/* 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) ((bool)!!(cbool))
+#define cBOOL(cbool) ((cbool) ? (bool)1 : (bool)0)
/* Try to figure out __func__ or __FUNCTION__ equivalent, if any.
* XXX Should really be a Configure probe, with HAS__FUNCTION__
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.
+locale and by C<use bytes>.
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
* compiler to optimize it out if possible. This is because Configure makes
* sure that the machine has an 8-bit byte, so if c is stored in a byte, the
* sizeof() guarantees that this evaluates to a constant true at compile time.
- * The use of the mask instead of '< 256' keeps gcc from complaining that it is
- * always true, when c's storage class is a byte. */
-#define FITS_IN_8_BITS(c) ((sizeof(c) == 1) \
- || (((WIDEST_UTYPE)(c) & 0xFF) == (WIDEST_UTYPE)(c)))
+ */
+#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 : 0)
+# define isASCII(c) (FITS_IN_8_BITS(c) && (NATIVE_TO_UNI((U8) (c)) < 128))
#else
# define isASCII(c) ((WIDEST_UTYPE)(c) < 128)
#endif
/* ASCII range only */
#ifdef H_PERL /* If have access to perl.h, lookup in its table */
-/* Bits for PL_charclass[] */
+/* 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_ALNUMC_A (1<<0)
# define _CC_ALNUMC_L1 (1<<1)
# define _CC_ALPHA_A (1<<2)
# define _CC_NONLATIN1_FOLD (1<<30)
/* Unused
* (1<<31)
+ * 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 occupy a single range like OCTAL_A, DIGIT_A,
+ * UPPER_A, and LOWER_A.
*/
# ifdef DOINIT
* 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))
#else /* No perl.h. */
-# define isOCTAL_A(c) ((c) >= '0' && (c) <= '9')
+# define isOCTAL_A(c) ((c) <= '7' && (c) >= '0')
# ifdef EBCDIC
# define isALNUMC_A(c) (isASCII(c) && isALNUMC(c))
# define isALPHA_A(c) (isASCII(c) && isALPHA(c))
# define isALNUMC_A(c) (isALPHA_A(c) || isDIGIT_A(c))
# 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) : 0)
-# define isDIGIT_A(c) ((c) >= '0' && (c) <= '9')
+# define isCNTRL_A(c) (FITS_IN_8_BITS(c) && ((U8) (c) < ' ' || (c) == 127))
+# 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 isLOWER_A(c) ((c) >= 'a' && (c) <= 'z')
# 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 isUPPER_A(c) ((c) >= 'A' && (c) <= 'Z')
+# 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) >= 'A' && (c) <= 'F'))
+# define isXDIGIT_A(c) (isDIGIT_A(c) || ((c) >= 'a' && (c) <= 'f') || ((c) <= 'F' && (c) >= 'A'))
# endif
#endif /* ASCII range definitions */
#define generic_uni(macro, function, c) ((c) < 256 \
? CAT2(macro, _L1)(c) \
: function(c))
+/* Note that all ignore 'use bytes' */
#define isALNUM_uni(c) generic_uni(isWORDCHAR, is_uni_alnum, c)
#define isIDFIRST_uni(c) generic_uni(isIDFIRST, is_uni_idfirst, 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) : 0)
+#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)
* 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(macro, _A)(*(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 ignore 'use bytes' */
+
#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
: (UTF8_IS_DOWNGRADEABLE_START(*(p))) \
? isIDFIRST_L1(TWO_BYTE_UTF8_TO_UNI(*(p), \
*((p)+1)))\
- : _is_utf8__perl_idstart(p))
+ : 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 isSPACE_utf8(p) generic_utf8(isSPACE, is_utf8_space, p)