+# 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) ? (U8)((c) + ('a' - 'A')) : (c))
+#define toUPPER(c) (isLOWER(c) ? (U8)((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 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; returns the input itself if is out-of-range */
+#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 */