicates that it was unwise.) Prior to 5.18, only the +locale forms of this macro (the ones with C in their names) matched +precisely what C does. In those releases, the only difference, +in the non-locale variants, was that C did not match a vertical tab. +(See L for a macro that matches a vertical tab in all releases.) See the L for an explanation of variants -C and C. +C, C, C, C, C, +C, and C. + +=for apidoc Am|bool|isPSXSPC|char ch +(short for Posix Space) +Starting in 5.18, this is identical (experimentally) in all its forms to the +corresponding C macros. ("Experimentally" means that this change +may be backed out in 5.20 or 5.22 if field experience indicates that it +was unwise.) +The locale forms of this macro are identical to their corresponding +C forms in all Perl releases. In releases prior to 5.18, the +non-locale forms differ from their C forms only in that the +C forms don't match a Vertical Tab, and the C forms do. +Otherwise they are identical. Thus this macro is analogous to what +C matches in a regular expression. +See the L for an explanation of variants +C, C, C, C, C, +C, and C. =for apidoc Am|bool|isUPPER|char ch Returns a boolean indicating whether the specified character is an -uppercase character in the platform's native character set. +uppercase character, analogous to C. See the L for an explanation of variants -C and C. +C, C, C, C, C, +C, and C. -=for apidoc Am|bool|isWORDCHAR|char ch +=for apidoc Am|bool|isPRINT|char ch Returns a boolean indicating whether the specified character is a -character that is any of: alphabetic, numeric, or an underscore. This is the -same as what C<\w> matches in a regular expression. -C is a synonym provided for backward compatibility. Note that it -does not have the standard C language meaning of alphanumeric, since it matches -an underscore and the standard meaning does not. +printable character, analogous to C. See the L for an explanation of variants -C and C. +C, C, C, C, C, +C, and C. + +=for apidoc Am|bool|isWORDCHAR|char ch +Returns a boolean indicating whether the specified character is a character +that is a word character, analogous to what C and C match +in a regular expression. A word character is an alphabetic character, a +decimal digit, a connecting punctuation character (such as an underscore), or +a "mark" character that attaches to one of those (like some sort of accent). +C is a synonym provided for backward compatibility, even though a +word character includes more than the standard C language meaning of +alphanumeric. +See the L for an explanation of variants +C, C, C, C, +C, C, and C. =for apidoc Am|bool|isXDIGIT|char ch Returns a boolean indicating whether the specified character is a hexadecimal -digit, [0-9A-Fa-f]. Variants C and C are -identical to C. +digit. In the ASCII range these are C<[0-9A-Fa-f]>. Variants C +and C are identical to C. +See the L for an explanation of variants +C, C, C, C, and +C. + +=for apidoc Am|bool|isIDFIRST|char ch +Returns a boolean indicating whether the specified character can be the first +character of an identifier. This is very close to, but not quite the same as +the official Unicode property C. The difference is that this +returns true only if the input character also matches L. +See the L for an explanation of variants +C, C, C, C, +C, C, and C. + +=for apidoc Am|bool|isIDCONT|char ch +Returns a boolean indicating whether the specified character can be the +second or succeeding character of an identifier. This is very close to, but +not quite the same as the official Unicode property C. The +difference is that this returns true only if the input character also matches +L. See the L for an +explanation of variants C, C, C, +C, C, C, and +C. + +=head1 Miscellaneous Functions + +=for apidoc Am|U8|READ_XDIGIT|char str* +Returns the value of an ASCII-range hex digit and advances the string pointer. +Behaviour is only well defined when isXDIGIT(*str) is true. =head1 Character case changing -=for apidoc Am|char|toUPPER|char ch -Converts the specified character to uppercase in the platform's native -character set, if possible; otherwise returns the input character itself. +=for apidoc Am|U8|toUPPER|U8 ch +Converts the specified character to uppercase. If the input is anything but an +ASCII lowercase character, that input character itself is returned. Variant +C is equivalent. + +=for apidoc Am|UV|toUPPER_uni|UV cp|U8* s|STRLEN* lenp +Converts the Unicode code point C to its uppercase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the uppercase version may be longer than the original character. + +The first code point of the uppercased version is returned +(but note, as explained just above, that there may be more.) + +=for apidoc Am|UV|toUPPER_utf8|U8* p|U8* s|STRLEN* lenp +Converts the UTF-8 encoded character at C

to its uppercase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the uppercase version may be longer than the original character. + +The first code point of the uppercased version is returned +(but note, as explained just above, that there may be more.) + +The input character at C

is assumed to be well-formed. + +=for apidoc Am|U8|toFOLD|U8 ch +Converts the specified character to foldcase. If the input is anything but an +ASCII uppercase character, that input character itself is returned. Variant +C is equivalent. (There is no equivalent C for the full +Latin1 range, as the full generality of L is needed there.) + +=for apidoc Am|UV|toFOLD_uni|UV cp|U8* s|STRLEN* lenp +Converts the Unicode code point C to its foldcase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the foldcase version may be longer than the original character. + +The first code point of the foldcased version is returned +(but note, as explained just above, that there may be more.) + +=for apidoc Am|UV|toFOLD_utf8|U8* p|U8* s|STRLEN* lenp +Converts the UTF-8 encoded character at C

to its foldcase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the foldcase version may be longer than the original character. + +The first code point of the foldcased version is returned +(but note, as explained just above, that there may be more.) + +The input character at C

is assumed to be well-formed. + +=for apidoc Am|U8|toLOWER|U8 ch +Converts the specified character to lowercase. If the input is anything but an +ASCII uppercase character, that input character itself is returned. Variant +C is equivalent. + +=for apidoc Am|U8|toLOWER_L1|U8 ch +Converts the specified Latin1 character to lowercase. The results are undefined if +the input doesn't fit in a byte. + +=for apidoc Am|U8|toLOWER_LC|U8 ch +Converts the specified character to lowercase using the current locale's rules, +if possible; otherwise returns the input character itself. + +=for apidoc Am|UV|toLOWER_uni|UV cp|U8* s|STRLEN* lenp +Converts the Unicode code point C to its lowercase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the lowercase version may be longer than the original character. + +The first code point of the lowercased version is returned +(but note, as explained just above, that there may be more.) + +=for apidoc Am|UV|toLOWER_utf8|U8* p|U8* s|STRLEN* lenp +Converts the UTF-8 encoded character at C

to its lowercase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the lowercase version may be longer than the original character. + +The first code point of the lowercased version is returned +(but note, as explained just above, that there may be more.) + +The input character at C

is assumed to be well-formed. + +=for apidoc Am|U8|toLOWER_LC|U8 ch +Converts the specified character to lowercase using the current locale's rules, +if possible; otherwise returns the input character itself. + +=for apidoc Am|U8|toTITLE|U8 ch +Converts the specified character to titlecase. If the input is anything but an +ASCII lowercase character, that input character itself is returned. Variant +C is equivalent. (There is no C for the full Latin1 range, +as the full generality of L is needed there. Titlecase is not a +concept used in locale handling, so there is no functionality for that.) + +=for apidoc Am|UV|toTITLE_uni|UV cp|U8* s|STRLEN* lenp +Converts the Unicode code point C to its titlecase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the titlecase version may be longer than the original character. + +The first code point of the titlecased version is returned +(but note, as explained just above, that there may be more.) + +=for apidoc Am|UV|toTITLE_utf8|U8* p|U8* s|STRLEN* lenp +Converts the UTF-8 encoded character at C

to its titlecase version, and +stores that in UTF-8 in C, and its length in bytes in C. Note +that the buffer pointed to by C needs to be at least C +bytes since the titlecase version may be longer than the original character. -=for apidoc Am|char|toLOWER|char ch -Converts the specified character to lowercase in the platform's native -character set, if possible; otherwise returns the input character itself. +The first code point of the titlecased version is returned +(but note, as explained just above, that there may be more.) + +The input character at C

is assumed to be well-formed. =cut +XXX Still undocumented isVERTWS_uni and _utf8; it's unclear what their names +really should be. Also toUPPER_LC and toFOLD_LC, which are subject to change. + Note that these macros are repeated in Devel::PPPort, so should also be patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc @@ -586,7 +851,15 @@ patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc #define FITS_IN_8_BITS(c) ((sizeof(c) == 1) || !(((WIDEST_UTYPE)(c)) & ~0xFF)) #ifdef EBCDIC -# define isASCII(c) (FITS_IN_8_BITS(c) && (NATIVE_TO_UNI((U8) (c)) < 128)) +# ifndef _ALL_SOURCE + /* This returns the wrong results on at least z/OS unless this is + * defined. */ +# error _ALL_SOURCE should probably be defined +# endif + + /* We could be called without perl.h, in which case NATIVE_TO_ASCII() is + * likely not defined, and so we use the native function */ +# define isASCII(c) isascii(c) #else # define isASCII(c) ((WIDEST_UTYPE)(c) < 128) #endif @@ -599,61 +872,120 @@ patched there. The file as of this writing is cpan/Devel-PPPort/parts/inc/misc * digits */ #define isOCTAL_A(c) cBOOL(FITS_IN_8_BITS(c) && (0xF8 & (c)) == '0') - /* ASCII range only */ #ifdef H_PERL /* If have access to perl.h, lookup in its table */ -/* Bits for PL_charclass[]. These use names used in l1_char_class_tab.h but - * their actual definitions are here. If that has a name not used here, it - * won't compile. */ -# define _CC_WORDCHAR_A (1<<0) -# define _CC_SPACE_A (1<<1) -# define _CC_DIGIT_A (1<<2) -# define _CC_ALNUMC_A (1<<3) -# define _CC_ALPHA_A (1<<4) -# define _CC_ASCII (1<<5) -# define _CC_CNTRL_A (1<<6) -# define _CC_GRAPH_A (1<<7) -# define _CC_LOWER_A (1<<8) -# define _CC_PRINT_A (1<<9) -# define _CC_PUNCT_A (1<<10) -# define _CC_UPPER_A (1<<11) -# define _CC_XDIGIT_A (1<<12) -# define _CC_PSXSPC_A (1<<13) -# define _CC_BLANK_A (1<<14) - -# define _CC_WORDCHAR_L1 (1<<15) -# define _CC_SPACE_L1 (1<<16) -# define _CC_ALNUMC_L1 (1<<17) -# define _CC_ALPHA_L1 (1<<18) -# define _CC_CNTRL_L1 (1<<19) -# define _CC_GRAPH_L1 (1<<20) -# define _CC_LOWER_L1 (1<<21) -# define _CC_PRINT_L1 (1<<22) -# define _CC_PUNCT_L1 (1<<23) -# define _CC_UPPER_L1 (1<<24) -# define _CC_PSXSPC_L1 (1<<25) -# define _CC_BLANK_L1 (1<<26) -# define _CC_IDFIRST_A (1<<27) -# define _CC_IDFIRST_L1 (1<<28) -# define _CC_CHARNAME_CONT (1<<29) -# define _CC_NONLATIN1_FOLD (1<<30) -# define _CC_QUOTEMETA (1U<<31) /* 1U keeps Solaris from griping */ -/* Unused: None - * If more are needed, can give up some of the above. The first ones to go - * would be those that require just two tests to verify; either there are two - * code points, like BLANK_A, or it occupies a single range like DIGIT_A, - * UPPER_A, and LOWER_A. Also consider the ones that can be replaced with two - * tests and an additional mask, so - * - * #define isCNTRL_A cBOOL(FITS_IN_8_BITS(c) \ - * && (( ! (~0x1F & NATIVE_TO_UNI(c)])) \ - * || UNLIKELY(NATIVE_TO_UNI(c) == 0x7f))) + +/* Character class numbers. For internal core Perl use only. The ones less + * than 32 are used in PL_charclass[] and the ones up through the one that + * corresponds to <_HIGHEST_REGCOMP_DOT_H_SYNC> are used by regcomp.h and + * related files. PL_charclass ones use names used in l1_char_class_tab.h but + * their actual definitions are here. If that file has a name not used here, + * it won't compile. * - * This takes advantage of the contiguous block of these with the first one's - * representation having the lower order bits all zero;, except the DELETE must - * be tested specially. A similar pattern can be used for for isCNTRL_L1, - * isPRINT_A, and isPRINT_L1 - */ + * The first group of these is ordered in what I (khw) estimate to be the + * frequency of their use. This gives a slight edge to exiting a loop earlier + * (in reginclass() in regexec.c) */ +# define _CC_WORDCHAR 0 /* \w and [:word:] */ +# define _CC_DIGIT 1 /* \d and [:digit:] */ +# define _CC_ALPHA 2 /* [:alpha:] */ +# define _CC_LOWER 3 /* [:lower:] */ +# define _CC_UPPER 4 /* [:upper:] */ +# define _CC_PUNCT 5 /* [:punct:] */ +# define _CC_PRINT 6 /* [:print:] */ +# define _CC_ALPHANUMERIC 7 /* [:alnum:] */ +# define _CC_GRAPH 8 /* [:graph:] */ +# define _CC_CASED 9 /* [:lower:] and [:upper:] under /i */ + +#define _FIRST_NON_SWASH_CC 10 +/* The character classes above are implemented with swashes. The second group + * (just below) contains the ones implemented without. These are also sorted + * in rough order of the frequency of their use, except that \v should be last, + * as it isn't a real Posix character class, and some (small) inefficiencies in + * regular expression handling would be introduced by putting it in the middle + * of those that are. Also, cntrl and ascii come after the others as it may be + * useful to group these which have no members that match above Latin1, (or + * above ASCII in the latter case) */ + +# define _CC_SPACE 10 /* \s */ +# define _CC_BLANK 11 /* [:blank:] */ +# define _CC_XDIGIT 12 /* [:xdigit:] */ +# define _CC_PSXSPC 13 /* [:space:] */ +# define _CC_CNTRL 14 /* [:cntrl:] */ +# define _CC_ASCII 15 /* [:ascii:] */ +# define _CC_VERTSPACE 16 /* \v */ + +# define _HIGHEST_REGCOMP_DOT_H_SYNC _CC_VERTSPACE + +/* The members of the third group below do not need to be coordinated with data + * structures in regcomp.[ch] and regexec.c. But they should be added to + * bootstrap_ctype() */ +# define _CC_IDFIRST 17 +# define _CC_CHARNAME_CONT 18 +# define _CC_NONLATIN1_FOLD 19 +# define _CC_QUOTEMETA 20 +# define _CC_NON_FINAL_FOLD 21 +# define _CC_IS_IN_SOME_FOLD 22 +# define _CC_BACKSLASH_FOO_LBRACE_IS_META 31 /* temp, see mk_PL_charclass.pl */ +/* Unused: 23-30 + * If more bits are needed, one could add a second word for non-64bit + * QUAD_IS_INT systems, using some #ifdefs to distinguish between having a 2nd + * word or not. The IS_IN_SOME_FOLD bit is the most easily expendable, as it + * is used only for optimization (as of this writing), and differs in the + * Latin1 range from the ALPHA bit only in two relatively unimportant + * characters: the masculine and feminine ordinal indicators, so removing it + * would just cause /i regexes which match them to run less efficiently */ + +#if defined(PERL_CORE) || defined(PERL_EXT) +/* An enum version of the character class numbers, to help compilers + * optimize */ +typedef enum { + _CC_ENUM_ALPHA = _CC_ALPHA, + _CC_ENUM_ALPHANUMERIC = _CC_ALPHANUMERIC, + _CC_ENUM_ASCII = _CC_ASCII, + _CC_ENUM_BLANK = _CC_BLANK, + _CC_ENUM_CASED = _CC_CASED, + _CC_ENUM_CNTRL = _CC_CNTRL, + _CC_ENUM_DIGIT = _CC_DIGIT, + _CC_ENUM_GRAPH = _CC_GRAPH, + _CC_ENUM_LOWER = _CC_LOWER, + _CC_ENUM_PRINT = _CC_PRINT, + _CC_ENUM_PSXSPC = _CC_PSXSPC, + _CC_ENUM_PUNCT = _CC_PUNCT, + _CC_ENUM_SPACE = _CC_SPACE, + _CC_ENUM_UPPER = _CC_UPPER, + _CC_ENUM_VERTSPACE = _CC_VERTSPACE, + _CC_ENUM_WORDCHAR = _CC_WORDCHAR, + _CC_ENUM_XDIGIT = _CC_XDIGIT +} _char_class_number; +#endif + +#define POSIX_SWASH_COUNT _FIRST_NON_SWASH_CC +#define POSIX_CC_COUNT (_HIGHEST_REGCOMP_DOT_H_SYNC + 1) + +#if defined(PERL_IN_UTF8_C) || defined(PERL_IN_REGCOMP_C) || defined(PERL_IN_REGEXEC_C) +# if _CC_WORDCHAR != 0 || _CC_DIGIT != 1 || _CC_ALPHA != 2 || _CC_LOWER != 3 \ + || _CC_UPPER != 4 || _CC_PUNCT != 5 || _CC_PRINT != 6 \ + || _CC_ALPHANUMERIC != 7 || _CC_GRAPH != 8 || _CC_CASED != 9 + #error Need to adjust order of swash_property_names[] +# endif + +/* This is declared static in each of the few files that this is #defined for + * to keep them from being publicly accessible. Hence there is a small amount + * of wasted space */ + +static const char* const swash_property_names[] = { + "XPosixWord", + "XPosixDigit", + "XPosixAlpha", + "XPosixLower", + "XPosixUpper", + "XPosixPunct", + "XPosixPrint", + "XPosixAlnum", + "XPosixGraph", + "Cased" +}; +#endif # ifdef DOINIT EXTCONST U32 PL_charclass[] = { @@ -663,379 +995,551 @@ EXTCONST U32 PL_charclass[] = { # else /* ! DOINIT */ EXTCONST U32 PL_charclass[]; # endif +#endif /* Has perl.h */ + +#if defined(H_PERL) && ! defined(BOOTSTRAP_CHARSET) + + /* The 1U keeps Solaris from griping when shifting sets the uppermost bit */ +# define _CC_mask(classnum) (1U << (classnum)) +# define _generic_isCC(c, classnum) cBOOL(FITS_IN_8_BITS(c) \ + && (PL_charclass[(U8) (c)] & _CC_mask(classnum))) + + /* The mask for the _A versions of the macros; it just adds in the bit for + * ASCII. */ +# define _CC_mask_A(classnum) (_CC_mask(classnum) | _CC_mask(_CC_ASCII)) + + /* The _A version makes sure that both the desired bit and the ASCII bit + * are present */ +# define _generic_isCC_A(c, classnum) (FITS_IN_8_BITS(c) \ + && ((PL_charclass[(U8) (c)] & _CC_mask_A(classnum)) \ + == _CC_mask_A(classnum))) + +# define isALPHA_A(c) _generic_isCC_A(c, _CC_ALPHA) +# define isALPHANUMERIC_A(c) _generic_isCC_A(c, _CC_ALPHANUMERIC) +# define isBLANK_A(c) _generic_isCC_A(c, _CC_BLANK) +# define isCNTRL_A(c) _generic_isCC_A(c, _CC_CNTRL) +# define isDIGIT_A(c) _generic_isCC(c, _CC_DIGIT) +# define isGRAPH_A(c) _generic_isCC_A(c, _CC_GRAPH) +# define isLOWER_A(c) _generic_isCC_A(c, _CC_LOWER) +# define isPRINT_A(c) _generic_isCC_A(c, _CC_PRINT) +# define isPSXSPC_A(c) _generic_isCC_A(c, _CC_PSXSPC) +# define isPUNCT_A(c) _generic_isCC_A(c, _CC_PUNCT) +# define isSPACE_A(c) _generic_isCC_A(c, _CC_SPACE) +# define isUPPER_A(c) _generic_isCC_A(c, _CC_UPPER) +# define isWORDCHAR_A(c) _generic_isCC_A(c, _CC_WORDCHAR) +# define isXDIGIT_A(c) _generic_isCC(c, _CC_XDIGIT) +# define isIDFIRST_A(c) _generic_isCC_A(c, _CC_IDFIRST) +# define isALPHA_L1(c) _generic_isCC(c, _CC_ALPHA) +# define isALPHANUMERIC_L1(c) _generic_isCC(c, _CC_ALPHANUMERIC) +# define isBLANK_L1(c) _generic_isCC(c, _CC_BLANK) + + /* continuation character for legal NAME in \N{NAME} */ +# define isCHARNAME_CONT(c) _generic_isCC(c, _CC_CHARNAME_CONT) + +# define isCNTRL_L1(c) _generic_isCC(c, _CC_CNTRL) +# define isGRAPH_L1(c) _generic_isCC(c, _CC_GRAPH) +# define isLOWER_L1(c) _generic_isCC(c, _CC_LOWER) +# define isPRINT_L1(c) _generic_isCC(c, _CC_PRINT) +# define isPSXSPC_L1(c) _generic_isCC(c, _CC_PSXSPC) +# define isPUNCT_L1(c) _generic_isCC(c, _CC_PUNCT) +# define isSPACE_L1(c) _generic_isCC(c, _CC_SPACE) +# define isUPPER_L1(c) _generic_isCC(c, _CC_UPPER) +# define isWORDCHAR_L1(c) _generic_isCC(c, _CC_WORDCHAR) +# define isIDFIRST_L1(c) _generic_isCC(c, _CC_IDFIRST) -# define _generic_isCC(c, bit) cBOOL(FITS_IN_8_BITS(c) && (PL_charclass[(U8) NATIVE_TO_UNI(c)] & (bit))) - -# define isALNUMC_A(c) _generic_isCC(c, _CC_ALNUMC_A) -# define isALPHA_A(c) _generic_isCC(c, _CC_ALPHA_A) -# define isBLANK_A(c) _generic_isCC(c, _CC_BLANK_A) -# define isCNTRL_A(c) _generic_isCC(c, _CC_CNTRL_A) -# define isDIGIT_A(c) _generic_isCC(c, _CC_DIGIT_A) -# define isGRAPH_A(c) _generic_isCC(c, _CC_GRAPH_A) -# define isIDFIRST_A(c) _generic_isCC(c, _CC_IDFIRST_A) -# define isLOWER_A(c) _generic_isCC(c, _CC_LOWER_A) -# define isPRINT_A(c) _generic_isCC(c, _CC_PRINT_A) -# define isPSXSPC_A(c) _generic_isCC(c, _CC_PSXSPC_A) -# define isPUNCT_A(c) _generic_isCC(c, _CC_PUNCT_A) -# define isSPACE_A(c) _generic_isCC(c, _CC_SPACE_A) -# define isUPPER_A(c) _generic_isCC(c, _CC_UPPER_A) -# define isWORDCHAR_A(c) _generic_isCC(c, _CC_WORDCHAR_A) -# define isXDIGIT_A(c) _generic_isCC(c, _CC_XDIGIT_A) /* Either participates in a fold with a character above 255, or is a * multi-char fold */ -# 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)) +# 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) (c)] & _CC_mask(_CC_NONLATIN1_FOLD))) + # define _isQUOTEMETA(c) _generic_isCC(c, _CC_QUOTEMETA) -#else /* No perl.h. */ +# define _IS_NON_FINAL_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c) \ + _generic_isCC(c, _CC_NON_FINAL_FOLD) +# define _IS_IN_SOME_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c) \ + _generic_isCC(c, _CC_IS_IN_SOME_FOLD) +#else /* Either don't have perl.h or don't want to use char_class_tab.h */ + + /* If we don't have perl.h, we are compiling a utility program. Below we + * hard-code various macro definitions that wouldn't otherwise be available + * to it. We can also get here if we are configured to bootstrap up Perl + * on a non-ASCII platform that doesn't have a working Perl (currently only + * EBCDIC). For these we currently use the native definitions to get + * things going. (It should also be possible to use the translation + * function NATIVE_TO_LATIN1(), but that is an extra layer of dependence on + * Perl, so it is currently avoided for the macros where it's possible to + * do so.) */ # ifdef EBCDIC -# define isALNUMC_A(c) (isASCII(c) && isALNUMC(c)) -# define isALPHA_A(c) (isASCII(c) && isALPHA(c)) -# define isBLANK_A(c) (isASCII(c) && isBLANK(c)) -# define isCNTRL_A(c) (isASCII(c) && isCNTRL(c)) -# define isDIGIT_A(c) (isASCII(c) && isDIGIT(c)) -# define isGRAPH_A(c) (isASCII(c) && isGRAPH(c)) -# define isIDFIRST_A(c) (isASCII(c) && isIDFIRST(c)) -# define isLOWER_A(c) (isASCII(c) && isLOWER(c)) -# define isPRINT_A(c) (isASCII(c) && isPRINT(c)) -# define isPSXSPC_A(c) (isASCII(c) && isPSXSPC(c)) -# define isPUNCT_A(c) (isASCII(c) && isPUNCT(c)) -# define isSPACE_A(c) (isASCII(c) && isSPACE(c)) -# define isUPPER_A(c) (isASCII(c) && isUPPER(c)) -# define isWORDCHAR_A(c) (isASCII(c) && isWORDCHAR(c)) -# define isXDIGIT_A(c) (isASCII(c) && isXDIGIT(c)) -# else /* ASCII platform, no perl.h */ -# define isALNUMC_A(c) (isALPHA_A(c) || isDIGIT_A(c)) + /* Use the native functions. They likely will return false for all + * non-ASCII values, but this makes sure */ +# define isALPHA_A(c) (isASCII(c) && isalpha(c)) +# define isALPHA_A(c) (isASCII(c) && isalpha(c)) +# define isALPHANUMERIC_A(c) (isASCII(c) && isalnum(c)) +# define isCNTRL_A(c) (isASCII(c) && iscntrl(c)) +# define isDIGIT_A(c) (isASCII(c) && isdigit(c)) +# define isGRAPH_A(c) (isASCII(c) && isgraph(c)) +# define isLOWER_A(c) (isASCII(c) && islower(c)) +# define isPRINT_A(c) (isASCII(c) && isprint(c)) +# define isPUNCT_A(c) (isASCII(c) && ispunct(c)) +# define isSPACE_A(c) (isASCII(c) && isspace(c)) +# define isUPPER_A(c) (isASCII(c) && isupper(c)) +# define isXDIGIT_A(c) (isASCII(c) && isxdigit(c)) +# else /* ASCII platform. These are coded based on first principals */ # define isALPHA_A(c) (isUPPER_A(c) || isLOWER_A(c)) -# define isBLANK_A(c) ((c) == ' ' || (c) == '\t') -# define isCNTRL_A(c) (FITS_IN_8_BITS(c) && ((U8) (c) < ' ' || (c) == 127)) +# define isALPHANUMERIC_A(c) (isALPHA_A(c) || isDIGIT_A(c)) +# define isCNTRL_A(c) (isASCII(c) && (! isPRINT_A(c))) # define isDIGIT_A(c) ((c) <= '9' && (c) >= '0') -# define isGRAPH_A(c) (isWORDCHAR_A(c) || isPUNCT_A(c)) -# define isIDFIRST_A(c) (isALPHA_A(c) || (c) == '_') +# define isGRAPH_A(c) (isPRINT_A(c) && (c) != ' ') # define isLOWER_A(c) ((c) >= 'a' && (c) <= 'z') # define isPRINT_A(c) (((c) >= 32 && (c) < 127)) -# define isPSXSPC_A(c) (isSPACE_A(c) || (c) == '\v') -# define isPUNCT_A(c) (((c) >= 33 && (c) <= 47) || ((c) >= 58 && (c) <= 64) || ((c) >= 91 && (c) <= 96) || ((c) >= 123 && (c) <= 126)) -# define isSPACE_A(c) ((c) == ' ' || (c) == '\t' || (c) == '\n' || (c) =='\r' || (c) == '\f') +# define isPUNCT_A(c) (isGRAPH_A(c) && (! isALPHANUMERIC_A(c))) +# define isSPACE_A(c) ((c) == ' ' \ + || (c) == '\t' \ + || (c) == '\n' \ + || (c) == '\r' \ + || (c) =='\v' \ + || (c) == '\f') # define isUPPER_A(c) ((c) <= 'Z' && (c) >= 'A') -# define isWORDCHAR_A(c) (isALPHA_A(c) || isDIGIT_A(c) || (c) == '_') -# define isXDIGIT_A(c) (isDIGIT_A(c) || ((c) >= 'a' && (c) <= 'f') || ((c) <= 'F' && (c) >= 'A')) -# endif -#endif /* ASCII range definitions */ - -/* Latin1 definitions */ -#ifdef H_PERL -# define isALNUMC_L1(c) _generic_isCC(c, _CC_ALNUMC_L1) -# define isALPHA_L1(c) _generic_isCC(c, _CC_ALPHA_L1) -# define isBLANK_L1(c) _generic_isCC(c, _CC_BLANK_L1) -/* continuation character for legal NAME in \N{NAME} */ -# define isCHARNAME_CONT(c) _generic_isCC(c, _CC_CHARNAME_CONT) -# define isCNTRL_L1(c) _generic_isCC(c, _CC_CNTRL_L1) -# define isGRAPH_L1(c) _generic_isCC(c, _CC_GRAPH_L1) -# define isIDFIRST_L1(c) _generic_isCC(c, _CC_IDFIRST_L1) -# define isLOWER_L1(c) _generic_isCC(c, _CC_LOWER_L1) -# define isPRINT_L1(c) _generic_isCC(c, _CC_PRINT_L1) -# define isPSXSPC_L1(c) _generic_isCC(c, _CC_PSXSPC_L1) -# define isPUNCT_L1(c) _generic_isCC(c, _CC_PUNCT_L1) -# define isSPACE_L1(c) _generic_isCC(c, _CC_SPACE_L1) -# define isUPPER_L1(c) _generic_isCC(c, _CC_UPPER_L1) -# define isWORDCHAR_L1(c) _generic_isCC(c, _CC_WORDCHAR_L1) -#else /* No access to perl.h. Only a few provided here, just in case needed - * for backwards compatibility */ - /* ALPHAU includes Unicode semantics for latin1 characters. It has an extra - * >= AA test to speed up ASCII-only tests at the expense of the others */ -# define isALPHA_L1(c) (isALPHA(c) || (NATIVE_TO_UNI((U8) c) >= 0xAA \ - && ((NATIVE_TO_UNI((U8) c) >= 0xC0 \ - && NATIVE_TO_UNI((U8) c) != 0xD7 && NATIVE_TO_UNI((U8) c) != 0xF7) \ - || NATIVE_TO_UNI((U8) c) == 0xAA \ - || NATIVE_TO_UNI((U8) c) == 0xB5 \ - || NATIVE_TO_UNI((U8) c) == 0xBA))) -# define isCHARNAME_CONT(c) (isALNUM_L1(c) || (c) == ' ' || (c) == '-' || (c) == '(' || (c) == ')' || (c) == ':' || NATIVE_TO_UNI((U8) c) == 0xA0) -#endif - -/* Macros for backwards compatibility and for completeness when the ASCII and - * Latin1 values are identical */ -#define isALNUM(c) isWORDCHAR(c) -#define isALNUMU(c) isWORDCHAR_L1(c) -#define isALPHAU(c) isALPHA_L1(c) -#define isDIGIT_L1(c) isDIGIT_A(c) -#define isOCTAL(c) isOCTAL_A(c) -#define isOCTAL_L1(c) isOCTAL_A(c) -#define isXDIGIT_L1(c) isXDIGIT_A(c) - -/* Macros that differ between EBCDIC and ASCII. Where C89 defines a function, - * that is used in the EBCDIC form, because in EBCDIC we do not do locales: - * therefore can use native functions. For those where C89 doesn't define a - * function, use our function, assuming that the EBCDIC code page is isomorphic - * with Latin1, which the three currently recognized by Perl are. Some libc's - * have an isblank(), but it's not guaranteed. */ -#ifdef EBCDIC -# define isALNUMC(c) isalnum(c) -# define isALPHA(c) isalpha(c) -# define isBLANK(c) ((c) == ' ' || (c) == '\t' || NATIVE_TO_UNI(c) == 0xA0) -# define isCNTRL(c) iscntrl(c) -# define isDIGIT(c) isdigit(c) -# define isGRAPH(c) isgraph(c) -# define isIDFIRST(c) (isALPHA(c) || (c) == '_') -# define isLOWER(c) islower(c) -# define isPRINT(c) isprint(c) -# define isPSXSPC(c) isspace(c) -# define isPUNCT(c) ispunct(c) -# define isSPACE(c) (isPSXSPC(c) && (c) != '\v') -# define isUPPER(c) isupper(c) -# define isXDIGIT(c) isxdigit(c) -# define isWORDCHAR(c) (isalnum(c) || (c) == '_') -# define toLOWER(c) tolower(c) -# define toUPPER(c) toupper(c) -#else /* Not EBCDIC: ASCII-only matching */ -# define isALNUMC(c) isALNUMC_A(c) /* Mnemonic: "C's alnum" = alpha + digit */ -# define isALPHA(c) isALPHA_A(c) -# define isBLANK(c) isBLANK_A(c) -# define isCNTRL(c) isCNTRL_A(c) -# define isDIGIT(c) isDIGIT_A(c) -# define isGRAPH(c) isGRAPH_A(c) -# define isIDFIRST(c) isIDFIRST_A(c) -# define isLOWER(c) isLOWER_A(c) -# define isPRINT(c) isPRINT_A(c) -# define isPSXSPC(c) isPSXSPC_A(c) -# define isPUNCT(c) isPUNCT_A(c) -# define isSPACE(c) isSPACE_A(c) -# define isUPPER(c) isUPPER_A(c) -# define isWORDCHAR(c) isWORDCHAR_A(c) -# define isXDIGIT(c) isXDIGIT_A(c) - - /* ASCII casing. These could also be written as - #define toLOWER(c) (isASCII(c) ? toLOWER_LATIN1(c) : (c)) - #define toUPPER(c) (isASCII(c) ? toUPPER_LATIN1_MOD(c) : (c)) - which uses table lookup and mask instead of subtraction. (This would - work because the _MOD does not apply in the ASCII range) */ -# define toLOWER(c) (isUPPER(c) ? (c) + ('a' - 'A') : (c)) -# define toUPPER(c) (isLOWER(c) ? (c) - ('a' - 'A') : (c)) -#endif - +# define isXDIGIT_A(c) (isDIGIT_A(c) \ + || ((c) >= 'a' && (c) <= 'f') \ + || ((c) <= 'F' && (c) >= 'A')) +# endif /* Below are common definitions for ASCII and non-ASCII */ +# define isBLANK_A(c) ((c) == ' ' || (c) == '\t') +# define isIDFIRST_A(c) (isALPHA_A(c) || (c) == '_') +# define isWORDCHAR_A(c) (isALPHANUMERIC_A(c) || (c) == '_') + + /* The _L1 macros may be unnecessary for both the utilities and for + * bootstrapping; I (khw) added them during debugging of bootstrapping, and + * it seems best to keep them. */ +# define isPSXSPC_A(c) isSPACE_A(c) /* XXX Assumes SPACE matches '\v' */ +# define isALPHA_L1(c) (isUPPER_L1(c) || isLOWER_L1(c)) +# define isALPHANUMERIC_L1(c) (isALPHA_L1(c) || isDIGIT_A(c)) +# define isBLANK_L1(c) (isBLANK_A(c) \ + || (FITS_IN_8_BITS(c) \ + && NATIVE_TO_LATIN1((U8) c) == 0xA0)) +# define isCNTRL_L1(c) (FITS_IN_8_BITS(c) && (! isPRINT_L1(c))) +# define isGRAPH_L1(c) (isPRINT_L1(c) && (! isBLANK_L1(c))) +# define isLOWER_L1(c) (isLOWER_A(c) \ + || (FITS_IN_8_BITS(c) \ + && ((NATIVE_TO_LATIN1((U8) c) >= 0xDF \ + && NATIVE_TO_LATIN1((U8) c) != 0xF7) \ + || NATIVE_TO_LATIN1((U8) c) == 0xAA \ + || NATIVE_TO_LATIN1((U8) c) == 0xBA \ + || NATIVE_TO_LATIN1((U8) c) == 0xB5))) +# define isPRINT_L1(c) (isPRINT_A(c) \ + || (FITS_IN_8_BITS(c) \ + && NATIVE_TO_LATIN1((U8) c) >= 0xA0)) +# define isPSXSPC_L1(c) isSPACE_L1(c) +# define isPUNCT_L1(c) (isPUNCT_A(c) \ + || (FITS_IN_8_BITS(c) \ + && (NATIVE_TO_LATIN1((U8) c) == 0xA1 \ + || NATIVE_TO_LATIN1((U8) c) == 0xA7 \ + || NATIVE_TO_LATIN1((U8) c) == 0xAB \ + || NATIVE_TO_LATIN1((U8) c) == 0xB6 \ + || NATIVE_TO_LATIN1((U8) c) == 0xB7 \ + || NATIVE_TO_LATIN1((U8) c) == 0xBB \ + || NATIVE_TO_LATIN1((U8) c) == 0xBF))) +# define isSPACE_L1(c) (isSPACE_A(c) \ + || (FITS_IN_8_BITS(c) \ + && (NATIVE_TO_LATIN1((U8) c) == 0x85 \ + || NATIVE_TO_LATIN1((U8) c) == 0xA0))) +# define isUPPER_L1(c) (isUPPER_A(c) \ + || (FITS_IN_8_BITS(c) \ + && (NATIVE_TO_LATIN1((U8) c) >= 0xC0 \ + && NATIVE_TO_LATIN1((U8) c) <= 0xDE \ + && NATIVE_TO_LATIN1((U8) c) != 0xD7))) +# define isWORDCHAR_L1(c) (isIDFIRST_L1(c) || isDIGIT_A(c)) +# define isIDFIRST_L1(c) (isALPHA_L1(c) || NATIVE_TO_LATIN1(c) == '_') +# define isCHARNAME_CONT(c) (isWORDCHAR_L1(c) \ + || isBLANK_L1(c) \ + || (c) == '-' \ + || (c) == '(' \ + || (c) == ')') + /* The following are not fully accurate in the above-ASCII range. I (khw) + * don't think it's necessary to be so for the purposes where this gets + * compiled */ +# define _isQUOTEMETA(c) (FITS_IN_8_BITS(c) && ! isWORDCHAR_L1(c)) +# define _IS_IN_SOME_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c) isALPHA_L1(c) + + /* And these aren't accurate at all. They are useful only for above + * Latin1, which utilities and bootstrapping don't deal with */ +# define _IS_NON_FINAL_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c) 0 +# define _HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(c) 0 + + /* Many of the macros later in this file are defined in terms of these. By + * implementing them with a function, which converts the class number into + * a call to the desired macro, all of the later ones work. However, that + * function won't be actually defined when building a utility program (no + * perl.h), and so a compiler error will be generated if one is attempted + * to be used. And the above-Latin1 code points require Unicode tables to + * be present, something unlikely to be the case when bootstrapping */ +# define _generic_isCC(c, classnum) \ + (FITS_IN_8_BITS(c) && S_bootstrap_ctype((U8) (c), (classnum), TRUE)) +# define _generic_isCC_A(c, classnum) \ + (FITS_IN_8_BITS(c) && S_bootstrap_ctype((U8) (c), (classnum), FALSE)) +#endif /* End of no perl.h or have BOOTSTRAP_CHARSET */ + +#define isALPHANUMERIC(c) isALPHANUMERIC_A(c) +#define isALPHA(c) isALPHA_A(c) +#define isBLANK(c) isBLANK_A(c) +#define isCNTRL(c) isCNTRL_A(c) +#define isDIGIT(c) isDIGIT_A(c) +#define isGRAPH(c) isGRAPH_A(c) +#define isIDFIRST(c) isIDFIRST_A(c) +#define isLOWER(c) isLOWER_A(c) +#define isPRINT(c) isPRINT_A(c) +#define isPSXSPC(c) isPSXSPC_A(c) +#define isPUNCT(c) isPUNCT_A(c) +#define isSPACE(c) isSPACE_A(c) +#define isUPPER(c) isUPPER_A(c) +#define isWORDCHAR(c) isWORDCHAR_A(c) +#define isXDIGIT(c) isXDIGIT_A(c) + +/* ASCII casing. These could also be written as + #define toLOWER(c) (isASCII(c) ? toLOWER_LATIN1(c) : (c)) + #define toUPPER(c) (isASCII(c) ? toUPPER_LATIN1_MOD(c) : (c)) + which uses table lookup and mask instead of subtraction. (This would + work because the _MOD does not apply in the ASCII range) */ +#define toLOWER(c) (isUPPER(c) ? (c) + ('a' - 'A') : (c)) +#define toUPPER(c) (isLOWER(c) ? (c) - ('a' - 'A') : (c)) + +/* In the ASCII range, these are equivalent to what they're here defined to be. + * But by creating these definitions, other code doesn't have to be aware of + * this detail */ +#define toFOLD(c) toLOWER(c) +#define toFOLD_LC(c) toLOWER_LC(c) +#define toTITLE(c) toUPPER(c) + +#define toLOWER_A(c) toLOWER(c) +#define toUPPER_A(c) toUPPER(c) +#define toFOLD_A(c) toFOLD(c) +#define toTITLE_A(c) toTITLE(c) /* Use table lookup for speed; return error character for input * out-of-range */ -#define toLOWER_LATIN1(c) (FITS_IN_8_BITS(c) \ - ? UNI_TO_NATIVE(PL_latin1_lc[ \ - NATIVE_TO_UNI( (U8) (c)) ]) \ - : UNICODE_REPLACEMENT) +#define toLOWER_LATIN1(c) ((! FITS_IN_8_BITS(c)) \ + ? (c) \ + : PL_latin1_lc[ (U8) (c) ]) +#define toLOWER_L1(c) toLOWER_LATIN1(c) /* Synonym for consistency */ + /* Modified uc. Is correct uc except for three non-ascii chars which are * all mapped to one of them, and these need special handling; error * character for input out-of-range */ -#define toUPPER_LATIN1_MOD(c) (FITS_IN_8_BITS(c) \ - ? UNI_TO_NATIVE(PL_mod_latin1_uc[ \ - NATIVE_TO_UNI( (U8) (c)) ]) \ - : UNICODE_REPLACEMENT) - +#define toUPPER_LATIN1_MOD(c) ((! FITS_IN_8_BITS(c)) \ + ? (c) \ + : PL_mod_latin1_uc[ (U8) (c) ]) #ifdef USE_NEXT_CTYPE -# define isALNUM_LC(c) \ - (NXIsAlNum((unsigned int)(c)) || (char)(c) == '_') -# define isIDFIRST_LC(c) \ - (NXIsAlpha((unsigned int)(c)) || (char)(c) == '_') +# define isALPHANUMERIC_LC(c) NXIsAlNum((unsigned int)(c)) # define isALPHA_LC(c) NXIsAlpha((unsigned int)(c)) # define isASCII_LC(c) isASCII((unsigned int)(c)) # define isBLANK_LC(c) isBLANK((unsigned int)(c)) -# define isSPACE_LC(c) NXIsSpace((unsigned int)(c)) -# define isDIGIT_LC(c) NXIsDigit((unsigned int)(c)) -# define isUPPER_LC(c) NXIsUpper((unsigned int)(c)) -# define isLOWER_LC(c) NXIsLower((unsigned int)(c)) -# define isALNUMC_LC(c) NXIsAlNum((unsigned int)(c)) # define isCNTRL_LC(c) NXIsCntrl((unsigned int)(c)) +# define isDIGIT_LC(c) NXIsDigit((unsigned int)(c)) # define isGRAPH_LC(c) NXIsGraph((unsigned int)(c)) +# define isIDFIRST_LC(c) (NXIsAlpha((unsigned int)(c)) || (char)(c) == '_') +# define isLOWER_LC(c) NXIsLower((unsigned int)(c)) # define isPRINT_LC(c) NXIsPrint((unsigned int)(c)) # define isPUNCT_LC(c) NXIsPunct((unsigned int)(c)) -# define toUPPER_LC(c) NXToUpper((unsigned int)(c)) +# define isSPACE_LC(c) NXIsSpace((unsigned int)(c)) +# define isUPPER_LC(c) NXIsUpper((unsigned int)(c)) +# define isWORDCHAR_LC(c) (NXIsAlNum((unsigned int)(c)) || (char)(c) == '_') +# define isXDIGIT_LC(c) NXIsXDigit((unsigned int)(c)) # define toLOWER_LC(c) NXToLower((unsigned int)(c)) +# define toUPPER_LC(c) NXToUpper((unsigned int)(c)) #else /* !USE_NEXT_CTYPE */ # if defined(CTYPE256) || (!defined(isascii) && !defined(HAS_ISASCII)) -/* Note that the foo_LC() macros in this case generally are defined only on - * code points 0-256, and give undefined, unwarned results if called with - * values outside that range */ +/* Use foo_LC_uvchr() instead of these for beyond the Latin1 range */ -# define isALNUM_LC(c) (isalnum((unsigned char)(c)) || (char)(c) == '_') -# define isIDFIRST_LC(c) (isalpha((unsigned char)(c)) || (char)(c) == '_') -# define isALPHA_LC(c) isalpha((unsigned char)(c)) +# define isALPHA_LC(c) (FITS_IN_8_BITS(c) && isalpha((unsigned char)(c))) +# define isALPHANUMERIC_LC(c) (FITS_IN_8_BITS(c) \ + && isalnum((unsigned char)(c))) # ifdef HAS_ISASCII -# define isASCII_LC(c) isascii((unsigned char)(c)) +# define isASCII_LC(c) (FITS_IN_8_BITS(c) && isascii((unsigned char)(c))) # else -# define isASCII_LC(c) isASCII((unsigned char)(c)) +# define isASCII_LC(c) (FITS_IN_8_BITS(c) && isASCII((unsigned char)(c))) # endif # ifdef HAS_ISBLANK -# define isBLANK_LC(c) isblank((unsigned char)(c)) +# define isBLANK_LC(c) (FITS_IN_8_BITS(c) && isblank((unsigned char)(c))) # else -# define isBLANK_LC(c) isBLANK((unsigned char)(c)) +# define isBLANK_LC(c) (FITS_IN_8_BITS(c) && isBLANK((unsigned char)(c))) # endif -# define isSPACE_LC(c) isspace((unsigned char)(c)) -# define isDIGIT_LC(c) isdigit((unsigned char)(c)) -# define isUPPER_LC(c) isupper((unsigned char)(c)) -# define isLOWER_LC(c) islower((unsigned char)(c)) -# define isALNUMC_LC(c) isalnum((unsigned char)(c)) -# define isCNTRL_LC(c) iscntrl((unsigned char)(c)) -# define isGRAPH_LC(c) isgraph((unsigned char)(c)) -# define isPRINT_LC(c) isprint((unsigned char)(c)) -# define isPUNCT_LC(c) ispunct((unsigned char)(c)) -# define toUPPER_LC(c) toupper((unsigned char)(c)) -# define toLOWER_LC(c) tolower((unsigned char)(c)) +# define isCNTRL_LC(c) (FITS_IN_8_BITS(c) && iscntrl((unsigned char)(c))) +# define isDIGIT_LC(c) (FITS_IN_8_BITS(c) && isdigit((unsigned char)(c))) +# define isGRAPH_LC(c) (FITS_IN_8_BITS(c) && isgraph((unsigned char)(c))) +# define isIDFIRST_LC(c) (FITS_IN_8_BITS(c) \ + && (isalpha((unsigned char)(c)) || (char)(c) == '_')) +# define isLOWER_LC(c) (FITS_IN_8_BITS(c) && islower((unsigned char)(c))) +# define isPRINT_LC(c) (FITS_IN_8_BITS(c) && isprint((unsigned char)(c))) +# define isPUNCT_LC(c) (FITS_IN_8_BITS(c) && ispunct((unsigned char)(c))) +# define isSPACE_LC(c) (FITS_IN_8_BITS(c) && isspace((unsigned char)(c))) +# define isUPPER_LC(c) (FITS_IN_8_BITS(c) && isupper((unsigned char)(c))) +# define isWORDCHAR_LC(c) (FITS_IN_8_BITS(c) \ + && (isalnum((unsigned char)(c)) || (char)(c) == '_')) +# define isXDIGIT_LC(c) (FITS_IN_8_BITS(c) && isxdigit((unsigned char)(c))) +# define toLOWER_LC(c) (FITS_IN_8_BITS(c) ? tolower((unsigned char)(c)) : (c)) +# define toUPPER_LC(c) (FITS_IN_8_BITS(c) ? toupper((unsigned char)(c)) : (c)) # else -# define isALNUM_LC(c) (isascii(c) && (isalnum(c) || (c) == '_')) -# define isIDFIRST_LC(c) (isascii(c) && (isalpha(c) || (c) == '_')) # define isALPHA_LC(c) (isascii(c) && isalpha(c)) +# define isALPHANUMERIC_LC(c) (isascii(c) && isalnum(c)) # define isASCII_LC(c) isascii(c) # ifdef HAS_ISBLANK # define isBLANK_LC(c) (isascii(c) && isblank(c)) # else -# define isBLANK_LC(c) isBLANK(c) +# define isBLANK_LC(c) isBLANK_A(c) # endif -# define isSPACE_LC(c) (isascii(c) && isspace(c)) -# define isDIGIT_LC(c) (isascii(c) && isdigit(c)) -# define isUPPER_LC(c) (isascii(c) && isupper(c)) -# define isLOWER_LC(c) (isascii(c) && islower(c)) -# define isALNUMC_LC(c) (isascii(c) && isalnum(c)) # define isCNTRL_LC(c) (isascii(c) && iscntrl(c)) +# define isDIGIT_LC(c) (isascii(c) && isdigit(c)) # define isGRAPH_LC(c) (isascii(c) && isgraph(c)) +# define isIDFIRST_LC(c) (isascii(c) && (isalpha(c) || (c) == '_')) +# define isLOWER_LC(c) (isascii(c) && islower(c)) # define isPRINT_LC(c) (isascii(c) && isprint(c)) # define isPUNCT_LC(c) (isascii(c) && ispunct(c)) -# define toUPPER_LC(c) toupper(c) -# define toLOWER_LC(c) tolower(c) +# define isSPACE_LC(c) (isascii(c) && isspace(c)) +# define isUPPER_LC(c) (isascii(c) && isupper(c)) +# define isWORDCHAR_LC(c) (isascii(c) && (isalnum(c) || (c) == '_')) +# define isXDIGIT_LC(c) (isascii(c) && isxdigit(c)) +# define toLOWER_LC(c) (isascii(c) ? tolower(c) : (c)) +# define toUPPER_LC(c) (isascii(c) ? toupper(c) : (c)) # endif #endif /* USE_NEXT_CTYPE */ -#define isPSXSPC_LC(c) (isSPACE_LC(c) || (c) == '\v') +#define isIDCONT(c) isWORDCHAR(c) +#define isIDCONT_A(c) isWORDCHAR_A(c) +#define isIDCONT_L1(c) isWORDCHAR_L1(c) +#define isIDCONT_LC(c) isWORDCHAR_LC(c) +#define isPSXSPC_LC(c) isSPACE_LC(c) + +/* For internal core Perl use only. If the input is Latin1, use the Latin1 + * macro; otherwise use the function 'above_latin1'. Won't compile if 'c' isn't unsigned, as + * won't match above_latin1 prototype. The macros do bounds checking, so have + * duplicate checks here, so could create versions of the macros that don't, + * but experiments show that gcc optimizes them out anyway. */ -/* For use in the macros just below. If the input is Latin1, use the Latin1 - * (_L1) version of the macro; otherwise use the function. Won't compile if - * 'c' isn't unsigned, as won't match function prototype. The macros do bounds - * checking, so have duplicate checks here, so could create versions of the - * macros that don't, but experiments show that gcc optimizes them out anyway. - */ -#define generic_uni(macro, function, c) ((c) < 256 \ - ? CAT2(macro, _L1)(c) \ - : function(c)) /* Note that all ignore 'use bytes' */ +#define _generic_uni(classnum, above_latin1, c) ((c) < 256 \ + ? _generic_isCC(c, classnum) \ + : above_latin1(c)) +#define _generic_swash_uni(classnum, c) ((c) < 256 \ + ? _generic_isCC(c, classnum) \ + : _is_uni_FOO(classnum, c)) +#define isALPHA_uni(c) _generic_swash_uni(_CC_ALPHA, c) +#define isALPHANUMERIC_uni(c) _generic_swash_uni(_CC_ALPHANUMERIC, c) +#define isASCII_uni(c) isASCII(c) +#define isBLANK_uni(c) _generic_uni(_CC_BLANK, is_HORIZWS_cp_high, c) +#define isCNTRL_uni(c) isCNTRL_L1(c) /* All controls are in Latin1 */ +#define isDIGIT_uni(c) _generic_swash_uni(_CC_DIGIT, c) +#define isGRAPH_uni(c) _generic_swash_uni(_CC_GRAPH, c) +#define isIDCONT_uni(c) _generic_uni(_CC_WORDCHAR, _is_uni_perl_idcont, c) +#define isIDFIRST_uni(c) _generic_uni(_CC_IDFIRST, _is_uni_perl_idstart, c) +#define isLOWER_uni(c) _generic_swash_uni(_CC_LOWER, c) +#define isPRINT_uni(c) _generic_swash_uni(_CC_PRINT, c) + +/* Posix and regular space are identical above Latin1 */ +#define isPSXSPC_uni(c) _generic_uni(_CC_PSXSPC, is_XPERLSPACE_cp_high, c) + +#define isPUNCT_uni(c) _generic_swash_uni(_CC_PUNCT, c) +#define isSPACE_uni(c) _generic_uni(_CC_SPACE, is_XPERLSPACE_cp_high, c) +#define isUPPER_uni(c) _generic_swash_uni(_CC_UPPER, c) +#define isVERTWS_uni(c) _generic_uni(_CC_VERTSPACE, is_VERTWS_cp_high, c) +#define isWORDCHAR_uni(c) _generic_swash_uni(_CC_WORDCHAR, c) +#define isXDIGIT_uni(c) _generic_uni(_CC_XDIGIT, is_XDIGIT_cp_high, c) -#define isALNUM_uni(c) generic_uni(isWORDCHAR, is_uni_alnum, c) -#define isBLANK_uni(c) generic_uni(isBLANK, is_uni_blank, c) -#define isIDFIRST_uni(c) generic_uni(isIDFIRST, is_uni_idfirst, c) -#define isALPHA_uni(c) generic_uni(isALPHA, is_uni_alpha, c) -#define isSPACE_uni(c) generic_uni(isSPACE, is_uni_space, c) -#define isDIGIT_uni(c) generic_uni(isDIGIT, is_uni_digit, c) -#define isUPPER_uni(c) generic_uni(isUPPER, is_uni_upper, c) -#define isLOWER_uni(c) generic_uni(isLOWER, is_uni_lower, c) -#define isASCII_uni(c) isASCII(c) -/* All controls are in Latin1 */ -#define isCNTRL_uni(c) ((c) < 256 && isCNTRL_L1(c)) -#define isGRAPH_uni(c) generic_uni(isGRAPH, is_uni_graph, c) -#define isPRINT_uni(c) generic_uni(isPRINT, is_uni_print, c) -#define isPUNCT_uni(c) generic_uni(isPUNCT, is_uni_punct, c) -#define isXDIGIT_uni(c) generic_uni(isXDIGIT, is_uni_xdigit, c) -#define toUPPER_uni(c,s,l) to_uni_upper(c,s,l) -#define toTITLE_uni(c,s,l) to_uni_title(c,s,l) -#define toLOWER_uni(c,s,l) to_uni_lower(c,s,l) #define toFOLD_uni(c,s,l) to_uni_fold(c,s,l) +#define toLOWER_uni(c,s,l) to_uni_lower(c,s,l) +#define toTITLE_uni(c,s,l) to_uni_title(c,s,l) +#define toUPPER_uni(c,s,l) to_uni_upper(c,s,l) -/* Posix and regular space differ only in U+000B, which is in Latin1 */ -#define isPSXSPC_uni(c) ((c) < 256 ? isPSXSPC_L1(c) : isSPACE_uni(c)) - -#define isALNUM_LC_uvchr(c) (c < 256 ? isALNUM_LC(c) : is_uni_alnum_lc(c)) -#define isIDFIRST_LC_uvchr(c) (c < 256 ? isIDFIRST_LC(c) : is_uni_idfirst_lc(c)) -#define isALPHA_LC_uvchr(c) (c < 256 ? isALPHA_LC(c) : is_uni_alpha_lc(c)) -#define isSPACE_LC_uvchr(c) (c < 256 ? isSPACE_LC(c) : is_uni_space_lc(c)) -#define isDIGIT_LC_uvchr(c) (c < 256 ? isDIGIT_LC(c) : is_uni_digit_lc(c)) -#define isUPPER_LC_uvchr(c) (c < 256 ? isUPPER_LC(c) : is_uni_upper_lc(c)) -#define isLOWER_LC_uvchr(c) (c < 256 ? isLOWER_LC(c) : is_uni_lower_lc(c)) -#define isCNTRL_LC_uvchr(c) (c < 256 ? isCNTRL_LC(c) : is_uni_cntrl_lc(c)) -#define isGRAPH_LC_uvchr(c) (c < 256 ? isGRAPH_LC(c) : is_uni_graph_lc(c)) -#define isPRINT_LC_uvchr(c) (c < 256 ? isPRINT_LC(c) : is_uni_print_lc(c)) -#define isPUNCT_LC_uvchr(c) (c < 256 ? isPUNCT_LC(c) : is_uni_punct_lc(c)) - -#define isPSXSPC_LC_uni(c) (isSPACE_LC_uni(c) ||(c) == '\f') -#define isBLANK_LC_uni(c) isBLANK(c) /* could be wrong */ - -/* For use in the macros just below. If the input is ASCII, use the ASCII (_A) - * version of the macro; if the input is in the upper Latin1 range, use the - * Latin1 (_L1) version of the macro, after converting from utf8; otherwise use - * the function. This relies on the fact that ASCII characters have the same - * representation whether utf8 or not */ -#define generic_utf8(macro, function, p) (isASCII(*(p)) \ - ? CAT2(CAT2(macro,_),A)(*(p)) \ - : (UTF8_IS_DOWNGRADEABLE_START(*(p))) \ - ? CAT2(macro, _L1) \ - (TWO_BYTE_UTF8_TO_UNI(*(p), \ - *((p)+1))) \ - : function(p)) - -/* Note that all assume that the utf8 has been validated, and ignore 'use +#define _generic_LC_uvchr(latin1, above_latin1, c) \ + (c < 256 ? latin1(c) : above_latin1(c)) +#define _generic_LC_swash_uvchr(latin1, classnum, c) \ + (c < 256 ? latin1(c) : _is_uni_FOO(classnum, c)) + +#define isALPHA_LC_uvchr(c) _generic_LC_swash_uvchr(isALPHA_LC, _CC_ALPHA, c) +#define isALPHANUMERIC_LC_uvchr(c) _generic_LC_swash_uvchr(isALPHANUMERIC_LC, \ + _CC_ALPHANUMERIC, c) +#define isASCII_LC_uvchr(c) isASCII_LC(c) +#define isBLANK_LC_uvchr(c) _generic_LC_uvchr(isBLANK_LC, is_HORIZWS_cp_high, c) +#define isCNTRL_LC_uvchr(c) (c < 256 ? isCNTRL_LC(c) : 0) +#define isDIGIT_LC_uvchr(c) _generic_LC_swash_uvchr(isDIGIT_LC, _CC_DIGIT, c) +#define isGRAPH_LC_uvchr(c) _generic_LC_swash_uvchr(isGRAPH_LC, _CC_GRAPH, c) +#define isIDCONT_LC_uvchr(c) _generic_LC_uvchr(isIDCONT_LC, \ + _is_uni_perl_idcont, c) +#define isIDFIRST_LC_uvchr(c) _generic_LC_uvchr(isIDFIRST_LC, \ + _is_uni_perl_idstart, c) +#define isLOWER_LC_uvchr(c) _generic_LC_swash_uvchr(isLOWER_LC, _CC_LOWER, c) +#define isPRINT_LC_uvchr(c) _generic_LC_swash_uvchr(isPRINT_LC, _CC_PRINT, c) +#define isPSXSPC_LC_uvchr(c) isSPACE_LC_uvchr(c) /* space is identical to posix + space under locale */ +#define isPUNCT_LC_uvchr(c) _generic_LC_swash_uvchr(isPUNCT_LC, _CC_PUNCT, c) +#define isSPACE_LC_uvchr(c) _generic_LC_uvchr(isSPACE_LC, \ + is_XPERLSPACE_cp_high, c) +#define isUPPER_LC_uvchr(c) _generic_LC_swash_uvchr(isUPPER_LC, _CC_UPPER, c) +#define isWORDCHAR_LC_uvchr(c) _generic_LC_swash_uvchr(isWORDCHAR_LC, \ + _CC_WORDCHAR, c) +#define isXDIGIT_LC_uvchr(c) _generic_LC_uvchr(isXDIGIT_LC, is_XDIGIT_cp_high, c) + +#define isBLANK_LC_uni(c) isBLANK_LC_uvchr(UNI_TO_NATIVE(c)) + +/* Everything whose name begins with an underscore is for internal core Perl + * use only. */ + +/* If the input is in the Latin1 range, use + * the Latin1 macro 'classnum' on 'p' which is a pointer to a UTF-8 string. + * Otherwise use the value given by the 'utf8' parameter. This relies on the + * fact that ASCII characters have the same representation whether utf8 or not. + * Note that it assumes that the utf8 has been validated, and ignores 'use * bytes' */ +#define _generic_utf8(classnum, p, utf8) (UTF8_IS_INVARIANT(*(p)) \ + ? _generic_isCC(*(p), classnum) \ + : (UTF8_IS_DOWNGRADEABLE_START(*(p))) \ + ? _generic_isCC( \ + TWO_BYTE_UTF8_TO_NATIVE(*(p), \ + *((p)+1 )), \ + classnum) \ + : utf8) +/* Like the above, but calls 'above_latin1(p)' to get the utf8 value. 'above_latin1' + * can be a macro */ +#define _generic_func_utf8(classnum, above_latin1, p) \ + _generic_utf8(classnum, p, above_latin1(p)) +/* Like the above, but passes classnum to _isFOO_utf8(), instead of having a + * 'above_latin1' parameter */ +#define _generic_swash_utf8(classnum, p) \ + _generic_utf8(classnum, p, _is_utf8_FOO(classnum, p)) + +/* Like the above, but should be used only when it is known that there are no + * characters in the range 128-255 which the class is TRUE for. Hence it can + * skip the tests for this range. 'above_latin1' should include its arguments */ +#define _generic_utf8_no_upper_latin1(classnum, p, above_latin1) \ + (UTF8_IS_INVARIANT(*(p)) \ + ? _generic_isCC(*(p), classnum) \ + : (UTF8_IS_ABOVE_LATIN1(*(p))) \ + ? above_latin1 \ + : 0) + +/* NOTE that some of these macros have very similar ones in regcharclass.h. + * For example, there is (at the time of this writing) an 'is_SPACE_utf8()' + * there, differing in name only by an underscore from the one here + * 'isSPACE_utf8(). The difference is that the ones here are probably more + * efficient and smaller, using an O(1) array lookup for Latin1-range code + * points; the regcharclass.h ones are implemented as a series of + * "if-else-if-else ..." */ + +#define isALPHA_utf8(p) _generic_swash_utf8(_CC_ALPHA, p) +#define isALPHANUMERIC_utf8(p) _generic_swash_utf8(_CC_ALPHANUMERIC, p) +#define isASCII_utf8(p) isASCII(*p) /* Because ASCII is invariant under + utf8, the non-utf8 macro works + */ +#define isBLANK_utf8(p) _generic_func_utf8(_CC_BLANK, is_HORIZWS_high, p) +#define isCNTRL_utf8(p) _generic_utf8(_CC_CNTRL, p, 0) +#define isDIGIT_utf8(p) _generic_utf8_no_upper_latin1(_CC_DIGIT, p, \ + _is_utf8_FOO(_CC_DIGIT, p)) +#define isGRAPH_utf8(p) _generic_swash_utf8(_CC_GRAPH, p) +#define isIDCONT_utf8(p) _generic_func_utf8(_CC_WORDCHAR, \ + _is_utf8_perl_idcont, p) -#define isALNUM_utf8(p) generic_utf8(isWORDCHAR, is_utf8_alnum, p) /* To prevent S_scan_word in toke.c from hanging, we have to make sure that * IDFIRST is an alnum. See * http://rt.perl.org/rt3/Ticket/Display.html?id=74022 for more detail than you - * ever wanted to know about. XXX It is unclear if this should extend to - * isIDFIRST_uni() which it hasn't so far. (In the ASCII range, there isn't a - * difference.) This used to be not the XID version, but we decided to go with - * the more modern Unicode definition */ -#define isIDFIRST_utf8(p) (isASCII(*(p)) \ - ? isIDFIRST_A(*(p)) \ - : (UTF8_IS_DOWNGRADEABLE_START(*(p))) \ - ? isIDFIRST_L1(TWO_BYTE_UTF8_TO_UNI(*(p), \ - *((p)+1)))\ - : Perl__is_utf8__perl_idstart(aTHX_ p)) -#define isIDCONT_utf8(p) generic_utf8(isWORDCHAR, is_utf8_xidcont, p) -#define isALPHA_utf8(p) generic_utf8(isALPHA, is_utf8_alpha, p) -#define isBLANK_utf8(p) generic_utf8(isBLANK, is_utf8_blank, p) -#define isSPACE_utf8(p) generic_utf8(isSPACE, is_utf8_space, p) -#define isDIGIT_utf8(p) generic_utf8(isDIGIT, is_utf8_digit, p) -#define isUPPER_utf8(p) generic_utf8(isUPPER, is_utf8_upper, p) -#define isLOWER_utf8(p) generic_utf8(isLOWER, is_utf8_lower, p) -/* Because ASCII is invariant under utf8, the non-utf8 macro works */ -#define isASCII_utf8(p) isASCII(*p) -#define isCNTRL_utf8(p) generic_utf8(isCNTRL, is_utf8_cntrl, p) -#define isGRAPH_utf8(p) generic_utf8(isGRAPH, is_utf8_graph, p) -#define isPRINT_utf8(p) generic_utf8(isPRINT, is_utf8_print, p) -#define isPUNCT_utf8(p) generic_utf8(isPUNCT, is_utf8_punct, p) -#define isXDIGIT_utf8(p) generic_utf8(isXDIGIT, is_utf8_xdigit, p) -#define toUPPER_utf8(p,s,l) to_utf8_upper(p,s,l) -#define toTITLE_utf8(p,s,l) to_utf8_title(p,s,l) + * ever wanted to know about. (In the ASCII range, there isn't a difference.) + * This used to be not the XID version, but we decided to go with the more + * modern Unicode definition */ +#define isIDFIRST_utf8(p) _generic_func_utf8(_CC_IDFIRST, \ + _is_utf8_perl_idstart, p) + +#define isLOWER_utf8(p) _generic_swash_utf8(_CC_LOWER, p) +#define isPRINT_utf8(p) _generic_swash_utf8(_CC_PRINT, p) + +/* Posix and regular space are identical above Latin1 */ +#define isPSXSPC_utf8(p) _generic_func_utf8(_CC_PSXSPC, is_XPERLSPACE_high, p) + +#define isPUNCT_utf8(p) _generic_swash_utf8(_CC_PUNCT, p) +#define isSPACE_utf8(p) _generic_func_utf8(_CC_SPACE, is_XPERLSPACE_high, p) +#define isUPPER_utf8(p) _generic_swash_utf8(_CC_UPPER, p) +#define isVERTWS_utf8(p) _generic_func_utf8(_CC_VERTSPACE, is_VERTWS_high, p) +#define isWORDCHAR_utf8(p) _generic_swash_utf8(_CC_WORDCHAR, p) +#define isXDIGIT_utf8(p) _generic_utf8_no_upper_latin1(_CC_XDIGIT, p, \ + is_XDIGIT_high(p)) + +#define toFOLD_utf8(p,s,l) to_utf8_fold(p,s,l) #define toLOWER_utf8(p,s,l) to_utf8_lower(p,s,l) +#define toTITLE_utf8(p,s,l) to_utf8_title(p,s,l) +#define toUPPER_utf8(p,s,l) to_utf8_upper(p,s,l) -/* Posix and regular space differ only in U+000B, which is in ASCII (and hence - * Latin1 */ -#define isPSXSPC_utf8(p) ((isASCII(*(p))) \ - ? isPSXSPC_A(*(p)) \ - : (UTF8_IS_DOWNGRADEABLE_START(*(p)) \ - ? isPSXSPC_L1(TWO_BYTE_UTF8_TO_UNI(*(p), \ - *((p)+1)))\ - : isSPACE_utf8(p))) -#define isALNUM_LC_utf8(p) isALNUM_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isIDFIRST_LC_utf8(p) isIDFIRST_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isALPHA_LC_utf8(p) isALPHA_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isBLANK_LC_utf8(p) isBLANK_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isSPACE_LC_utf8(p) isSPACE_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isDIGIT_LC_utf8(p) isDIGIT_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isUPPER_LC_utf8(p) isUPPER_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isLOWER_LC_utf8(p) isLOWER_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isALNUMC_LC_utf8(p) isALNUMC_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isCNTRL_LC_utf8(p) isCNTRL_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isGRAPH_LC_utf8(p) isGRAPH_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isPRINT_LC_utf8(p) isPRINT_LC_uvchr(valid_utf8_to_uvchr(p, 0)) -#define isPUNCT_LC_utf8(p) isPUNCT_LC_uvchr(valid_utf8_to_uvchr(p, 0)) - -#define isPSXSPC_LC_utf8(c) (isSPACE_LC_utf8(c) ||(c) == '\f') +/* For internal core Perl use only. If the input is in the Latin1 range, use + * the macro 'macro' on 'p' which is a pointer to a UTF-8 string. Otherwise + * use the value given by the 'utf8' parameter. This relies on the fact that + * ASCII characters have the same representation whether utf8 or not. Note + * that it assumes that the utf8 has been validated, and ignores 'use bytes' */ +#define _generic_LC_utf8(macro, p, utf8) \ + (UTF8_IS_INVARIANT(*(p)) \ + ? macro(*(p)) \ + : (UTF8_IS_DOWNGRADEABLE_START(*(p))) \ + ? macro(TWO_BYTE_UTF8_TO_NATIVE(*(p), *((p)+1))) \ + : utf8) + +#define _generic_LC_swash_utf8(macro, classnum, p) \ + _generic_LC_utf8(macro, p, _is_utf8_FOO(classnum, p)) +#define _generic_LC_func_utf8(macro, above_latin1, p) \ + _generic_LC_utf8(macro, p, above_latin1(p)) + +#define isALPHANUMERIC_LC_utf8(p) _generic_LC_swash_utf8(isALPHANUMERIC_LC, \ + _CC_ALPHANUMERIC, p) +#define isALPHA_LC_utf8(p) _generic_LC_swash_utf8(isALPHA_LC, _CC_ALPHA, p) +#define isASCII_LC_utf8(p) isASCII_LC(*p) +#define isBLANK_LC_utf8(p) _generic_LC_func_utf8(isBLANK_LC, is_HORIZWS_high, p) +#define isCNTRL_LC_utf8(p) _generic_LC_utf8(isCNTRL_LC, p, 0) +#define isDIGIT_LC_utf8(p) _generic_LC_swash_utf8(isDIGIT_LC, _CC_DIGIT, p) +#define isGRAPH_LC_utf8(p) _generic_LC_swash_utf8(isGRAPH_LC, _CC_GRAPH, p) +#define isIDCONT_LC_utf8(p) _generic_LC_func_utf8(isIDCONT_LC, _is_utf8_perl_idcont, p) +#define isIDFIRST_LC_utf8(p) _generic_LC_func_utf8(isIDFIRST_LC, _is_utf8_perl_idstart, p) +#define isLOWER_LC_utf8(p) _generic_LC_swash_utf8(isLOWER_LC, _CC_LOWER, p) +#define isPRINT_LC_utf8(p) _generic_LC_swash_utf8(isPRINT_LC, _CC_PRINT, p) +#define isPSXSPC_LC_utf8(p) isSPACE_LC_utf8(p) /* space is identical to posix + space under locale */ +#define isPUNCT_LC_utf8(p) _generic_LC_swash_utf8(isPUNCT_LC, _CC_PUNCT, p) +#define isSPACE_LC_utf8(p) _generic_LC_func_utf8(isSPACE_LC, is_XPERLSPACE_high, p) +#define isUPPER_LC_utf8(p) _generic_LC_swash_utf8(isUPPER_LC, _CC_UPPER, p) +#define isWORDCHAR_LC_utf8(p) _generic_LC_swash_utf8(isWORDCHAR_LC, \ + _CC_WORDCHAR, p) +#define isXDIGIT_LC_utf8(p) _generic_LC_func_utf8(isXDIGIT_LC, is_XDIGIT_high, p) + +/* Macros for backwards compatibility and for completeness when the ASCII and + * Latin1 values are identical */ +#define isALPHAU(c) isALPHA_L1(c) +#define isDIGIT_L1(c) isDIGIT_A(c) +#define isOCTAL(c) isOCTAL_A(c) +#define isOCTAL_L1(c) isOCTAL_A(c) +#define isXDIGIT_L1(c) isXDIGIT_A(c) +#define isALNUM(c) isWORDCHAR(c) +#define isALNUMU(c) isWORDCHAR_L1(c) +#define isALNUM_LC(c) isWORDCHAR_LC(c) +#define isALNUM_uni(c) isWORDCHAR_uni(c) +#define isALNUM_LC_uvchr(c) isWORDCHAR_LC_uvchr(c) +#define isALNUM_utf8(p) isWORDCHAR_utf8(p) +#define isALNUM_LC_utf8(p) isWORDCHAR_LC_utf8(p) +#define isALNUMC_A(c) isALPHANUMERIC_A(c) /* Mnemonic: "C's alnum" */ +#define isALNUMC_L1(c) isALPHANUMERIC_L1(c) +#define isALNUMC(c) isALPHANUMERIC(c) +#define isALNUMC_LC(c) isALPHANUMERIC_LC(c) +#define isALNUMC_uni(c) isALPHANUMERIC_uni(c) +#define isALNUMC_LC_uvchr(c) isALPHANUMERIC_LC_uvchr(c) +#define isALNUMC_utf8(p) isALPHANUMERIC_utf8(p) +#define isALNUMC_LC_utf8(p) isALPHANUMERIC_LC_utf8(p) /* This conversion works both ways, strangely enough. On EBCDIC platforms, - * CTRL-@ is 0, CTRL-A is 1, etc, just like on ASCII */ -# define toCTRL(c) (toUPPER(NATIVE_TO_UNI(c)) ^ 64) + * CTRL-@ is 0, CTRL-A is 1, etc, just like on ASCII, except that they don't + * necessarily mean the same characters, e.g. CTRL-D is 4 on both systems, but + * that is EOT on ASCII; ST on EBCDIC */ +# define toCTRL(c) (toUPPER(NATIVE_TO_LATIN1(c)) ^ 64) /* Line numbers are unsigned, 32 bits. */ typedef U32 line_t; -#define NOLINE ((line_t) 4294967295UL) +#define NOLINE ((line_t) 4294967295UL) /* = FFFFFFFF */ /* Helpful alias for version prescan */ #define is_LAX_VERSION(a,b) \ @@ -1050,6 +1554,8 @@ typedef U32 line_t; } \ return a; +#define READ_XDIGIT(s) (isALPHA(*(s)) ? ((*(s)++ + 9) & 0xf) : (*(s)++ & 0xf)) + /* =head1 Memory Management @@ -1082,8 +1588,8 @@ The XSUB-writer's interface to the C C function. =for apidoc Am|void|Move|void* src|void* dest|int nitems|type The XSUB-writer's interface to the C C function. The C is the -source, C is the destination, C is the number of items, and C is -the type. Can do overlapping moves. See also C. +source, C is the destination, C is the number of items, and +C is the type. Can do overlapping moves. See also C. =for apidoc Am|void *|MoveD|void* src|void* dest|int nitems|type Like C but returns dest. Useful for encouraging compilers to tail-call @@ -1091,8 +1597,8 @@ optimise. =for apidoc Am|void|Copy|void* src|void* dest|int nitems|type The XSUB-writer's interface to the C C function. The C is the -source, C is the destination, C is the number of items, and C is -the type. May fail on overlapping copies. See also C. +source, C is the destination, C is the number of items, and +C is the type. May fail on overlapping copies. See also C. =for apidoc Am|void *|CopyD|void* src|void* dest|int nitems|type @@ -1142,13 +1648,13 @@ PoisonWith(0xEF) for catching access to freed memory. * overly eager compilers that will bleat about e.g. * (U16)n > (size_t)~0/sizeof(U16) always being false. */ #ifdef PERL_MALLOC_WRAP -#define MEM_WRAP_CHECK(n,t) MEM_WRAP_CHECK_1(n,t,PL_memory_wrap) +#define MEM_WRAP_CHECK(n,t) \ + (void)(sizeof(t) > 1 && ((MEM_SIZE)(n)+0.0) > MEM_SIZE_MAX/sizeof(t) && (croak_memory_wrap(),0)) #define MEM_WRAP_CHECK_1(n,t,a) \ (void)(sizeof(t) > 1 && ((MEM_SIZE)(n)+0.0) > MEM_SIZE_MAX/sizeof(t) && (Perl_croak_nocontext("%s",(a)),0)) #define MEM_WRAP_CHECK_(n,t) MEM_WRAP_CHECK(n,t), -#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))) - +#define PERL_STRLEN_ROUNDUP(n) ((void)(((n) > MEM_SIZE_MAX - 2 * PERL_STRLEN_ROUNDUP_QUANTUM) ? (croak_memory_wrap(),0):0),((n-1+PERL_STRLEN_ROUNDUP_QUANTUM)&~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM-1))) #else #define MEM_WRAP_CHECK(n,t) @@ -1274,6 +1780,12 @@ void Perl_mem_log_del_sv(const SV *sv, const char *filename, const int linenumbe #define PoisonFree(d,n,t) PoisonWith(d,n,t,0xEF) #define Poison(d,n,t) PoisonFree(d,n,t) +#ifdef PERL_POISON +# define PERL_POISON_EXPR(x) x +#else +# define PERL_POISON_EXPR(x) +#endif + #ifdef USE_STRUCT_COPY #define StructCopy(s,d,t) (*((t*)(d)) = *((t*)(s))) #else @@ -1319,6 +1831,39 @@ void Perl_mem_log_del_sv(const SV *sv, const char *filename, const int linenumbe # define deprecate(s) Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), "Use of " s " is deprecated") #endif +/* Internal macros to deal with gids and uids */ +#ifdef PERL_CORE + +# if Uid_t_size > IVSIZE +# define sv_setuid(sv, uid) sv_setnv((sv), (NV)(uid)) +# define SvUID(sv) SvNV(sv) +# else +# if Uid_t_sign <= 0 +# define sv_setuid(sv, uid) sv_setiv((sv), (IV)(uid)) +# define SvUID(sv) SvIV(sv) +# else +# define sv_setuid(sv, uid) sv_setuv((sv), (UV)(uid)) +# define SvUID(sv) SvUV(sv) +# endif +# endif /* Uid_t_size */ + +# if Gid_t_size > IVSIZE +# define sv_setgid(sv, gid) sv_setnv((sv), (NV)(gid)) +# define SvGID(sv) SvNV(sv) +# else +# if Gid_t_sign <= 0 +# define sv_setgid(sv, gid) sv_setiv((sv), (IV)(gid)) +# define SvGID(sv) SvIV(sv) +# else +# define sv_setgid(sv, gid) sv_setuv((sv), (UV)(gid)) +# define SvGID(sv) SvUV(sv) +# endif +# endif /* Gid_t_size */ + +#endif + +#endif /* HANDY_H */ + /* * Local variables: * c-indentation-style: bsd