win32.c: Add mutexes around some calls

[perl5.git] / handy.h

/*    handy.h
 *
 *    Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1999, 2000,
 *    2001, 2002, 2004, 2005, 2006, 2007, 2008, 2012 by Larry Wall and others
 *
 *    You may distribute under the terms of either the GNU General Public
 *    License or the Artistic License, as specified in the README file.
 *
 */

/* IMPORTANT NOTE: Everything whose name begins with an underscore is for
 * internal core Perl use only. */

#ifndef PERL_HANDY_H_ /* Guard against nested #inclusion */
#define PERL_HANDY_H_

#ifndef PERL_CORE
#  define Null(type) ((type)NULL)

/*
=for apidoc_section $string
=for apidoc AmnU||Nullch
Null character pointer.  (No longer available when C<PERL_CORE> is
defined.)

=for apidoc_section $SV
=for apidoc AmnU||Nullsv
Null SV pointer.  (No longer available when C<PERL_CORE> is defined.)

=cut

Below are signatures of functions from config.h which can't easily be gleaned
from it, and are very unlikely to change

=for apidoc_section $signals
=for apidoc Am|int|Sigsetjmp|jmp_buf env|int savesigs
=for apidoc Am|void|Siglongjmp|jmp_buf env|int val

=for apidoc_section $filesystem
=for apidoc Am|void *|FILE_ptr|FILE * f
=for apidoc Am|Size_t|FILE_cnt|FILE * f
=for apidoc Am|void *|FILE_base|FILE * f
=for apidoc Am|Size_t|FILE_bufsiz|FILE *f

=for apidoc_section $string
=for apidoc Amu|token|CAT2|token x|token y
=for apidoc Amu|string|STRINGIFY|token x

=for apidoc_section $numeric
=for apidoc Am|double|Drand01
=for apidoc Am|void|seedDrand01|Rand_seed_t x
=for apidoc Am|char *|Gconvert|double x|Size_t n|bool t|char * b

=cut
*/

#  define Nullch Null(char*)
#  define Nullfp Null(PerlIO*)
#  define Nullsv Null(SV*)
#endif

#ifdef TRUE
#undef TRUE
#endif
#ifdef FALSE
#undef FALSE
#endif
#define TRUE (1)
#define FALSE (0)

/*
=for apidoc_section $SV
=for apidoc   Am |AV *  |MUTABLE_AV |AV * p
=for apidoc_item |CV *  |MUTABLE_CV |CV * p
=for apidoc_item |GV *  |MUTABLE_GV |GV * p
=for apidoc_item |HV *  |MUTABLE_HV |HV * p
=for apidoc_item |IO *  |MUTABLE_IO |IO * p
=for apidoc_item |void *|MUTABLE_PTR|void * p
=for apidoc_item |SV *  |MUTABLE_SV |SV * p

The C<MUTABLE_I<*>>() macros cast pointers to the types shown, in such a way
(compiler permitting) that casting away const-ness will give a warning;
e.g.:

 const SV *sv = ...;
 AV *av1 = (AV*)sv;        <== BAD:  the const has been silently
                                     cast away
 AV *av2 = MUTABLE_AV(sv); <== GOOD: it may warn

C<MUTABLE_PTR> is the base macro used to derive new casts.  The other
already-built-in ones return pointers to what their names indicate.

=cut

The brace group version will raise a diagnostic if 'p' is const; the other
blindly casts away const.
 */
#if defined(PERL_USE_GCC_BRACE_GROUPS)
#  define MUTABLE_PTR(p) ({ void *p_ = (p); p_; })
#else
#  define MUTABLE_PTR(p) ((void *) (p))
#endif

#define MUTABLE_AV(p)	((AV *)MUTABLE_PTR(p))
#define MUTABLE_CV(p)	((CV *)MUTABLE_PTR(p))
#define MUTABLE_GV(p)	((GV *)MUTABLE_PTR(p))
#define MUTABLE_HV(p)	((HV *)MUTABLE_PTR(p))
#define MUTABLE_IO(p)	((IO *)MUTABLE_PTR(p))
#define MUTABLE_SV(p)	((SV *)MUTABLE_PTR(p))

#ifndef __cplusplus
#  include <stdbool.h>
#endif

/*
=for apidoc_section $casting
=for apidoc Am|bool|cBOOL|bool expr

Cast-to-bool.  When Perl was able to be compiled on pre-C99 compilers, a
C<(bool)> cast didn't necessarily do the right thing, so this macro was
created (and made somewhat complicated to work around bugs in old
compilers).  Now, many years later, and C99 is used, this is no longer
required, but is kept for backwards compatibility.

=cut
*/
#define cBOOL(cbool) ((bool) (cbool))

/* Try to figure out __func__ or __FUNCTION__ equivalent, if any.
 * XXX Should really be a Configure probe, with HAS__FUNCTION__
 *     and FUNCTION__ as results.
 * XXX Similarly, a Configure probe for __FILE__ and __LINE__ is needed. */
#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) || (defined(__SUNPRO_C)) /* C99 or close enough. */
#  define FUNCTION__ __func__
#  define SAFE_FUNCTION__ __func__
#elif (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__ ("")
#  define SAFE_FUNCTION__ ("UNKNOWN")
#else
#  define FUNCTION__ __FUNCTION__ /* Common extension. */
#  define SAFE_FUNCTION__ __FUNCTION__ /* Common extension. */
#endif

/* XXX A note on the perl source internal type system.  The
   original intent was that I32 be *exactly* 32 bits.

   Currently, we only guarantee that I32 is *at least* 32 bits.
   Specifically, if int is 64 bits, then so is I32.  (This is the case
   for the Cray.)  This has the advantage of meshing nicely with
   standard library calls (where we pass an I32 and the library is
   expecting an int), but the disadvantage that an I32 is not 32 bits.
   Andy Dougherty	August 1996

   There is no guarantee that there is *any* integral type with
   exactly 32 bits.  It is perfectly legal for a system to have
   sizeof(short) == sizeof(int) == sizeof(long) == 8.

   Similarly, there is no guarantee that I16 and U16 have exactly 16
   bits.

   For dealing with issues that may arise from various 32/64-bit
   systems, we will ask Configure to check out

        SHORTSIZE == sizeof(short)
        INTSIZE == sizeof(int)
        LONGSIZE == sizeof(long)
        LONGLONGSIZE == sizeof(long long) (if HAS_LONG_LONG)
        PTRSIZE == sizeof(void *)
        DOUBLESIZE == sizeof(double)
        LONG_DOUBLESIZE == sizeof(long double) (if HAS_LONG_DOUBLE).

*/

#ifdef I_INTTYPES /* e.g. Linux has int64_t without <inttypes.h> */
#   include <inttypes.h>
#   ifdef INT32_MIN_BROKEN
#       undef  INT32_MIN
#       define INT32_MIN (-2147483647-1)
#   endif
#   ifdef INT64_MIN_BROKEN
#       undef  INT64_MIN
#       define INT64_MIN (-9223372036854775807LL-1)
#   endif
#endif

typedef I8TYPE I8;
typedef U8TYPE U8;
typedef I16TYPE I16;
typedef U16TYPE U16;
typedef I32TYPE I32;
typedef U32TYPE U32;

#ifdef QUADKIND
typedef I64TYPE I64;
typedef U64TYPE U64;
#endif

/* I8_MAX and I8_MIN constants are not defined, as I8 is an ambiguous type.
   Please search CHAR_MAX in perl.h for further details. */
#ifdef UINT8_MAX
#  define U8_MAX UINT8_MAX
#else
#  define U8_MAX PERL_UCHAR_MAX
#endif
#ifdef UINT8_MIN
#  define U8_MIN UINT8_MIN
#else
#  define U8_MIN PERL_UCHAR_MIN
#endif

#ifdef INT16_MAX
#  define I16_MAX INT16_MAX
#else
#  define I16_MAX PERL_SHORT_MAX
#endif
#ifdef INT16_MIN
#  define I16_MIN INT16_MIN
#else
#  define I16_MIN PERL_SHORT_MIN
#endif
#ifdef UINT16_MAX
#  define U16_MAX UINT16_MAX
#else
#  define U16_MAX PERL_USHORT_MAX
#endif
#ifdef UINT16_MIN
#  define U16_MIN UINT16_MIN
#else
#  define U16_MIN PERL_USHORT_MIN
#endif

#ifdef INT32_MAX
#  define I32_MAX INT32_MAX
#elif LONGSIZE > 4
#  define I32_MAX PERL_INT_MAX
#else
#  define I32_MAX PERL_LONG_MAX
#endif
#ifdef INT32_MIN
#  define I32_MIN INT32_MIN
#elif LONGSIZE > 4
#  define I32_MIN PERL_INT_MIN
#else
#  define I32_MIN PERL_LONG_MIN
#endif
#ifdef UINT32_MAX
#  ifndef UINT32_MAX_BROKEN /* e.g. HP-UX with gcc messes this up */
#    define U32_MAX UINT_MAX
#  else
#    define U32_MAX 4294967295U
#  endif
#elif LONGSIZE > 4
#  define U32_MAX PERL_UINT_MAX
#else
#  define U32_MAX PERL_ULONG_MAX
#endif
#ifdef UINT32_MIN
#  define U32_MIN UINT32_MIN
#elif LONGSIZE > 4
#  define U32_MIN PERL_UINT_MIN
#else
#  define U32_MIN PERL_ULONG_MIN
#endif

/*
=for apidoc_section $integer
=for apidoc Ay|| PERL_INT_FAST8_T
=for apidoc_item PERL_INT_FAST16_T
=for apidoc_item PERL_UINT_FAST8_T
=for apidoc_item PERL_UINT_FAST16_T

These are equivalent to the correspondingly-named C99 typedefs on platforms
that have those; they evaluate to C<int> and C<unsigned int> on platforms that
don't, so that you can portably take advantage of this C99 feature.

=cut
*/
#ifdef I_STDINT
    typedef  int_fast8_t  PERL_INT_FAST8_T;
    typedef uint_fast8_t  PERL_UINT_FAST8_T;
    typedef  int_fast16_t PERL_INT_FAST16_T;
    typedef uint_fast16_t PERL_UINT_FAST16_T;
#else
    typedef int           PERL_INT_FAST8_T;
    typedef unsigned int  PERL_UINT_FAST8_T;
    typedef int           PERL_INT_FAST16_T;
    typedef unsigned int  PERL_UINT_FAST16_T;
#endif

/* log(2) (i.e., log base 10 of 2) is pretty close to 0.30103, just in case
 * anyone is grepping for it.  So BIT_DIGITS gives the number of decimal digits
 * required to represent any possible unsigned number containing N bits.
 * TYPE_DIGITS gives the number of decimal digits required to represent any
 * possible unsigned number of type T. */
#define BIT_DIGITS(N)   (((N)*146)/485 + 1)  /* log10(2) =~ 146/485 */
#define TYPE_DIGITS(T)  BIT_DIGITS(sizeof(T) * 8)
#define TYPE_CHARS(T)   (TYPE_DIGITS(T) + 2) /* sign, NUL */

/* Unused by core; should be deprecated */
#define Ctl(ch) ((ch) & 037)

#if defined(PERL_CORE) || defined(PERL_EXT)
#  ifndef MIN
#    define MIN(a,b) ((a) < (b) ? (a) : (b))
#  endif
#  ifndef MAX
#    define MAX(a,b) ((a) > (b) ? (a) : (b))
#  endif
#endif

/* Returns a boolean as to whether the input unsigned number is a power of 2
 * (2**0, 2**1, etc).  In other words if it has just a single bit set.
 * If not, subtracting 1 would leave the uppermost bit set, so the & would
 * yield non-zero */
#if defined(PERL_CORE) || defined(PERL_EXT)
#  define isPOWER_OF_2(n) ((n) && ((n) & ((n)-1)) == 0)
#endif

/* Returns a mask with the lowest n bits set */
#define nBIT_MASK(n) ((UINTMAX_C(1) << (n)) - 1)

/* The largest unsigned number that will fit into n bits */
#define nBIT_UMAX(n)  nBIT_MASK(n)

/*
=for apidoc_section $directives
=for apidoc Am||__ASSERT_|bool expr

This is a helper macro to avoid preprocessor issues, replaced by nothing
unless under DEBUGGING, where it expands to an assert of its argument,
followed by a comma (hence the comma operator).  If we just used a straight
assert(), we would get a comma with nothing before it when not DEBUGGING.

=cut

We also use empty definition under Coverity since the __ASSERT_
checks often check for things that Really Cannot Happen, and Coverity
detects that and gets all excited. */

#if   defined(DEBUGGING) && !defined(__COVERITY__)                        \
 && ! defined(PERL_SMALL_MACRO_BUFFER)
#   define __ASSERT_(statement)  assert(statement),
#else
#   define __ASSERT_(statement)
#endif

/*
=for apidoc_section $SV

=for apidoc Ama|SV*|newSVpvs|"literal string"
Like C<newSVpvn>, but takes a literal string instead of a
string/length pair.

=for apidoc Ama|SV*|newSVpvs_flags|"literal string"|U32 flags
Like C<newSVpvn_flags>, but takes a literal string instead of
a string/length pair.

=for apidoc Ama|SV*|newSVpvs_share|"literal string"
Like C<newSVpvn_share>, but takes a literal string instead of
a string/length pair and omits the hash parameter.

=for apidoc Am|void|sv_catpvs_flags|SV* sv|"literal string"|I32 flags
Like C<sv_catpvn_flags>, but takes a literal string instead
of a string/length pair.

=for apidoc Am|void|sv_catpvs_nomg|SV* sv|"literal string"
Like C<sv_catpvn_nomg>, but takes a literal string instead of
a string/length pair.

=for apidoc Am|void|sv_catpvs|SV* sv|"literal string"
Like C<sv_catpvn>, but takes a literal string instead of a
string/length pair.

=for apidoc Am|void|sv_catpvs_mg|SV* sv|"literal string"
Like C<sv_catpvn_mg>, but takes a literal string instead of a
string/length pair.

=for apidoc Am|SV *|sv_setref_pvs|SV *const rv|const char *const classname|"literal string"
Like C<sv_setref_pvn>, but takes a literal string instead of
a string/length pair.

=for apidoc_section $string

=for apidoc Ama|char*|savepvs|"literal string"
Like C<savepvn>, but takes a literal string instead of a
string/length pair.

=for apidoc Ama|char*|savesharedpvs|"literal string"
A version of C<savepvs()> which allocates the duplicate string in memory
which is shared between threads.

=for apidoc_section $GV

=for apidoc Am|HV*|gv_stashpvs|"name"|I32 create
Like C<gv_stashpvn>, but takes a literal string instead of a
string/length pair.

=for apidoc_section $HV

=for apidoc Am|SV**|hv_fetchs|HV* tb|"key"|I32 lval
Like C<hv_fetch>, but takes a literal string instead of a
string/length pair.
=for apidoc_section $lexer

=for apidoc Amx|void|lex_stuff_pvs|"pv"|U32 flags

Like L</lex_stuff_pvn>, but takes a literal string instead of
a string/length pair.

=cut
*/

#define ASSERT_IS_LITERAL(s) ("" s "")

/*
=for apidoc_section $string

=for apidoc Amu|pair|STR_WITH_LEN|"literal string"

Returns two comma separated tokens of the input literal string, and its length.
This is convenience macro which helps out in some API calls.
Note that it can't be used as an 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.

=cut
*/

#define STR_WITH_LEN(s)  ASSERT_IS_LITERAL(s), (sizeof(s)-1)

/* STR_WITH_LEN() shortcuts */
#define newSVpvs(str) Perl_newSVpvn(aTHX_ STR_WITH_LEN(str))
#define newSVpvs_flags(str,flags)	\
    Perl_newSVpvn_flags(aTHX_ STR_WITH_LEN(str), flags)
#define newSVpvs_share(str) Perl_newSVpvn_share(aTHX_ STR_WITH_LEN(str), 0)
#define sv_catpvs_flags(sv, str, flags) \
    Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), flags)
#define sv_catpvs_nomg(sv, str) \
    Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), 0)
#define sv_catpvs(sv, str) \
    Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC)
#define sv_catpvs_mg(sv, str) \
    Perl_sv_catpvn_flags(aTHX_ sv, STR_WITH_LEN(str), SV_GMAGIC|SV_SMAGIC)
#define sv_setpvs(sv, str) Perl_sv_setpvn(aTHX_ sv, STR_WITH_LEN(str))
#define sv_setpvs_mg(sv, str) Perl_sv_setpvn_mg(aTHX_ sv, STR_WITH_LEN(str))
#define sv_setref_pvs(rv, classname, str) \
    Perl_sv_setref_pvn(aTHX_ rv, classname, STR_WITH_LEN(str))
#define savepvs(str) Perl_savepvn(aTHX_ STR_WITH_LEN(str))
#define savesharedpvs(str) Perl_savesharedpvn(aTHX_ STR_WITH_LEN(str))
#define gv_stashpvs(str, create) \
    Perl_gv_stashpvn(aTHX_ STR_WITH_LEN(str), create)

#define gv_fetchpvs(namebeg, flags, sv_type) \
    Perl_gv_fetchpvn_flags(aTHX_ STR_WITH_LEN(namebeg), flags, sv_type)
#define  gv_fetchpvn  gv_fetchpvn_flags
#define sv_catxmlpvs(dsv, str, utf8) \
    Perl_sv_catxmlpvn(aTHX_ dsv, STR_WITH_LEN(str), utf8)


#define lex_stuff_pvs(pv,flags) Perl_lex_stuff_pvn(aTHX_ STR_WITH_LEN(pv), flags)

#define get_cvs(str, flags)					\
        Perl_get_cvn_flags(aTHX_ STR_WITH_LEN(str), (flags))

/* internal helpers */
/* Transitional */
#ifndef PERL_VERSION_MAJOR
#  define PERL_VERSION_MAJOR  PERL_REVISION
#else
#  undef  PERL_REVISION     /* We don't want code to be using these */
#endif
#ifndef PERL_VERSION_MINOR
#  define PERL_VERSION_MINOR  PERL_VERSION
#else
#  undef  PERL_VERSION
#endif
#ifndef PERL_VERSION_PATCH
#  define PERL_VERSION_PATCH  PERL_SUBVERSION
#else
#  undef  PERL_SUBVERSION
#endif

#define PERL_JNP_TO_DECIMAL_(maJor,miNor,Patch)                             \
            /* '10*' leaves room for things like alpha, beta, releases */   \
                    (10 * ((maJor) * 1000000) + ((miNor) * 1000) + (Patch))
#define PERL_DECIMAL_VERSION_                                               \
        PERL_JNP_TO_DECIMAL_(PERL_VERSION_MAJOR, PERL_VERSION_MINOR,        \
                                                        PERL_VERSION_PATCH)

/*
=for apidoc_section $versioning
=for apidoc AmR|bool|PERL_VERSION_EQ|const U8 major|const U8 minor|const U8 patch
=for apidoc_item PERL_VERSION_GE
=for apidoc_item PERL_VERSION_GT
=for apidoc_item PERL_VERSION_LE
=for apidoc_item PERL_VERSION_LT
=for apidoc_item PERL_VERSION_NE

Returns whether or not the perl currently being compiled has the specified
relationship to the perl given by the parameters.  For example,

 #if PERL_VERSION_GT(5,24,2)
   code that will only be compiled on perls after v5.24.2
 #else
   fallback code
 #endif

Note that this is usable in making compile-time decisions

You may use the special value '*' for the final number to mean ALL possible
values for it.  Thus,

 #if PERL_VERSION_EQ(5,31,'*')

means all perls in the 5.31 series.  And

 #if PERL_VERSION_NE(5,24,'*')

means all perls EXCEPT 5.24 ones.  And

 #if PERL_VERSION_LE(5,9,'*')

is effectively

 #if PERL_VERSION_LT(5,10,0)

This means you don't have to think so much when converting from the existing
deprecated C<PERL_VERSION> to using this macro:

 #if PERL_VERSION <= 9

becomes

 #if PERL_VERSION_LE(5,9,'*')

=cut
*/

/* N.B. These don't work if the patch version is 42 or 92, as those are what
 * '*' is in ASCII and EBCDIC respectively */
# define PERL_VERSION_EQ(j,n,p)                                             \
              (((p) == '*')                                                 \
               ? (   (j) == PERL_VERSION_MAJOR                              \
                  && (n) == PERL_VERSION_MINOR)                             \
               : (PERL_DECIMAL_VERSION_ == PERL_JNP_TO_DECIMAL_(j,n,p)))
# define PERL_VERSION_NE(j,n,p) (! PERL_VERSION_EQ(j,n,p))

# define PERL_VERSION_LT(j,n,p) /* < '*' effectively means < 0 */           \
    (PERL_DECIMAL_VERSION_ < PERL_JNP_TO_DECIMAL_( (j),                     \
                                                   (n),                     \
                                                 (((p) == '*') ? 0 : p)))
# define PERL_VERSION_GE(j,n,p)  (! PERL_VERSION_LT(j,n,p))

# define PERL_VERSION_LE(j,n,p)  /* <= '*' effectively means < n+1 */       \
    (PERL_DECIMAL_VERSION_ < PERL_JNP_TO_DECIMAL_(                  (j),    \
                                          (((p) == '*') ? ((n)+1) : (n)),   \
                                          (((p) == '*') ? 0 : p)))
# define PERL_VERSION_GT(j,n,p) (! PERL_VERSION_LE(j,n,p))

/*
=for apidoc_section $string

=for apidoc Am|bool|strNE|char* s1|char* s2
Test two C<NUL>-terminated strings to see if they are different.  Returns true
or false.

=for apidoc Am|bool|strEQ|char* s1|char* s2
Test two C<NUL>-terminated strings to see if they are equal.  Returns true or
false.

=for apidoc Am|bool|strLT|char* s1|char* s2
Test two C<NUL>-terminated strings to see if the first, C<s1>, is less than the
second, C<s2>.  Returns true or false.

=for apidoc Am|bool|strLE|char* s1|char* s2
Test two C<NUL>-terminated strings to see if the first, C<s1>, is less than or
equal to the second, C<s2>.  Returns true or false.

=for apidoc Am|bool|strGT|char* s1|char* s2
Test two C<NUL>-terminated strings to see if the first, C<s1>, is greater than
the second, C<s2>.  Returns true or false.

=for apidoc Am|bool|strGE|char* s1|char* s2
Test two C<NUL>-terminated strings to see if the first, C<s1>, is greater than
or equal to the second, C<s2>.  Returns true or false.

=for apidoc Am|bool|strnNE|char* s1|char* s2|STRLEN len
Test two C<NUL>-terminated strings to see if they are different.  The C<len>
parameter indicates the number of bytes to compare.  Returns true or false.  (A
wrapper for C<strncmp>).

=for apidoc Am|bool|strnEQ|char* s1|char* s2|STRLEN len
Test two C<NUL>-terminated strings to see if they are equal.  The C<len>
parameter indicates the number of bytes to compare.  Returns true or false.  (A
wrapper for C<strncmp>).

=for apidoc Am|bool|memEQ|char* s1|char* s2|STRLEN len
Test two buffers (which may contain embedded C<NUL> characters, to see if they
are equal.  The C<len> parameter indicates the number of bytes to compare.
Returns true or false.  It is undefined behavior if either of the buffers
doesn't contain at least C<len> bytes.

=for apidoc Am|bool|memEQs|char* s1|STRLEN l1|"s2"
Like L</memEQ>, but the second string is a literal enclosed in double quotes,
C<l1> gives the number of bytes in C<s1>.
Returns true or false.

=for apidoc Am|bool|memNE|char* s1|char* s2|STRLEN len
Test two buffers (which may contain embedded C<NUL> characters, to see if they
are not equal.  The C<len> parameter indicates the number of bytes to compare.
Returns true or false.  It is undefined behavior if either of the buffers
doesn't contain at least C<len> bytes.

=for apidoc Am|bool|memNEs|char* s1|STRLEN l1|"s2"
Like L</memNE>, but the second string is a literal enclosed in double quotes,
C<l1> gives the number of bytes in C<s1>.
Returns true or false.

=for apidoc Am|bool|memCHRs|"list"|char c
Returns the position of the first occurence of the byte C<c> in the literal
string C<"list">, or NULL if C<c> doesn't appear in C<"list">.  All bytes are
treated as unsigned char.  Thus this macro can be used to determine if C<c> is
in a set of particular characters.  Unlike L<strchr(3)>, it works even if C<c>
is C<NUL> (and the set doesn't include C<NUL>).

=cut

New macros should use the following conventions for their names (which are
based on the underlying C library functions):

  (mem | str n? ) (EQ | NE | LT | GT | GE | (( BEGIN | END ) P? )) l? s?

  Each has two main parameters, string-like operands that are compared
  against each other, as specified by the macro name.  Some macros may
  additionally have one or potentially even two length parameters.  If a length
  parameter applies to both string parameters, it will be positioned third;
  otherwise any length parameter immediately follows the string parameter it
  applies to.

  If the prefix to the name is 'str', the string parameter is a pointer to a C
  language string.  Such a string does not contain embedded NUL bytes; its
  length may be unknown, but can be calculated by C<strlen()>, since it is
  terminated by a NUL, which isn't included in its length.

  The optional 'n' following 'str' means that there is a third parameter,
  giving the maximum number of bytes to look at in each string.  Even if both
  strings are longer than the length parameter, those extra bytes will be
  unexamined.

  The 's' suffix means that the 2nd byte string parameter is a literal C
  double-quoted string.  Its length will automatically be calculated by the
  macro, so no length parameter will ever be needed for it.

  If the prefix is 'mem', the string parameters don't have to be C strings;
  they may contain embedded NUL bytes, do not necessarily have a terminating
  NUL, and their lengths can be known only through other means, which in
  practice are additional parameter(s) passed to the function.  All 'mem'
  functions have at least one length parameter.  Barring any 'l' or 's' suffix,
  there is a single length parameter, in position 3, which applies to both
  string parameters.  The 's' suffix means, as described above, that the 2nd
  string is a literal double-quoted C string (hence its length is calculated by
  the macro, and the length parameter to the function applies just to the first
  string parameter, and hence is positioned just after it).  An 'l' suffix
  means that the 2nd string parameter has its own length parameter, and the
  signature will look like memFOOl(s1, l1, s2, l2).

  BEGIN (and END) are for testing if the 2nd string is an initial (or final)
  substring  of the 1st string.  'P' if present indicates that the substring
  must be a "proper" one in tha mathematical sense that the first one must be
  strictly larger than the 2nd.

*/


#define strNE(s1,s2) (strcmp(s1,s2) != 0)
#define strEQ(s1,s2) (strcmp(s1,s2) == 0)
#define strLT(s1,s2) (strcmp(s1,s2) < 0)
#define strLE(s1,s2) (strcmp(s1,s2) <= 0)
#define strGT(s1,s2) (strcmp(s1,s2) > 0)
#define strGE(s1,s2) (strcmp(s1,s2) >= 0)

#define strnNE(s1,s2,l) (strncmp(s1,s2,l) != 0)
#define strnEQ(s1,s2,l) (strncmp(s1,s2,l) == 0)

#define memEQ(s1,s2,l) (memcmp(((const void *) (s1)), ((const void *) (s2)), l) == 0)
#define memNE(s1,s2,l) (! memEQ(s1,s2,l))

/* memEQ and memNE where second comparand is a string constant */
#define memEQs(s1, l, s2) \
        (((sizeof(s2)-1) == (l)) && memEQ((s1), ASSERT_IS_LITERAL(s2), (sizeof(s2)-1)))
#define memNEs(s1, l, s2) (! memEQs(s1, l, s2))

/* Keep these private until we decide it was a good idea */
#if defined(PERL_CORE) || defined(PERL_EXT) || defined(PERL_EXT_POSIX)

#define strBEGINs(s1,s2) (strncmp(s1,ASSERT_IS_LITERAL(s2), sizeof(s2)-1) == 0)

#define memBEGINs(s1, l, s2)                                                \
            (   (Ptrdiff_t) (l) >= (Ptrdiff_t) sizeof(s2) - 1               \
             && memEQ(s1, ASSERT_IS_LITERAL(s2), sizeof(s2)-1))
#define memBEGINPs(s1, l, s2)                                               \
            (   (Ptrdiff_t) (l) > (Ptrdiff_t) sizeof(s2) - 1                \
             && memEQ(s1, ASSERT_IS_LITERAL(s2), sizeof(s2)-1))
#define memENDs(s1, l, s2)                                                  \
            (   (Ptrdiff_t) (l) >= (Ptrdiff_t) sizeof(s2) - 1               \
             && memEQ(s1 + (l) - (sizeof(s2) - 1), ASSERT_IS_LITERAL(s2), sizeof(s2)-1))
#define memENDPs(s1, l, s2)                                                 \
            (   (Ptrdiff_t) (l) > (Ptrdiff_t) sizeof(s2)                    \
             && memEQ(s1 + (l) - (sizeof(s2) - 1), ASSERT_IS_LITERAL(s2), sizeof(s2)-1))
#endif  /* End of making macros private */

#define memLT(s1,s2,l) (memcmp(s1,s2,l) < 0)
#define memLE(s1,s2,l) (memcmp(s1,s2,l) <= 0)
#define memGT(s1,s2,l) (memcmp(s1,s2,l) > 0)
#define memGE(s1,s2,l) (memcmp(s1,s2,l) >= 0)

#define memCHRs(s1,c) ((const char *) memchr(ASSERT_IS_LITERAL(s1) , c, sizeof(s1)-1))

/*
 * Character classes.
 *
 * Unfortunately, the introduction of locales means that we
 * can't trust isupper(), etc. to tell the truth.  And when
 * it comes to /\w+/ with tainting enabled, we *must* be able
 * to trust our character classes.
 *
 * Therefore, the default tests in the text of Perl will be independent of
 * locale.  Any code that wants to depend on the current locale will use the
 * macros that contain _LC in their names
 */

#ifdef USE_LOCALE_CTYPE
#  ifndef CTYPE256
#    define CTYPE256
#  endif
#endif

/*

=head1 Character classification
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<perlrecharclass/POSIX Character Classes>.)  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<use bytes>, and only the ones
with C<LC> in the name are affected by the current locale.

The base function, e.g., C<isALPHA()>, takes any signed or unsigned value,
treating it as a code point, and returns a boolean as to whether or not the
character represented by it 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<isI<FOO>_A> (e.g., C<isALPHA_A()>) is identical to the base function
with no suffix C<"_A">.  This variant is used to emphasize by its name that
only ASCII-range characters can return TRUE.

Variant C<isI<FOO>_L1> imposes the Latin-1 (or EBCDIC equivalent) 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<isWORDCHAR_L1()> will return
true when called with the code point 0xDF, which is a word character in both
ASCII and EBCDIC (though it represents different characters in each).
If the input is a number that doesn't fit in an octet, FALSE is returned.
(Perl's documentation uses a colloquial definition of Latin-1, to include all
code points below 256.)

Variant C<isI<FOO>_uvchr> is exactly like the C<isI<FOO>_L1> variant, for
inputs below 256, but if the code point is larger than 255, Unicode rules are
used to determine if it is in the character class.  For example,
C<isWORDCHAR_uvchr(0x100)> returns TRUE, since 0x100 is LATIN CAPITAL LETTER A
WITH MACRON in Unicode, and is a word character.

Variants C<isI<FOO>_utf8> and C<isI<FOO>_utf8_safe> are like C<isI<FOO>_uvchr>,
but are used for UTF-8 encoded strings.  The two forms are different names for
the same thing.  Each call to one of these classifies the first character of
the string starting at C<p>.  The second parameter, C<e>, points to anywhere in
the string beyond the first character, up to one byte past the end of the
entire string.  Although both variants are identical, the suffix C<_safe> in
one name emphasizes that it will not attempt to read beyond S<C<e - 1>>,
provided that the constraint S<C<s E<lt> e>> is true (this is asserted for in
C<-DDEBUGGING> builds).  If the UTF-8 for the input character is malformed in
some way, the program may croak, or the function may return FALSE, at the
discretion of the implementation, and subject to change in future releases.

Variant C<isI<FOO>_LC> is like the C<isI<FOO>_A> and C<isI<FOO>_L1> variants,
but the result is based on the current locale, which is what C<LC> in the name
stands for.  If Perl can determine that the current locale is a UTF-8 locale,
it uses the published Unicode rules; otherwise, it uses the C library function
that gives the named classification.  For example, C<isDIGIT_LC()> when not in
a UTF-8 locale returns the result of calling C<isdigit()>.  FALSE is always
returned if the input won't fit into an octet.  On some platforms where the C
library function is known to be defective, Perl changes its result to follow
the POSIX standard's rules.

Variant C<isI<FOO>_LC_uvchr> acts exactly like C<isI<FOO>_LC> for inputs less
than 256, but for larger ones it returns the Unicode classification of the code
point.

Variants C<isI<FOO>_LC_utf8> and C<isI<FOO>_LC_utf8_safe> are like
C<isI<FOO>_LC_uvchr>, but are used for UTF-8 encoded strings.  The two forms
are different names for the same thing.  Each call to one of these classifies
the first character of the string starting at C<p>.  The second parameter,
C<e>, points to anywhere in the string beyond the first character, up to one
byte past the end of the entire string.  Although both variants are identical,
the suffix C<_safe> in one name emphasizes that it will not attempt to read
beyond S<C<e - 1>>, provided that the constraint S<C<s E<lt> e>> is true (this
is asserted for in C<-DDEBUGGING> builds).  If the UTF-8 for the input
character is malformed in some way, the program may croak, or the function may
return FALSE, at the discretion of the implementation, and subject to change in
future releases.

=for apidoc Am|bool|isALPHA|UV ch
=for apidoc_item ||isALPHA_A|UV ch
=for apidoc_item ||isALPHA_LC|UV ch
=for apidoc_item ||isALPHA_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isALPHA_LC_uvchr|UV ch
=for apidoc_item ||isALPHA_L1|UV ch
=for apidoc_item ||isALPHA_utf8|U8 * s|U8 * end
=for apidoc_item ||isALPHA_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isALPHA_uvchr|UV ch
Returns a boolean indicating whether the specified input is one of C<[A-Za-z]>,
analogous to C<m/[[:alpha:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=cut

Here and below, we add the prototypes of these macros for downstream programs
that would be interested in them, such as Devel::PPPort

=for apidoc Am|bool|isALPHANUMERIC|UV ch
=for apidoc_item ||isALPHANUMERIC_A|UV ch
=for apidoc_item ||isALPHANUMERIC_LC|UV ch
=for apidoc_item ||isALPHANUMERIC_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isALPHANUMERIC_LC_uvchr|UV ch
=for apidoc_item ||isALPHANUMERIC_L1|UV ch
=for apidoc_item ||isALPHANUMERIC_utf8|U8 * s|U8 * end
=for apidoc_item ||isALPHANUMERIC_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isALPHANUMERIC_uvchr|UV ch
Returns a boolean indicating whether the specified character is one of
C<[A-Za-z0-9]>, analogous to C<m/[[:alnum:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isALNUMC|UV ch
=for apidoc_item  ||isALNUMC_A|UV ch
=for apidoc_item  ||isALNUMC_LC|UV ch
=for apidoc_item  ||isALNUMC_LC_uvchr|UV ch
=for apidoc_item  ||isALNUMC_L1|UV ch
These are discouraged, backward compatibility macros for L</C<isALPHANUMERIC>>.
That is, each returns a boolean indicating whether the specified character is
one of C<[A-Za-z0-9]>, analogous to C<m/[[:alnum:]]/>.

The C<C> suffix in the names was meant to indicate that they correspond to the
C language L<C<isalnum(3)>>.

=for apidoc Am|bool|isASCII|UV ch
=for apidoc_item ||isASCII_A|UV ch
=for apidoc_item ||isASCII_LC|UV ch
=for apidoc_item ||isASCII_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isASCII_LC_uvchr|UV ch
=for apidoc_item ||isASCII_L1|UV ch
=for apidoc_item ||isASCII_utf8|U8 * s|U8 * end
=for apidoc_item ||isASCII_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isASCII_uvchr|UV ch
Returns a boolean indicating whether the specified character is one of the 128
characters in the ASCII character set, analogous to C<m/[[:ascii:]]/>.
On non-ASCII platforms, it returns TRUE iff this
character corresponds to an ASCII character.  Variants C<isASCII_A()> and
C<isASCII_L1()> are identical to C<isASCII()>.
See the L<top of this section|/Character classification> for an explanation of
the variants.
Note, however, that some platforms do not have the C library routine
C<isascii()>.  In these cases, the variants whose names contain C<LC> 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<isASCII> will give the correct results when called with any
byte in any string encoded or not in UTF-8.  And similarly C<isASCII_utf8> and
C<isASCII_utf8_safe> will work properly on any string encoded or not in UTF-8.

=for apidoc Am|bool|isBLANK|UV ch
=for apidoc_item ||isBLANK_A|UV ch
=for apidoc_item ||isBLANK_LC|UV ch
=for apidoc_item ||isBLANK_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isBLANK_LC_uvchr|UV ch
=for apidoc_item ||isBLANK_L1|UV ch
=for apidoc_item ||isBLANK_utf8|U8 * s|U8 * end
=for apidoc_item ||isBLANK_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isBLANK_uvchr|UV ch
Returns a boolean indicating whether the specified character is a
character considered to be a blank, analogous to C<m/[[:blank:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants.
Note,
however, that some platforms do not have the C library routine
C<isblank()>.  In these cases, the variants whose names contain C<LC> are
the same as the corresponding ones without.

=for apidoc Am|bool|isCNTRL|UV ch
=for apidoc_item ||isCNTRL_A|UV ch
=for apidoc_item ||isCNTRL_LC|UV ch
=for apidoc_item ||isCNTRL_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isCNTRL_LC_uvchr|UV ch
=for apidoc_item ||isCNTRL_L1|UV ch
=for apidoc_item ||isCNTRL_utf8|U8 * s|U8 * end
=for apidoc_item ||isCNTRL_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isCNTRL_uvchr|UV ch

Returns a boolean indicating whether the specified character is a
control character, analogous to C<m/[[:cntrl:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants.
On EBCDIC platforms, you almost always want to use the C<isCNTRL_L1> variant.

=for apidoc Am|bool|isDIGIT|UV ch
=for apidoc_item ||isDIGIT_A|UV ch
=for apidoc_item ||isDIGIT_LC|UV ch
=for apidoc_item ||isDIGIT_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isDIGIT_LC_uvchr|UV ch
=for apidoc_item ||isDIGIT_L1|UV ch
=for apidoc_item ||isDIGIT_utf8|U8 * s|U8 * end
=for apidoc_item ||isDIGIT_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isDIGIT_uvchr|UV ch

Returns a boolean indicating whether the specified character is a
digit, analogous to C<m/[[:digit:]]/>.
Variants C<isDIGIT_A> and C<isDIGIT_L1> are identical to C<isDIGIT>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isGRAPH|UV ch
=for apidoc_item ||isGRAPH_A|UV ch
=for apidoc_item ||isGRAPH_LC|UV ch
=for apidoc_item ||isGRAPH_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isGRAPH_LC_uvchr|UV ch
=for apidoc_item ||isGRAPH_L1|UV ch
=for apidoc_item ||isGRAPH_utf8|U8 * s|U8 * end
=for apidoc_item ||isGRAPH_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isGRAPH_uvchr|UV ch
Returns a boolean indicating whether the specified character is a
graphic character, analogous to C<m/[[:graph:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isLOWER|UV ch
=for apidoc_item ||isLOWER_A|UV ch
=for apidoc_item ||isLOWER_LC|UV ch
=for apidoc_item ||isLOWER_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isLOWER_LC_uvchr|UV ch
=for apidoc_item ||isLOWER_L1|UV ch
=for apidoc_item ||isLOWER_utf8|U8 * s|U8 * end
=for apidoc_item ||isLOWER_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isLOWER_uvchr|UV ch
Returns a boolean indicating whether the specified character is a
lowercase character, analogous to C<m/[[:lower:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants

=for apidoc Am|bool|isOCTAL|UV ch
=for apidoc_item ||isOCTAL_A|UV ch
=for apidoc_item ||isOCTAL_L1|UV ch
Returns a boolean indicating whether the specified character is an
octal digit, [0-7].
The only two variants are C<isOCTAL_A> and C<isOCTAL_L1>; each is identical to
C<isOCTAL>.

=for apidoc Am|bool|isPUNCT|UV ch
=for apidoc_item ||isPUNCT_A|UV ch
=for apidoc_item ||isPUNCT_LC|UV ch
=for apidoc_item ||isPUNCT_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isPUNCT_LC_uvchr|UV ch
=for apidoc_item ||isPUNCT_L1|UV ch
=for apidoc_item ||isPUNCT_utf8|U8 * s|U8 * end
=for apidoc_item ||isPUNCT_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isPUNCT_uvchr|UV ch
Returns a boolean indicating whether the specified character is a
punctuation character, analogous to C<m/[[:punct:]]/>.
Note that the definition of what is punctuation isn't as
straightforward as one might desire.  See L<perlrecharclass/POSIX Character
Classes> for details.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isSPACE|UV ch
=for apidoc_item ||isSPACE_A|UV ch
=for apidoc_item ||isSPACE_LC|UV ch
=for apidoc_item ||isSPACE_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isSPACE_LC_uvchr|UV ch
=for apidoc_item ||isSPACE_L1|UV ch
=for apidoc_item ||isSPACE_utf8|U8 * s|U8 * end
=for apidoc_item ||isSPACE_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isSPACE_uvchr|UV ch
Returns a boolean indicating whether the specified character is a
whitespace character.  This is analogous
to what C<m/\s/> matches in a regular expression.  Starting in Perl 5.18
this also matches what C<m/[[:space:]]/> does.  Prior to 5.18, only the
locale forms of this macro (the ones with C<LC> in their names) matched
precisely what C<m/[[:space:]]/> does.  In those releases, the only difference,
in the non-locale variants, was that C<isSPACE()> did not match a vertical tab.
(See L</isPSXSPC> for a macro that matches a vertical tab in all releases.)
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isPSXSPC|UV ch
=for apidoc_item ||isPSXSPC_A|UV ch
=for apidoc_item ||isPSXSPC_LC|UV ch
=for apidoc_item ||isPSXSPC_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isPSXSPC_LC_uvchr|UV ch
=for apidoc_item ||isPSXSPC_L1|UV ch
=for apidoc_item ||isPSXSPC_utf8|U8 * s|U8 * end
=for apidoc_item ||isPSXSPC_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isPSXSPC_uvchr|UV ch
(short for Posix Space)
Starting in 5.18, this is identical in all its forms to the
corresponding C<isSPACE()> macros.
The locale forms of this macro are identical to their corresponding
C<isSPACE()> forms in all Perl releases.  In releases prior to 5.18, the
non-locale forms differ from their C<isSPACE()> forms only in that the
C<isSPACE()> forms don't match a Vertical Tab, and the C<isPSXSPC()> forms do.
Otherwise they are identical.  Thus this macro is analogous to what
C<m/[[:space:]]/> matches in a regular expression.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isUPPER|UV ch
=for apidoc_item ||isUPPER_A|UV ch
=for apidoc_item ||isUPPER_LC|UV ch
=for apidoc_item ||isUPPER_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isUPPER_LC_uvchr|UV ch
=for apidoc_item ||isUPPER_L1|UV ch
=for apidoc_item ||isUPPER_utf8|U8 * s|U8 * end
=for apidoc_item ||isUPPER_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isUPPER_uvchr|UV ch
Returns a boolean indicating whether the specified character is an
uppercase character, analogous to C<m/[[:upper:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isPRINT|UV ch
=for apidoc_item ||isPRINT_A|UV ch
=for apidoc_item ||isPRINT_LC|UV ch
=for apidoc_item ||isPRINT_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isPRINT_LC_uvchr|UV ch
=for apidoc_item ||isPRINT_L1|UV ch
=for apidoc_item ||isPRINT_utf8|U8 * s|U8 * end
=for apidoc_item ||isPRINT_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isPRINT_uvchr|UV ch
Returns a boolean indicating whether the specified character is a
printable character, analogous to C<m/[[:print:]]/>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isWORDCHAR|UV ch
=for apidoc_item ||isWORDCHAR_A|UV ch
=for apidoc_item ||isWORDCHAR_LC|UV ch
=for apidoc_item ||isWORDCHAR_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isWORDCHAR_LC_uvchr|UV ch
=for apidoc_item ||isWORDCHAR_L1|UV ch
=for apidoc_item ||isWORDCHAR_utf8|U8 * s|U8 * end
=for apidoc_item ||isWORDCHAR_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isWORDCHAR_uvchr|UV ch
Returns a boolean indicating whether the specified character is a character
that is a word character, analogous to what C<m/\w/> and C<m/[[:word:]]/> 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).

See the L<top of this section|/Character classification> for an explanation of
the variants.

C<isWORDCHAR_A>, C<isWORDCHAR_L1>, C<isWORDCHAR_uvchr>,
C<isWORDCHAR_LC>, C<isWORDCHAR_LC_uvchr>, C<isWORDCHAR_LC_utf8>, and
C<isWORDCHAR_LC_utf8_safe> are also as described there, but additionally
include the platform's native underscore.

=for apidoc Am|bool|isALNUM         |UV ch
=for apidoc_item  ||isALNUM_A       |UV ch
=for apidoc_item  ||isALNUM_LC      |UV ch
=for apidoc_item  ||isALNUM_LC_uvchr|UV ch
These are each a synonym for their respectively named L</C<isWORDCHAR>>
variant.

They are provided for backward compatibility, even though a word character
includes more than the standard C language meaning of alphanumeric.
To get the C language definition, use the corresponding L</C<isALPHANUMERIC>>
variant.

=for apidoc Am|bool|isXDIGIT|UV ch
=for apidoc_item ||isXDIGIT_A|UV ch
=for apidoc_item ||isXDIGIT_LC|UV ch
=for apidoc_item ||isXDIGIT_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isXDIGIT_LC_uvchr|UV ch
=for apidoc_item ||isXDIGIT_L1|UV ch
=for apidoc_item ||isXDIGIT_utf8|U8 * s|U8 * end
=for apidoc_item ||isXDIGIT_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isXDIGIT_uvchr|UV ch
Returns a boolean indicating whether the specified character is a hexadecimal
digit.  In the ASCII range these are C<[0-9A-Fa-f]>.  Variants C<isXDIGIT_A()>
and C<isXDIGIT_L1()> are identical to C<isXDIGIT()>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isIDFIRST|UV ch
=for apidoc_item ||isIDFIRST_A|UV ch
=for apidoc_item ||isIDFIRST_LC|UV ch
=for apidoc_item ||isIDFIRST_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isIDFIRST_LC_uvchr|UV ch
=for apidoc_item ||isIDFIRST_L1|UV ch
=for apidoc_item ||isIDFIRST_utf8|U8 * s|U8 * end
=for apidoc_item ||isIDFIRST_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isIDFIRST_uvchr|UV 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<XID_Start>.  The difference is that this
returns true only if the input character also matches L</isWORDCHAR>.
See the L<top of this section|/Character classification> for an explanation of
the variants.

=for apidoc Am|bool|isIDCONT|UV ch
=for apidoc_item ||isIDCONT_A|UV ch
=for apidoc_item ||isIDCONT_LC|UV ch
=for apidoc_item ||isIDCONT_LC_utf8_safe|U8 * s| U8 *end
=for apidoc_item ||isIDCONT_LC_uvchr|UV ch
=for apidoc_item ||isIDCONT_L1|UV ch
=for apidoc_item ||isIDCONT_utf8|U8 * s|U8 * end
=for apidoc_item ||isIDCONT_utf8_safe|U8 * s|U8 * end
=for apidoc_item ||isIDCONT_uvchr|UV 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<XID_Continue>.  The
difference is that this returns true only if the input character also matches
L</isWORDCHAR>.  See the L<top of this section|/Character classification> for
an explanation of the variants.

=for apidoc_section $numeric

=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
Perl uses "full" Unicode case mappings.  This means that converting a single
character to another case may result in a sequence of more than one character.
For example, the uppercase of C<E<223>> (LATIN SMALL LETTER SHARP S) is the two
character sequence C<SS>.  This presents some complications   The lowercase of
all characters in the range 0..255 is a single character, and thus
C<L</toLOWER_L1>> is furnished.  But, C<toUPPER_L1> can't exist, as it couldn't
return a valid result for all legal inputs.  Instead C<L</toUPPER_uvchr>> has
an API that does allow every possible legal result to be returned.)  Likewise
no other function that is crippled by not being able to give the correct
results for the full range of possible inputs has been implemented here.

=for apidoc Am|UV|toUPPER|UV cp
=for apidoc_item |UV|toUPPER_A|UV cp
=for apidoc_item |UV|toUPPER_utf8|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toUPPER_utf8_safe|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toUPPER_uvchr|UV cp|U8* s|STRLEN* lenp

These all return the uppercase of a character.  The differences are what domain
they operate on, and whether the input is specified as a code point (those
forms with a C<cp> parameter) or as a UTF-8 string (the others).  In the latter
case, the code point to use is the first one in the buffer of UTF-8 encoded
code points, delineated by the arguments S<C<p .. e - 1>>.

C<toUPPER> and C<toUPPER_A> are synonyms of each other.  They return the
uppercase of any lowercase ASCII-range code point.  All other inputs are
returned unchanged.  Since these are macros, the input type may be any integral
one, and the output will occupy the same number of bits as the input.

There is no C<toUPPER_L1> nor C<toUPPER_LATIN1> as the uppercase of some code
points in the 0..255 range is above that range or consists of multiple
characters.  Instead use C<toUPPER_uvchr>.

C<toUPPER_uvchr> returns the uppercase of any Unicode code point.  The return
value is identical to that of C<toUPPER_A> for input code points in the ASCII
range.  The uppercase of the vast majority of Unicode code points is the same
as the code point itself.  For these, and for code points above the legal
Unicode maximum, this returns the input code point unchanged.  It additionally
stores the UTF-8 of the result into the buffer beginning at C<s>, and its
length in bytes into C<*lenp>.  The caller must have made C<s> large enough to
contain at least C<UTF8_MAXBYTES_CASE+1> bytes to avoid possible overflow.

NOTE: the uppercase of a code point may be more than one code point.  The
return value of this function is only the first of these.  The entire uppercase
is returned in C<s>.  To determine if the result is more than a single code
point, you can do something like this:

 uc = toUPPER_uvchr(cp, s, &len);
 if (len > UTF8SKIP(s)) { is multiple code points }
 else { is a single code point }

C<toUPPER_utf8> and C<toUPPER_utf8_safe> are synonyms of each other.  The only
difference between these and C<toUPPER_uvchr> is that the source for these is
encoded in UTF-8, instead of being a code point.  It is passed as a buffer
starting at C<p>, with C<e> pointing to one byte beyond its end.  The C<p>
buffer may certainly contain more than one code point; but only the first one
(up through S<C<e - 1>>) is examined.  If the UTF-8 for the input character is
malformed in some way, the program may croak, or the function may return the
REPLACEMENT CHARACTER, at the discretion of the implementation, and subject to
change in future releases.

=for apidoc Am|UV|toFOLD|UV cp
=for apidoc_item |UV|toFOLD_A|UV cp
=for apidoc_item |UV|toFOLD_utf8|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toFOLD_utf8_safe|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toFOLD_uvchr|UV cp|U8* s|STRLEN* lenp

These all return the foldcase of a character.  "foldcase" is an internal case
for C</i> pattern matching. If the foldcase of character A and the foldcase of
character B are the same, they match caselessly; otherwise they don't.

The differences in the forms are what domain they operate on, and whether the
input is specified as a code point (those forms with a C<cp> parameter) or as a
UTF-8 string (the others).  In the latter case, the code point to use is the
first one in the buffer of UTF-8 encoded code points, delineated by the
arguments S<C<p .. e - 1>>.

C<toFOLD> and C<toFOLD_A> are synonyms of each other.  They return the
foldcase of any ASCII-range code point.  In this range, the foldcase is
identical to the lowercase.  All other inputs are returned unchanged.  Since
these are macros, the input type may be any integral one, and the output will
occupy the same number of bits as the input.

There is no C<toFOLD_L1> nor C<toFOLD_LATIN1> as the foldcase of some code
points in the 0..255 range is above that range or consists of multiple
characters.  Instead use C<toFOLD_uvchr>.

C<toFOLD_uvchr> returns the foldcase of any Unicode code point.  The return
value is identical to that of C<toFOLD_A> for input code points in the ASCII
range.  The foldcase of the vast majority of Unicode code points is the same
as the code point itself.  For these, and for code points above the legal
Unicode maximum, this returns the input code point unchanged.  It additionally
stores the UTF-8 of the result into the buffer beginning at C<s>, and its
length in bytes into C<*lenp>.  The caller must have made C<s> large enough to
contain at least C<UTF8_MAXBYTES_CASE+1> bytes to avoid possible overflow.

NOTE: the foldcase of a code point may be more than one code point.  The
return value of this function is only the first of these.  The entire foldcase
is returned in C<s>.  To determine if the result is more than a single code
point, you can do something like this:

 uc = toFOLD_uvchr(cp, s, &len);
 if (len > UTF8SKIP(s)) { is multiple code points }
 else { is a single code point }

C<toFOLD_utf8> and C<toFOLD_utf8_safe> are synonyms of each other.  The only
difference between these and C<toFOLD_uvchr> is that the source for these is
encoded in UTF-8, instead of being a code point.  It is passed as a buffer
starting at C<p>, with C<e> pointing to one byte beyond its end.  The C<p>
buffer may certainly contain more than one code point; but only the first one
(up through S<C<e - 1>>) is examined.  If the UTF-8 for the input character is
malformed in some way, the program may croak, or the function may return the
REPLACEMENT CHARACTER, at the discretion of the implementation, and subject to
change in future releases.

=for apidoc Am|UV|toLOWER|UV cp
=for apidoc_item |UV|toLOWER_A|UV cp
=for apidoc_item |UV|toLOWER_LATIN1|UV cp
=for apidoc_item |UV|toLOWER_LC|UV cp
=for apidoc_item |UV|toLOWER_L1|UV cp
=for apidoc_item |UV|toLOWER_utf8|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toLOWER_utf8_safe|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toLOWER_uvchr|UV cp|U8* s|STRLEN* lenp

These all return the lowercase of a character.  The differences are what domain
they operate on, and whether the input is specified as a code point (those
forms with a C<cp> parameter) or as a UTF-8 string (the others).  In the latter
case, the code point to use is the first one in the buffer of UTF-8 encoded
code points, delineated by the arguments S<C<p .. e - 1>>.

C<toLOWER> and C<toLOWER_A> are synonyms of each other.  They return the
lowercase of any uppercase ASCII-range code point.  All other inputs are
returned unchanged.  Since these are macros, the input type may be any integral
one, and the output will occupy the same number of bits as the input.

C<toLOWER_L1> and C<toLOWER_LATIN1> are synonyms of each other.  They behave
identically as C<toLOWER> for ASCII-range input.  But additionally will return
the lowercase of any uppercase code point in the entire 0..255 range, assuming
a Latin-1 encoding (or the EBCDIC equivalent on such platforms).

C<toLOWER_LC> returns the lowercase of the input code point according to the
rules of the current POSIX locale.  Input code points outside the range 0..255
are returned unchanged.

C<toLOWER_uvchr> returns the lowercase of any Unicode code point.  The return
value is identical to that of C<toLOWER_L1> for input code points in the 0..255
range.  The lowercase of the vast majority of Unicode code points is the same
as the code point itself.  For these, and for code points above the legal
Unicode maximum, this returns the input code point unchanged.  It additionally
stores the UTF-8 of the result into the buffer beginning at C<s>, and its
length in bytes into C<*lenp>.  The caller must have made C<s> large enough to
contain at least C<UTF8_MAXBYTES_CASE+1> bytes to avoid possible overflow.

NOTE: the lowercase of a code point may be more than one code point.  The
return value of this function is only the first of these.  The entire lowercase
is returned in C<s>.  To determine if the result is more than a single code
point, you can do something like this:

 uc = toLOWER_uvchr(cp, s, &len);
 if (len > UTF8SKIP(s)) { is multiple code points }
 else { is a single code point }

C<toLOWER_utf8> and C<toLOWER_utf8_safe> are synonyms of each other.  The only
difference between these and C<toLOWER_uvchr> is that the source for these is
encoded in UTF-8, instead of being a code point.  It is passed as a buffer
starting at C<p>, with C<e> pointing to one byte beyond its end.  The C<p>
buffer may certainly contain more than one code point; but only the first one
(up through S<C<e - 1>>) is examined.  If the UTF-8 for the input character is
malformed in some way, the program may croak, or the function may return the
REPLACEMENT CHARACTER, at the discretion of the implementation, and subject to
change in future releases.

=for apidoc Am|UV|toTITLE|UV cp
=for apidoc_item |UV|toTITLE_A|UV cp
=for apidoc_item |UV|toTITLE_utf8|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toTITLE_utf8_safe|U8* p|U8* e|U8* s|STRLEN* lenp
=for apidoc_item |UV|toTITLE_uvchr|UV cp|U8* s|STRLEN* lenp

These all return the titlecase of a character.  The differences are what domain
they operate on, and whether the input is specified as a code point (those
forms with a C<cp> parameter) or as a UTF-8 string (the others).  In the latter
case, the code point to use is the first one in the buffer of UTF-8 encoded
code points, delineated by the arguments S<C<p .. e - 1>>.

C<toTITLE> and C<toTITLE_A> are synonyms of each other.  They return the
titlecase of any lowercase ASCII-range code point.  In this range, the
titlecase is identical to the uppercase.  All other inputs are returned
unchanged.  Since these are macros, the input type may be any integral one, and
the output will occupy the same number of bits as the input.

There is no C<toTITLE_L1> nor C<toTITLE_LATIN1> as the titlecase of some code
points in the 0..255 range is above that range or consists of multiple
characters.  Instead use C<toTITLE_uvchr>.

C<toTITLE_uvchr> returns the titlecase of any Unicode code point.  The return
value is identical to that of C<toTITLE_A> for input code points in the ASCII
range.  The titlecase of the vast majority of Unicode code points is the same
as the code point itself.  For these, and for code points above the legal
Unicode maximum, this returns the input code point unchanged.  It additionally
stores the UTF-8 of the result into the buffer beginning at C<s>, and its
length in bytes into C<*lenp>.  The caller must have made C<s> large enough to
contain at least C<UTF8_MAXBYTES_CASE+1> bytes to avoid possible overflow.

NOTE: the titlecase of a code point may be more than one code point.  The
return value of this function is only the first of these.  The entire titlecase
is returned in C<s>.  To determine if the result is more than a single code
point, you can do something like this:

 uc = toTITLE_uvchr(cp, s, &len);
 if (len > UTF8SKIP(s)) { is multiple code points }
 else { is a single code point }

C<toTITLE_utf8> and C<toTITLE_utf8_safe> are synonyms of each other.  The only
difference between these and C<toTITLE_uvchr> is that the source for these is
encoded in UTF-8, instead of being a code point.  It is passed as a buffer
starting at C<p>, with C<e> pointing to one byte beyond its end.  The C<p>
buffer may certainly contain more than one code point; but only the first one
(up through S<C<e - 1>>) is examined.  If the UTF-8 for the input character is
malformed in some way, the program may croak, or the function may return the
REPLACEMENT CHARACTER, at the discretion of the implementation, and subject to
change in future releases.

=cut

XXX Still undocumented isVERTWS_uvchr and _utf8; it's unclear what their names
really should be.  Also toUPPER_LC and toFOLD_LC, which are subject to change,
and aren't general purpose as they don't work on U+DF, and assert against that.
and isCASED_LC, as it really is more of an internal thing.

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

*/

/*
   void below because that's the best fit, and works for Devel::PPPort
=for apidoc_section $integer
=for apidoc AyT||WIDEST_UTYPE

Yields the widest unsigned integer type on the platform, currently either
C<U32> or C<U64>.  This can be used in declarations such as

 WIDEST_UTYPE my_uv;

or casts

 my_uv = (WIDEST_UTYPE) val;

=cut

*/
#define WIDEST_UTYPE PERL_UINTMAX_T

/* Where there could be some confusion, use this as a static assert in macros
 * to make sure that a parameter isn't a pointer.  But some compilers can't
 * handle this.  The only one known so far that doesn't is gcc 3.3.6; the check
 * below isn't thorough for such an old compiler, so may have to be revised if
 * experience so dictates. */
#if  ! PERL_IS_GCC || PERL_GCC_VERSION_GT(3,3,6)
#  define ASSERT_NOT_PTR(x) ((x) | 0)
#else
#  define ASSERT_NOT_PTR(x) (x)
#endif

/* Likewise, this is effectively a static assert to be used to guarantee the
 * parameter is a pointer
 *
 * NOT suitable for void* 
 */
#define ASSERT_IS_PTR(x) (__ASSERT_(sizeof(*(x))) (x))

/* FITS_IN_8_BITS(c) returns true if c doesn't have  a bit set other than in
 * the lower 8.  It is designed to be hopefully bomb-proof, making sure that no
 * bits of information are lost even on a 64-bit machine, but to get 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.
 *
 * For Coverity, be always true, because otherwise Coverity thinks
 * it finds several expressions that are always true, independent
 * of operands.  Well, they are, but that is kind of the point.
 */
#ifndef __COVERITY__
  /* The '| 0' part in ASSERT_NOT_PTR ensures a compiler error if c is not
   * integer (like e.g., a pointer) */
#  define FITS_IN_8_BITS(c) (   (sizeof(c) == 1)                            \
                             || (((WIDEST_UTYPE) ASSERT_NOT_PTR(c)) >> 8) == 0)
#else
#  define FITS_IN_8_BITS(c) (1)
#endif

/* Returns true if l <= c <= (l + n), where 'l' and 'n' are non-negative
 * Written this way so that after optimization, only one conditional test is
 * needed.  (The NV casts stop any warnings about comparison always being true
 * if called with an unsigned.  The cast preserves the sign, which is all we
 * care about.) */
#define withinCOUNT(c, l, n)  (__ASSERT_((NV) (l) >= 0)                 \
                               __ASSERT_((NV) (n) >= 0)                 \
                               withinCOUNT_KNOWN_VALID_((c), (l), (n)))

/* For internal use only, this can be used in places where it is known that the
 * parameters to withinCOUNT() are valid, to avoid the asserts.  For example,
 * inRANGE() below, calls this several times, but does all the necessary
 * asserts itself, once.  The reason that this is necessary is that the
 * duplicate asserts were exceeding the internal limits of some compilers */
#define withinCOUNT_KNOWN_VALID_(c, l, n)                                   \
    ((((WIDEST_UTYPE) (c)) - ASSERT_NOT_PTR(l))                             \
                                   <= ((WIDEST_UTYPE) ASSERT_NOT_PTR(n)))

/* Returns true if c is in the range l..u, where 'l' is non-negative
 * Written this way so that after optimization, only one conditional test is
 * needed. */
#define inRANGE(c, l, u) (__ASSERT_((NV) (l) >= 0) __ASSERT_((u) >= (l))    \
   (  (sizeof(c) == sizeof(U8))  ? inRANGE_helper_(U8, (c), (l), ((u)))     \
    : (sizeof(c) == sizeof(U16)) ? inRANGE_helper_(U16,(c), (l), ((u)))     \
    : (sizeof(c) == sizeof(U32)) ? inRANGE_helper_(U32,(c), (l), ((u)))     \
             : (__ASSERT_(sizeof(c) == sizeof(WIDEST_UTYPE))                \
                          inRANGE_helper_(WIDEST_UTYPE,(c), (l), ((u))))))

/* For internal use, this is used by machine-generated code which generates
 * known valid calls, with a known sizeof().  This avoids the extra code and
 * asserts that were exceeding internal limits of some compilers. */
#define inRANGE_helper_(cast, c, l, u)                                      \
                    withinCOUNT_KNOWN_VALID_(((cast) (c)), (l), ((u) - (l)))

#ifdef EBCDIC
#   ifndef _ALL_SOURCE
        /* The native libc isascii() et.al. functions return the wrong results
         * on at least z/OS unless this is defined. */
#       error   _ALL_SOURCE should probably be defined
#   endif
#else
    /* There is a simple definition of ASCII for ASCII platforms.  But the
     * EBCDIC one isn't so simple, so is defined using table look-up like the
     * other macros below.
     *
     * The cast here is used instead of '(c) >= 0', because some compilers emit
     * a warning that that test is always true when the parameter is an
     * unsigned type.  khw supposes that it could be written as
     *      && ((c) == '\0' || (c) > 0)
     * to avoid the message, but the cast will likely avoid extra branches even
     * with stupid compilers. */
#   define isASCII(c)    (((WIDEST_UTYPE) ASSERT_NOT_PTR(c)) < 128)
#endif

/* Take the eight possible bit patterns of the lower 3 bits and you get the
 * lower 3 bits of the 8 octal digits, in both ASCII and EBCDIC, so those bits
 * can be ignored.  If the rest match '0', we have an octal */
#define isOCTAL_A(c)  ((((WIDEST_UTYPE) ASSERT_NOT_PTR(c)) & ~7) == '0')

#ifdef H_PERL       /* If have access to perl.h, lookup in its table */

/* 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.
 *
 * 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).  Except \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_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:] or [:upper:] under /i */
#  define CC_SPACE_             10      /* \s, [:space:] */
#  define CC_BLANK_             11      /* [:blank:] */
#  define CC_XDIGIT_            12      /* [:xdigit:] */
#  define CC_CNTRL_             13      /* [:cntrl:] */
#  define CC_ASCII_             14      /* [:ascii:] */
#  define CC_VERTSPACE_         15      /* \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. */
#  define CC_IDFIRST_                  16
#  define CC_CHARNAME_CONT_            17
#  define CC_NONLATIN1_FOLD_           18
#  define CC_NONLATIN1_SIMPLE_FOLD_    19
#  define CC_QUOTEMETA_                20
#  define CC_NON_FINAL_FOLD_           21
#  define CC_IS_IN_SOME_FOLD_          22
#  define CC_BINDIGIT_                 23
#  define CC_OCTDIGIT_                 24
#  define CC_MNEMONIC_CNTRL_           25

/* Unused: 26-31
 * 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.
 * Similarly the EBCDIC-only bits are used just for speed, and could be
 * replaced by other means */

#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_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_CC_COUNT    (HIGHEST_REGCOMP_DOT_H_SYNC_ + 1)

START_EXTERN_C
#  ifdef DOINIT
EXTCONST  U32 PL_charclass[] = {
#    include "l1_char_class_tab.h"
};

#  else /* ! DOINIT */
EXTCONST U32 PL_charclass[];
#  endif
END_EXTERN_C

    /* The 1U keeps Solaris from griping when shifting sets the uppermost bit */
#   define CC_mask_(classnum) (1U << (classnum))

    /* For internal core Perl use only: the base macro for defining macros like
     * isALPHA */
#   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_))

    /* For internal core Perl use only: the base macro for defining macros like
     * isALPHA_A.  The foo_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)))

/* On ASCII platforms certain classes form a single range.  It's faster to
 * special case these.  isDIGIT is a single range on all platforms */
#   ifdef EBCDIC
#     define isALPHA_A(c)  generic_isCC_A_(c, CC_ALPHA_)
#     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 isUPPER_A(c)  generic_isCC_A_(c, CC_UPPER_)
#   else
      /* By folding the upper and lowercase, we can use a single range */
#     define isALPHA_A(c)  inRANGE((~('A' ^ 'a') & (c)), 'A', 'Z')
#     define isGRAPH_A(c)  inRANGE(c, ' ' + 1, 0x7e)
#     define isLOWER_A(c)  inRANGE(c, 'a', 'z')
#     define isPRINT_A(c)  inRANGE(c, ' ', 0x7e)
#     define isUPPER_A(c)  inRANGE(c, 'A', 'Z')
#   endif
#   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)  inRANGE(c, '0', '9')
#   define isPUNCT_A(c)  generic_isCC_A_(c, CC_PUNCT_)
#   define isSPACE_A(c)  generic_isCC_A_(c, CC_SPACE_)
#   define isWORDCHAR_A(c) generic_isCC_A_(c, CC_WORDCHAR_)
#   define isXDIGIT_A(c)  generic_isCC_(c, CC_XDIGIT_) /* No non-ASCII xdigits
                                                        */
#   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)  isSPACE_L1(c)
#   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_)

#   ifdef EBCDIC
#       define isASCII(c) generic_isCC_(c, CC_ASCII_)
#   endif

    /* Participates in a single-character fold with a character above 255 */
#   if defined(PERL_IN_REGCOMP_C) || defined(PERL_IN_REGEXEC_C)
#     define HAS_NONLATIN1_SIMPLE_FOLD_CLOSURE(c)                           \
        ((   ! cBOOL(FITS_IN_8_BITS(c)))                                    \
          || (PL_charclass[(U8) (c)] & CC_mask_(CC_NONLATIN1_SIMPLE_FOLD_)))

#   define IS_NON_FINAL_FOLD(c)   generic_isCC_(c, CC_NON_FINAL_FOLD_)
#   define IS_IN_SOME_FOLD_L1(c)  generic_isCC_(c, CC_IS_IN_SOME_FOLD_)
#  endif

    /* Like the above, but also can be part of a multi-char fold */
#   define HAS_NONLATIN1_FOLD_CLOSURE(c)                                    \
      (   (! cBOOL(FITS_IN_8_BITS(c)))                                      \
       || (PL_charclass[(U8) (c)] & CC_mask_(CC_NONLATIN1_FOLD_)))

#   define _isQUOTEMETA(c) generic_isCC_(c, CC_QUOTEMETA_)

/* is c a control character for which we have a mnemonic? */
#  if defined(PERL_CORE) || defined(PERL_EXT)
#     define isMNEMONIC_CNTRL(c) generic_isCC_(c, CC_MNEMONIC_CNTRL_)
#  endif
#else   /* else we don't have perl.h H_PERL */

    /* 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. Most are coded based on first principles.  These are written to
     * avoid EBCDIC vs. ASCII #ifdef's as much as possible. */
#   define isDIGIT_A(c)  inRANGE(c, '0', '9')
#   define isBLANK_A(c)  ((c) == ' ' || (c) == '\t')
#   define isSPACE_A(c)  (isBLANK_A(c)                                   \
                          || (c) == '\n'                                 \
                          || (c) == '\r'                                 \
                          || (c) == '\v'                                 \
                          || (c) == '\f')
    /* On EBCDIC, there are gaps between 'i' and 'j'; 'r' and 's'.  Same for
     * uppercase.  The tests for those aren't necessary on ASCII, but hurt only
     * performance (if optimization isn't on), and allow the same code to be
     * used for both platform types */
#   define isLOWER_A(c)  inRANGE((c), 'a', 'i')                         \
                      || inRANGE((c), 'j', 'r')                         \
                      || inRANGE((c), 's', 'z')
#   define isUPPER_A(c)  inRANGE((c), 'A', 'I')                         \
                      || inRANGE((c), 'J', 'R')                         \
                      || inRANGE((c), 'S', 'Z')
#   define isALPHA_A(c)  (isUPPER_A(c) || isLOWER_A(c))
#   define isALPHANUMERIC_A(c) (isALPHA_A(c) || isDIGIT_A(c))
#   define isWORDCHAR_A(c)   (isALPHANUMERIC_A(c) || (c) == '_')
#   define isIDFIRST_A(c)    (isALPHA_A(c) || (c) == '_')
#   define isXDIGIT_A(c) (   isDIGIT_A(c)                               \
                          || inRANGE((c), 'a', 'f')                     \
                          || inRANGE((c), 'A', 'F')
#   define isPUNCT_A(c)  ((c) == '-' || (c) == '!' || (c) == '"'        \
                       || (c) == '#' || (c) == '$' || (c) == '%'        \
                       || (c) == '&' || (c) == '\'' || (c) == '('       \
                       || (c) == ')' || (c) == '*' || (c) == '+'        \
                       || (c) == ',' || (c) == '.' || (c) == '/'        \
                       || (c) == ':' || (c) == ';' || (c) == '<'        \
                       || (c) == '=' || (c) == '>' || (c) == '?'        \
                       || (c) == '@' || (c) == '[' || (c) == '\\'       \
                       || (c) == ']' || (c) == '^' || (c) == '_'        \
                       || (c) == '`' || (c) == '{' || (c) == '|'        \
                       || (c) == '}' || (c) == '~')
#   define isGRAPH_A(c)  (isALPHANUMERIC_A(c) || isPUNCT_A(c))
#   define isPRINT_A(c)  (isGRAPH_A(c) || (c) == ' ')

#   ifdef EBCDIC
        /* The below is accurate for the 3 EBCDIC code pages traditionally
         * supported by perl.  The only difference between them in the controls
         * is the position of \n, and that is represented symbolically below */
#       define isCNTRL_A(c)  ((c) == '\0' || (c) == '\a' || (c) == '\b'     \
                          ||  (c) == '\f' || (c) == '\n' || (c) == '\r'     \
                          ||  (c) == '\t' || (c) == '\v'                    \
                          || inRANGE((c), 1, 3)     /* SOH, STX, ETX */     \
                          ||  (c) == 7F   /* U+7F DEL */                    \
                          || inRANGE((c), 0x0E, 0x13) /* SO SI DLE          \
                                                         DC[1-3] */         \
                          ||  (c) == 0x18 /* U+18 CAN */                    \
                          ||  (c) == 0x19 /* U+19 EOM */                    \
                          || inRANGE((c), 0x1C, 0x1F) /* [FGRU]S */         \
                          ||  (c) == 0x26 /* U+17 ETB */                    \
                          ||  (c) == 0x27 /* U+1B ESC */                    \
                          ||  (c) == 0x2D /* U+05 ENQ */                    \
                          ||  (c) == 0x2E /* U+06 ACK */                    \
                          ||  (c) == 0x32 /* U+16 SYN */                    \
                          ||  (c) == 0x37 /* U+04 EOT */                    \
                          ||  (c) == 0x3C /* U+14 DC4 */                    \
                          ||  (c) == 0x3D /* U+15 NAK */                    \
                          ||  (c) == 0x3F)/* U+1A SUB */
#       define isASCII(c)    (isCNTRL_A(c) || isPRINT_A(c))
#   else /* isASCII is already defined for ASCII platforms, so can use that to
            define isCNTRL */
#       define isCNTRL_A(c)  (isASCII(c) && ! isPRINT_A(c))
#   endif

    /* The _L1 macros may be unnecessary for the utilities; I (khw) added them
     * during debugging, and it seems best to keep them.  We may be called
     * without NATIVE_TO_LATIN1 being defined.  On ASCII platforms, it doesn't
     * do anything anyway, so make it not a problem */
#   if ! defined(EBCDIC) && ! defined(NATIVE_TO_LATIN1)
#       define NATIVE_TO_LATIN1(ch) (ch)
#   endif
#   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 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)                          \
                                  && (   IN_RANGE(NATIVE_TO_LATIN1((U8) c),  \
                                                  0xC0, 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_SIMPLE_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_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 H_PERL */

#define isALPHANUMERIC(c)  isALPHANUMERIC_A(c)
#define isALPHA(c)   isALPHA_A(c)
#define isASCII_A(c)  isASCII(c)
#define isASCII_L1(c)  isASCII(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_A(c) isSPACE_A(c)
#define isPSXSPC(c)  isPSXSPC_A(c)
#define isPSXSPC_L1(c) isSPACE_L1(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).

   These actually are UTF-8 invariant casing, not just ASCII, as any non-ASCII
   UTF-8 invariants are neither upper nor lower.  (Only on EBCDIC platforms are
   there non-ASCII invariants, and all of them are controls.) */
#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.  Actually this works for all UTF-8 invariants, not just the
 * ASCII range. (EBCDIC platforms can have non-ASCII invariants.) */
#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 */

/* Modified uc.  Is correct uc except for three non-ascii chars which are
 * all mapped to one of them, and these need special handling; returns the
 * input itself if is out-of-range */
#define toUPPER_LATIN1_MOD(c) ((! FITS_IN_8_BITS(c))                       \
                               ? (c)                                       \
                               : PL_mod_latin1_uc[ (U8) (c) ])
#define IN_UTF8_CTYPE_LOCALE PL_in_utf8_CTYPE_locale

/* Use foo_LC_uvchr() instead  of these for beyond the Latin1 range */

/* For internal core Perl use only: the base macro for defining macros like
 * isALPHA_LC, which uses the current LC_CTYPE locale.  'c' is the code point
 * (0-255) to check.  In a UTF-8 locale, the result is the same as calling
 * isFOO_L1(); 'classnum' is something like CC_UPPER_, which gives the class
 * number for doing this.  For non-UTF-8 locales, the code to actually do the
 * test this is passed in 'non_utf8'.  If 'c' is above 255, 0 is returned.  For
 * accessing the full range of possible code points under locale rules, use the
 * macros based on generic_LC_uvchr_ instead of this. */
#define generic_LC_base_(c, classnum, non_utf8_func)                        \
       (! FITS_IN_8_BITS(c)                                                 \
       ? 0                                                                  \
       : IN_UTF8_CTYPE_LOCALE                                               \
         ? cBOOL(PL_charclass[(U8) (c)] & CC_mask_(classnum))               \
         : cBOOL(non_utf8_func(c)))

/* A helper macro for defining macros like isALPHA_LC.  On systems without
 * proper locales, these reduce to, e.g., isALPHA_A */
#ifdef CTYPE256
#  define generic_LC_(c, classnum, non_utf8_func)   \
     generic_LC_base_(c, classnum, non_utf8_func)
#else
# define generic_LC_(c, classnum, non_utf8_func)    \
     generic_isCC_A_(c, classnum)
#endif

/* Below are the definitions for the locale-sensitive character classification
 * macros whose input domain is a byte, and the locale isn't UTF-8.  These are
 * as close as possible to the bare versions on the platform and still yield
 * POSIX Standard-compliant results.
 *
 * There is currently only one place these definitions should be used, in
 * certain function calls like Perl_iswordchar_() in inline.h.
 *
 * Most likely you want to use the macros a ways below with names like
 * isALPHA_LC().  Rarely, you may want isU8_ALPHA_LC(), somewhat below.
 *
 * The first two aren't in C89, so the fallback is to use the non-locale
 * sensitive versions; these are the same for all platforms */
#if defined(HAS_ISASCII)
#   define is_base_ASCII(c) isascii((U8) (c))
#else
#   define is_base_ASCII(c) isASCII(c)
#endif

#if defined(HAS_ISBLANK)
#   define is_base_BLANK(c) isblank((U8) (c))
#else
#   define is_base_BLANK(c) isBLANK(c)
#endif

/* The next few are the same in all platforms. */
#define is_base_CNTRL(c)     iscntrl((U8) (c))
#define is_base_IDFIRST(c)  (UNLIKELY((c) == '_') || is_base_ALPHA(c))
#define is_base_SPACE(c)     isspace((U8) (c))
#define is_base_WORDCHAR(c) (UNLIKELY((c) == '_') || is_base_ALPHANUMERIC(c))

/* The base-level case changing macros are also the same in all platforms */
#define to_base_LOWER(c)     tolower((U8) (c))
#define to_base_UPPER(c)     toupper((U8) (c))
#define to_base_FOLD(c)      to_base_LOWER(c)

#ifdef WIN32

/* The Windows functions don't bother to follow the POSIX standard, which for
 * example says that something can't both be a printable and a control.  But
 * Windows treats \t as both a control and a printable, and does such things as
 * making superscripts into both digits and punctuation.  These #defines tame
 * these flaws by assuming that the definitions of controls (and the other few
 * ones defined above) are correct, and then making sure that other definitions
 * don't have weirdnesses, by adding a check that \w and its subsets aren't
 * ispunct(), and things that are \W, like ispunct(), arent't controls.  Not
 * all possible weirdnesses are checked for, just ones that were detected on
 * actual Microsoft code pages */
#  define is_base_ALPHA(c)                                          \
                          (isalpha((U8) (c)) && ! is_base_PUNCT(c))
#  define is_base_ALPHANUMERIC(c)                                   \
                          (isalnum((U8) (c)) && ! is_base_PUNCT(c))
#  define is_base_CASED(c)                                          \
   ((isupper((U8) (c)) || islower((U8) (c))) && ! is_base_PUNCT(c))
#  define is_base_DIGIT(c)                                          \
                          (isdigit((U8) (c)) && ! is_base_PUNCT(c))
#  define is_base_GRAPH(c)                                          \
                          (isgraph((U8) (c)) && ! is_base_CNTRL(c))
#  define is_base_LOWER(c)                                          \
                          (islower((U8) (c)) && ! is_base_PUNCT(c))
#  define is_base_PRINT(c)                                          \
                          (isprint((U8) (c)) && ! is_base_CNTRL(c))
#  define is_base_PUNCT(c)                                          \
                          (ispunct((U8) (c)) && ! is_base_CNTRL(c))
#  define is_base_UPPER(c)                                          \
                          (isupper((U8) (c)) && ! is_base_PUNCT(c))
#  define is_base_XDIGIT(c)                                         \
                         (isxdigit((U8) (c)) && ! is_base_PUNCT(c))
#else

/* For all other platforms, as far as we know, isdigit(), etc. work sanely
 * enough */
#  define is_base_ALPHA(c)         isalpha((U8) (c))
#  define is_base_ALPHANUMERIC(c)  isalnum((U8) (c))
#  define is_base_CASED(c)        (islower((U8) (c)) || isupper((U8) (c)))
#  define is_base_DIGIT(c)         isdigit((U8) (c))

     /* ... But it seems that IBM products treat NBSP as both a space and a
      * graphic; these are the two platforms that we have active test beds for.
      */
#  if defined(OS390) || defined(_AIX)
#    define is_base_GRAPH(c)      (isgraph((U8) (c)) && ! isspace((U8) (c)))
#  else
#    define is_base_GRAPH(c)       isgraph((U8) (c))
#  endif
#  define is_base_LOWER(c)         islower((U8) (c))
#  define is_base_PRINT(c)         isprint((U8) (c))
#  define is_base_PUNCT(c)         ispunct((U8) (c))
#  define is_base_UPPER(c)         isupper((U8) (c))
#  define is_base_XDIGIT(c)        isxdigit((U8) (c))
#endif

/* Below is the next level up, which currently expands to nothing more
 * than the previous layer.  These are the macros to use if you really need
 * something whose input domain is a byte, and the locale isn't UTF-8; that is,
 * where you normally would have to use things like bare isalnum().
 *
 * But most likely you should instead use the layer defined further below which
 * has names like isALPHA_LC.  They deal with larger-than-byte inputs, and
 * UTF-8 locales.
 *
 * (Note, proper general operation of the bare libc functons requires you to
 * cast to U8.  These do that for you automatically.) */

#  define WRAP_U8_LC_(c, classnum, base)  base(c)

#define isU8_ALPHANUMERIC_LC(c)                                                \
              WRAP_U8_LC_((c), CC_ALPHANUMERIC_, is_base_ALPHANUMERIC)
#define isU8_ALPHA_LC(c)    WRAP_U8_LC_((c), CC_ALPHA_, is_base_ALPHA)
#define isU8_ASCII_LC(c)    WRAP_U8_LC_((c), CC_ASCII_, is_base_ASCII)
#define isU8_BLANK_LC(c)    WRAP_U8_LC_((c), CC_BLANK_, is_base_BLANK)
#define isU8_CASED_LC(c)    WRAP_U8_LC_((c), CC_CASED_, is_base_CASED)
#define isU8_CNTRL_LC(c)    WRAP_U8_LC_((c), CC_CNTRL_, is_base_CNTRL)
#define isU8_DIGIT_LC(c)    WRAP_U8_LC_((c), CC_DIGIT_, is_base_DIGIT)
#define isU8_GRAPH_LC(c)    WRAP_U8_LC_((c), CC_GRAPH_, is_base_GRAPH)
#define isU8_IDFIRST_LC(c)  WRAP_U8_LC_((c), CC_IDFIRST_, is_base_IDFIRST)
#define isU8_LOWER_LC(c)    WRAP_U8_LC_((c), CC_LOWER_, is_base_LOWER)
#define isU8_PRINT_LC(c)    WRAP_U8_LC_((c), CC_PRINT_, is_base_PRINT)
#define isU8_PUNCT_LC(c)    WRAP_U8_LC_((c), CC_PUNCT_, is_base_PUNCT)
#define isU8_SPACE_LC(c)    WRAP_U8_LC_((c), CC_SPACE_, is_base_SPACE)
#define isU8_UPPER_LC(c)    WRAP_U8_LC_((c), CC_UPPER_, is_base_UPPER)
#define isU8_WORDCHAR_LC(c) WRAP_U8_LC_((c), CC_WORDCHAR_, is_base_WORDCHAR)
#define isU8_XDIGIT_LC(c)   WRAP_U8_LC_((c), CC_XDIGIT_, is_base_XDIGIT)

#define toU8_LOWER_LC(c)    WRAP_U8_LC_((c), CC_TOLOWER_, to_base_LOWER)
#define toU8_UPPER_LC(c)    WRAP_U8_LC_((c), CC_TOUPPER_, to_base_UPPER)
#define toU8_FOLD_LC(c)     toU8_LOWER_LC(c)

/* The definitions below use the ones above to create versions in which the
 * input domain isn't restricted to bytes (though always returning false if the
 * input doesn't fit in a byte), and to behave properly should the locale be
 * UTF-8.  These are the documented ones, suitable for general use (though
 * toUPPER_LC and toFOLD_LC aren't documented because they need special
 * handling to deal with SHARP S expanding to two characters). */

#define isASCII_LC(c)               (FITS_IN_8_BITS(c) && isU8_ASCII_LC(c))
#define isALPHA_LC(c)               generic_LC_(c, CC_ALPHA_, isU8_ALPHA_LC)
#define isALPHANUMERIC_LC(c)                                                \
                      generic_LC_(c, CC_ALPHANUMERIC_, isU8_ALPHANUMERIC_LC)
#define isBLANK_LC(c)               generic_LC_(c, CC_BLANK_, isU8_BLANK_LC)
#define isCASED_LC(c)               generic_LC_(c, CC_CASED_, isU8_CASED_LC)
#define isCNTRL_LC(c)               generic_LC_(c, CC_CNTRL_, isU8_CNTRL_LC)
#define isDIGIT_LC(c)               generic_LC_(c, CC_DIGIT_, isU8_DIGIT_LC)
#define isGRAPH_LC(c)               generic_LC_(c, CC_GRAPH_, isU8_GRAPH_LC)
#define isIDFIRST_LC(c)         generic_LC_(c, CC_IDFIRST_, isU8_IDFIRST_LC)
#define isLOWER_LC(c)               generic_LC_(c, CC_LOWER_, isU8_LOWER_LC)
#define isPRINT_LC(c)               generic_LC_(c, CC_PRINT_, isU8_PRINT_LC)
#define isPUNCT_LC(c)               generic_LC_(c, CC_PUNCT_, isU8_PUNCT_LC)
#define isSPACE_LC(c)               generic_LC_(c, CC_SPACE_, isU8_SPACE_LC)
#define isUPPER_LC(c)               generic_LC_(c, CC_UPPER_, isU8_UPPER_LC)
#define isWORDCHAR_LC(c)      generic_LC_(c, CC_WORDCHAR_, isU8_WORDCHAR_LC)
#define isXDIGIT_LC(c)            generic_LC_(c, CC_XDIGIT_, isU8_XDIGIT_LC)

#ifndef CTYPE256
#  define toLOWER_LC(c)             toLOWER_A(c)
#  define toUPPER_LC(c)             toUPPER_A(c)
#  define toFOLD_LC(c)              toFOLD_A(c)
#else

/* In the next three macros, the reason for using the PL_latin arrays is in
 * case the system function is defective; it ensures uniform results that
 * conform to the Unicode standard. */

/* This does not handle the anomalies in UTF-8 Turkic locales. */
#  define toLOWER_LC(c)  ((! FITS_IN_8_BITS(c))                             \
                          ? (c)                                             \
                          : ((IN_UTF8_CTYPE_LOCALE)                         \
                             ? PL_latin1_lc[ (U8) (c) ]                     \
                             : ((U8) toU8_LOWER_LC(c))))

/* In this macro, note that the result can be larger than a byte in a UTF-8
 * locale.  It returns a single value, so can't adequately return the upper
 * case of LATIN SMALL LETTER SHARP S in a UTF-8 locale (which should be a
 * string of two values "SS");  instead it asserts against that under
 * DEBUGGING, and otherwise returns its input.  It does not handle the
 * anomalies in UTF-8 Turkic locales. */
#  define toUPPER_LC(c)                                                     \
    ((! FITS_IN_8_BITS(c))                                                  \
     ? (c)                                                                  \
     : ((! IN_UTF8_CTYPE_LOCALE)                                            \
        ? ((U8) toU8_UPPER_LC(c))                                           \
        : (UNLIKELY(((U8)(c)) == MICRO_SIGN)                                \
           ? GREEK_CAPITAL_LETTER_MU                                        \
           : ((UNLIKELY(((U8) (c)) == LATIN_SMALL_LETTER_Y_WITH_DIAERESIS)  \
              ? LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS                       \
              : (UNLIKELY(((U8)(c)) == LATIN_SMALL_LETTER_SHARP_S)          \
                ? (__ASSERT_(0) (c)) /* Fail on Sharp S in DEBUGGING */     \
                : PL_mod_latin1_uc[ (U8) (c) ]))))))

/* In this macro, note that the result can be larger than a byte in a UTF-8
 * locale.  It returns a single value, so can't adequately return the fold case
 * of LATIN SMALL LETTER SHARP S in a UTF-8 locale (which should be a string of
 * two values "ss"); instead it asserts against that under DEBUGGING, and
 * otherwise returns its input.  It does not handle the anomalies in UTF-8
 * Turkic locales */
#  define toFOLD_LC(c)                                                      \
                ((UNLIKELY((c) == MICRO_SIGN) && IN_UTF8_CTYPE_LOCALE)      \
                 ? GREEK_SMALL_LETTER_MU                                    \
                 : (__ASSERT_(   ! IN_UTF8_CTYPE_LOCALE                     \
                              || LIKELY((c) != LATIN_SMALL_LETTER_SHARP_S)) \
                    toLOWER_LC(c)))
#endif

#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: the base macros for defining macros like
 * isALPHA_uvchr.  'c' is the code point to check.  'classnum' is the POSIX class
 * number defined earlier in this file.  generic_uvchr_() is used for POSIX
 * classes where there is a macro or function 'above_latin1' that takes the
 * single argument 'c' and returns the desired value.  These exist for those
 * classes which have simple definitions, avoiding the overhead of an inversion
 * list binary search.  generic_invlist_uvchr_() can be used
 * for classes where that overhead is faster than a direct lookup.
 * generic_uvchr_() won't compile if 'c' isn't unsigned, as it won't match the
 * 'above_latin1' prototype. generic_isCC_() macro does 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. */

/* Note that all ignore 'use bytes' */
#define generic_uvchr_(classnum, above_latin1, c) ((c) < 256                \
                                             ? generic_isCC_(c, classnum)   \
                                             : above_latin1(c))
#define generic_invlist_uvchr_(classnum, c) ((c) < 256                      \
                                             ? generic_isCC_(c, classnum)   \
                                             : _is_uni_FOO(classnum, c))
#define isALPHA_uvchr(c)      generic_invlist_uvchr_(CC_ALPHA_, c)
#define isALPHANUMERIC_uvchr(c) generic_invlist_uvchr_(CC_ALPHANUMERIC_, c)
#define isASCII_uvchr(c)      isASCII(c)
#define isBLANK_uvchr(c)      generic_uvchr_(CC_BLANK_, is_HORIZWS_cp_high, c)
#define isCNTRL_uvchr(c)      isCNTRL_L1(c) /* All controls are in Latin1 */
#define isDIGIT_uvchr(c)      generic_invlist_uvchr_(CC_DIGIT_, c)
#define isGRAPH_uvchr(c)      generic_invlist_uvchr_(CC_GRAPH_, c)
#define isIDCONT_uvchr(c)                                                   \
                    generic_uvchr_(CC_WORDCHAR_, _is_uni_perl_idcont, c)
#define isIDFIRST_uvchr(c)                                                  \
                    generic_uvchr_(CC_IDFIRST_, _is_uni_perl_idstart, c)
#define isLOWER_uvchr(c)      generic_invlist_uvchr_(CC_LOWER_, c)
#define isPRINT_uvchr(c)      generic_invlist_uvchr_(CC_PRINT_, c)

#define isPUNCT_uvchr(c)      generic_invlist_uvchr_(CC_PUNCT_, c)
#define isSPACE_uvchr(c)      generic_uvchr_(CC_SPACE_, is_XPERLSPACE_cp_high, c)
#define isPSXSPC_uvchr(c)     isSPACE_uvchr(c)

#define isUPPER_uvchr(c)      generic_invlist_uvchr_(CC_UPPER_, c)
#define isVERTWS_uvchr(c)     generic_uvchr_(CC_VERTSPACE_, is_VERTWS_cp_high, c)
#define isWORDCHAR_uvchr(c)   generic_invlist_uvchr_(CC_WORDCHAR_, c)
#define isXDIGIT_uvchr(c)     generic_uvchr_(CC_XDIGIT_, is_XDIGIT_cp_high, c)

#define toFOLD_uvchr(c,s,l)	to_uni_fold(c,s,l)
#define toLOWER_uvchr(c,s,l)	to_uni_lower(c,s,l)
#define toTITLE_uvchr(c,s,l)	to_uni_title(c,s,l)
#define toUPPER_uvchr(c,s,l)	to_uni_upper(c,s,l)

/* For backwards compatibility, even though '_uni' should mean official Unicode
 * code points, in Perl it means native for those below 256 */
#define isALPHA_uni(c)          isALPHA_uvchr(c)
#define isALPHANUMERIC_uni(c)   isALPHANUMERIC_uvchr(c)
#define isASCII_uni(c)          isASCII_uvchr(c)
#define isBLANK_uni(c)          isBLANK_uvchr(c)
#define isCNTRL_uni(c)          isCNTRL_uvchr(c)
#define isDIGIT_uni(c)          isDIGIT_uvchr(c)
#define isGRAPH_uni(c)          isGRAPH_uvchr(c)
#define isIDCONT_uni(c)         isIDCONT_uvchr(c)
#define isIDFIRST_uni(c)        isIDFIRST_uvchr(c)
#define isLOWER_uni(c)          isLOWER_uvchr(c)
#define isPRINT_uni(c)          isPRINT_uvchr(c)
#define isPUNCT_uni(c)          isPUNCT_uvchr(c)
#define isSPACE_uni(c)          isSPACE_uvchr(c)
#define isPSXSPC_uni(c)         isPSXSPC_uvchr(c)
#define isUPPER_uni(c)          isUPPER_uvchr(c)
#define isVERTWS_uni(c)         isVERTWS_uvchr(c)
#define isWORDCHAR_uni(c)       isWORDCHAR_uvchr(c)
#define isXDIGIT_uni(c)         isXDIGIT_uvchr(c)
#define toFOLD_uni(c,s,l)       toFOLD_uvchr(c,s,l)
#define toLOWER_uni(c,s,l)      toLOWER_uvchr(c,s,l)
#define toTITLE_uni(c,s,l)      toTITLE_uvchr(c,s,l)
#define toUPPER_uni(c,s,l)      toUPPER_uvchr(c,s,l)

/* For internal core Perl use only: the base macros for defining macros like
 * isALPHA_LC_uvchr.  These are like isALPHA_LC, but the input can be any code
 * point, not just 0-255.  Like generic_uvchr_, there are two versions, one for
 * simple class definitions; the other for more complex.  These are like
 * generic_uvchr_, so see it for more info. */
#define generic_LC_uvchr_(latin1, above_latin1, c)                            \
                                    (c < 256 ? latin1(c) : above_latin1(c))
#define generic_LC_invlist_uvchr_(latin1, classnum, c)                        \
                            (c < 256 ? latin1(c) : _is_uni_FOO(classnum, c))

#define isALPHA_LC_uvchr(c)  generic_LC_invlist_uvchr_(isALPHA_LC, CC_ALPHA_, c)
#define isALPHANUMERIC_LC_uvchr(c)  generic_LC_invlist_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_invlist_uvchr_(isDIGIT_LC, CC_DIGIT_, c)
#define isGRAPH_LC_uvchr(c)  generic_LC_invlist_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_invlist_uvchr_(isLOWER_LC, CC_LOWER_, c)
#define isPRINT_LC_uvchr(c)  generic_LC_invlist_uvchr_(isPRINT_LC, CC_PRINT_, c)
#define isPSXSPC_LC_uvchr(c)  isSPACE_LC_uvchr(c)
#define isPUNCT_LC_uvchr(c)  generic_LC_invlist_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_invlist_uvchr_(isUPPER_LC, CC_UPPER_, c)
#define isWORDCHAR_LC_uvchr(c) generic_LC_invlist_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))

/* The "_safe" macros make sure that we don't attempt to read beyond 'e', but
 * they don't otherwise go out of their way to look for malformed UTF-8.  If
 * they can return accurate results without knowing if the input is otherwise
 * malformed, they do so.  For example isASCII is accurate in spite of any
 * non-length malformations because it looks only at a single byte. Likewise
 * isDIGIT looks just at the first byte for code points 0-255, as all UTF-8
 * variant ones return FALSE.  But, if the input has to be well-formed in order
 * for the results to be accurate, the macros will test and if malformed will
 * call a routine to die
 *
 * Except for toke.c, the macros do assume that e > p, asserting that on
 * DEBUGGING builds.  Much code that calls these depends on this being true,
 * for other reasons.  toke.c is treated specially as using the regular
 * assertion breaks it in many ways.  All strings that these operate on there
 * are supposed to have an extra NUL character at the end,  so that *e = \0. A
 * bunch of code in toke.c assumes that this is true, so the assertion allows
 * for that */
#ifdef PERL_IN_TOKE_C
#  define _utf8_safe_assert(p,e) ((e) > (p) || ((e) == (p) && *(p) == '\0'))
#else
#  define _utf8_safe_assert(p,e) ((e) > (p))
#endif

#define generic_utf8_safe_(classnum, p, e, above_latin1)                    \
    ((! _utf8_safe_assert(p, e))                                            \
      ? (_force_out_malformed_utf8_message((U8 *) (p), (U8 *) (e), 0, 1), 0)\
      : (UTF8_IS_INVARIANT(*(p)))                                           \
          ? generic_isCC_(*(p), classnum)                                   \
          : (UTF8_IS_DOWNGRADEABLE_START(*(p))                              \
             ? ((LIKELY((e) - (p) > 1 && UTF8_IS_CONTINUATION(*((p)+1))))   \
                ? generic_isCC_(EIGHT_BIT_UTF8_TO_NATIVE(*(p), *((p)+1 )),  \
                                classnum)                                   \
                : (_force_out_malformed_utf8_message(                       \
                                        (U8 *) (p), (U8 *) (e), 0, 1), 0))  \
             : above_latin1))
/* Like the above, but calls 'above_latin1(p)' to get the utf8 value.
 * 'above_latin1' can be a macro */
#define generic_func_utf8_safe_(classnum, above_latin1, p, e)               \
                    generic_utf8_safe_(classnum, p, e, above_latin1(p, e))
#define generic_non_invlist_utf8_safe_(classnum, above_latin1, p, e)        \
          generic_utf8_safe_(classnum, p, e,                                \
                             (UNLIKELY((e) - (p) < UTF8SKIP(p))             \
                              ? (_force_out_malformed_utf8_message(         \
                                      (U8 *) (p), (U8 *) (e), 0, 1), 0)     \
                              : above_latin1(p)))
/* Like the above, but passes classnum to _isFOO_utf8(), instead of having an
 * 'above_latin1' parameter */
#define generic_invlist_utf8_safe_(classnum, p, e)                          \
            generic_utf8_safe_(classnum, p, e, _is_utf8_FOO(classnum, p, e))

/* Like the above, but should be used only when it is known that there are no
 * characters in the upper-Latin1 range (128-255 on ASCII platforms) which the
 * class is TRUE for.  Hence it can skip the tests for this range.
 * 'above_latin1' should include its arguments */
#define generic_utf8_safe_no_upper_latin1_(classnum, p, e, above_latin1)    \
         (__ASSERT_(_utf8_safe_assert(p, e))                                \
         (isASCII(*(p)))                                                    \
          ? generic_isCC_(*(p), classnum)                                   \
          : (UTF8_IS_DOWNGRADEABLE_START(*(p)))                             \
             ? 0 /* Note that doesn't check validity for latin1 */          \
             : above_latin1)


#define isALPHA_utf8(p, e)         isALPHA_utf8_safe(p, e)
#define isALPHANUMERIC_utf8(p, e)  isALPHANUMERIC_utf8_safe(p, e)
#define isASCII_utf8(p, e)         isASCII_utf8_safe(p, e)
#define isBLANK_utf8(p, e)         isBLANK_utf8_safe(p, e)
#define isCNTRL_utf8(p, e)         isCNTRL_utf8_safe(p, e)
#define isDIGIT_utf8(p, e)         isDIGIT_utf8_safe(p, e)
#define isGRAPH_utf8(p, e)         isGRAPH_utf8_safe(p, e)
#define isIDCONT_utf8(p, e)        isIDCONT_utf8_safe(p, e)
#define isIDFIRST_utf8(p, e)       isIDFIRST_utf8_safe(p, e)
#define isLOWER_utf8(p, e)         isLOWER_utf8_safe(p, e)
#define isPRINT_utf8(p, e)         isPRINT_utf8_safe(p, e)
#define isPSXSPC_utf8(p, e)        isPSXSPC_utf8_safe(p, e)
#define isPUNCT_utf8(p, e)         isPUNCT_utf8_safe(p, e)
#define isSPACE_utf8(p, e)         isSPACE_utf8_safe(p, e)
#define isUPPER_utf8(p, e)         isUPPER_utf8_safe(p, e)
#define isVERTWS_utf8(p, e)        isVERTWS_utf8_safe(p, e)
#define isWORDCHAR_utf8(p, e)      isWORDCHAR_utf8_safe(p, e)
#define isXDIGIT_utf8(p, e)        isXDIGIT_utf8_safe(p, e)

#define isALPHA_utf8_safe(p, e)  generic_invlist_utf8_safe_(CC_ALPHA_, p, e)
#define isALPHANUMERIC_utf8_safe(p, e)                                      \
                        generic_invlist_utf8_safe_(CC_ALPHANUMERIC_, p, e)
#define isASCII_utf8_safe(p, e)                                             \
    /* Because ASCII is invariant under utf8, the non-utf8 macro            \
    * works */                                                              \
    (__ASSERT_(_utf8_safe_assert(p, e)) isASCII(*(p)))
#define isBLANK_utf8_safe(p, e)                                             \
        generic_non_invlist_utf8_safe_(CC_BLANK_, is_HORIZWS_high, p, e)

#ifdef EBCDIC
    /* Because all controls are UTF-8 invariants in EBCDIC, we can use this
     * more efficient macro instead of the more general one */
#   define isCNTRL_utf8_safe(p, e)                                          \
                    (__ASSERT_(_utf8_safe_assert(p, e)) isCNTRL_L1(*(p)))
#else
#   define isCNTRL_utf8_safe(p, e)  generic_utf8_safe_(CC_CNTRL_, p, e, 0)
#endif

#define isDIGIT_utf8_safe(p, e)                                             \
            generic_utf8_safe_no_upper_latin1_(CC_DIGIT_, p, e,             \
                                            _is_utf8_FOO(CC_DIGIT_, p, e))
#define isGRAPH_utf8_safe(p, e)    generic_invlist_utf8_safe_(CC_GRAPH_, p, e)
#define isIDCONT_utf8_safe(p, e)   generic_func_utf8_safe_(CC_WORDCHAR_,    \
                                                 _is_utf8_perl_idcont, p, e)

/* To prevent S_scan_word in toke.c from hanging, we have to make sure that
 * IDFIRST is an alnum.  See
 * https://github.com/Perl/perl5/issues/10275 for more detail than you
 * 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_safe(p, e)                                           \
    generic_func_utf8_safe_(CC_IDFIRST_,                                    \
                            _is_utf8_perl_idstart, (U8 *) (p), (U8 *) (e))

#define isLOWER_utf8_safe(p, e)     generic_invlist_utf8_safe_(CC_LOWER_, p, e)
#define isPRINT_utf8_safe(p, e)     generic_invlist_utf8_safe_(CC_PRINT_, p, e)
#define isPSXSPC_utf8_safe(p, e)     isSPACE_utf8_safe(p, e)
#define isPUNCT_utf8_safe(p, e)     generic_invlist_utf8_safe_(CC_PUNCT_, p, e)
#define isSPACE_utf8_safe(p, e)                                             \
    generic_non_invlist_utf8_safe_(CC_SPACE_, is_XPERLSPACE_high, p, e)
#define isUPPER_utf8_safe(p, e)  generic_invlist_utf8_safe_(CC_UPPER_, p, e)
#define isVERTWS_utf8_safe(p, e)                                            \
        generic_non_invlist_utf8_safe_(CC_VERTSPACE_, is_VERTWS_high, p, e)
#define isWORDCHAR_utf8_safe(p, e)                                          \
                             generic_invlist_utf8_safe_(CC_WORDCHAR_, p, e)
#define isXDIGIT_utf8_safe(p, e)                                            \
                   generic_utf8_safe_no_upper_latin1_(CC_XDIGIT_, p, e,     \
                             (UNLIKELY((e) - (p) < UTF8SKIP(p))             \
                              ? (_force_out_malformed_utf8_message(         \
                                      (U8 *) (p), (U8 *) (e), 0, 1), 0)     \
                              : is_XDIGIT_high(p)))

#define toFOLD_utf8(p,e,s,l)	toFOLD_utf8_safe(p,e,s,l)
#define toLOWER_utf8(p,e,s,l)	toLOWER_utf8_safe(p,e,s,l)
#define toTITLE_utf8(p,e,s,l)	toTITLE_utf8_safe(p,e,s,l)
#define toUPPER_utf8(p,e,s,l)	toUPPER_utf8_safe(p,e,s,l)

/* For internal core use only, subject to change */
#define _toFOLD_utf8_flags(p,e,s,l,f)  _to_utf8_fold_flags (p,e,s,l,f)
#define _toLOWER_utf8_flags(p,e,s,l,f) _to_utf8_lower_flags(p,e,s,l,f)
#define _toTITLE_utf8_flags(p,e,s,l,f) _to_utf8_title_flags(p,e,s,l,f)
#define _toUPPER_utf8_flags(p,e,s,l,f) _to_utf8_upper_flags(p,e,s,l,f)

#define toFOLD_utf8_safe(p,e,s,l)   _toFOLD_utf8_flags(p,e,s,l, FOLD_FLAGS_FULL)
#define toLOWER_utf8_safe(p,e,s,l)  _toLOWER_utf8_flags(p,e,s,l, 0)
#define toTITLE_utf8_safe(p,e,s,l)  _toTITLE_utf8_flags(p,e,s,l, 0)
#define toUPPER_utf8_safe(p,e,s,l)  _toUPPER_utf8_flags(p,e,s,l, 0)

#define isALPHA_LC_utf8(p, e)         isALPHA_LC_utf8_safe(p, e)
#define isALPHANUMERIC_LC_utf8(p, e)  isALPHANUMERIC_LC_utf8_safe(p, e)
#define isASCII_LC_utf8(p, e)         isASCII_LC_utf8_safe(p, e)
#define isBLANK_LC_utf8(p, e)         isBLANK_LC_utf8_safe(p, e)
#define isCNTRL_LC_utf8(p, e)         isCNTRL_LC_utf8_safe(p, e)
#define isDIGIT_LC_utf8(p, e)         isDIGIT_LC_utf8_safe(p, e)
#define isGRAPH_LC_utf8(p, e)         isGRAPH_LC_utf8_safe(p, e)
#define isIDCONT_LC_utf8(p, e)        isIDCONT_LC_utf8_safe(p, e)
#define isIDFIRST_LC_utf8(p, e)       isIDFIRST_LC_utf8_safe(p, e)
#define isLOWER_LC_utf8(p, e)         isLOWER_LC_utf8_safe(p, e)
#define isPRINT_LC_utf8(p, e)         isPRINT_LC_utf8_safe(p, e)
#define isPSXSPC_LC_utf8(p, e)        isPSXSPC_LC_utf8_safe(p, e)
#define isPUNCT_LC_utf8(p, e)         isPUNCT_LC_utf8_safe(p, e)
#define isSPACE_LC_utf8(p, e)         isSPACE_LC_utf8_safe(p, e)
#define isUPPER_LC_utf8(p, e)         isUPPER_LC_utf8_safe(p, e)
#define isWORDCHAR_LC_utf8(p, e)      isWORDCHAR_LC_utf8_safe(p, e)
#define isXDIGIT_LC_utf8(p, e)        isXDIGIT_LC_utf8_safe(p, e)

/* For internal core Perl use only: the base macros for defining macros like
 * isALPHA_LC_utf8_safe.  These are like generic_utf8_, but if the first code
 * point in 'p' is within the 0-255 range, it uses locale rules from the
 * passed-in 'macro' parameter */
#define generic_LC_utf8_safe_(macro, p, e, above_latin1)                    \
         (__ASSERT_(_utf8_safe_assert(p, e))                                \
         (UTF8_IS_INVARIANT(*(p)))                                          \
          ? macro(*(p))                                                     \
          : (UTF8_IS_DOWNGRADEABLE_START(*(p))                              \
             ? ((LIKELY((e) - (p) > 1 && UTF8_IS_CONTINUATION(*((p)+1))))   \
                ? macro(EIGHT_BIT_UTF8_TO_NATIVE(*(p), *((p)+1)))           \
                : (_force_out_malformed_utf8_message(                       \
                                        (U8 *) (p), (U8 *) (e), 0, 1), 0))  \
              : above_latin1))

#define generic_LC_invlist_utf8_safe_(macro, classnum, p, e)                  \
            generic_LC_utf8_safe_(macro, p, e,                              \
                                            _is_utf8_FOO(classnum, p, e))

#define generic_LC_func_utf8_safe_(macro, above_latin1, p, e)               \
            generic_LC_utf8_safe_(macro, p, e, above_latin1(p, e))

#define generic_LC_non_invlist_utf8_safe_(classnum, above_latin1, p, e)       \
          generic_LC_utf8_safe_(classnum, p, e,                             \
                             (UNLIKELY((e) - (p) < UTF8SKIP(p))             \
                              ? (_force_out_malformed_utf8_message(         \
                                      (U8 *) (p), (U8 *) (e), 0, 1), 0)     \
                              : above_latin1(p)))

#define isALPHANUMERIC_LC_utf8_safe(p, e)                                   \
            generic_LC_invlist_utf8_safe_(isALPHANUMERIC_LC,                \
                                        CC_ALPHANUMERIC_, p, e)
#define isALPHA_LC_utf8_safe(p, e)                                          \
            generic_LC_invlist_utf8_safe_(isALPHA_LC, CC_ALPHA_, p, e)
#define isASCII_LC_utf8_safe(p, e)                                          \
                    (__ASSERT_(_utf8_safe_assert(p, e)) isASCII_LC(*(p)))
#define isBLANK_LC_utf8_safe(p, e)                                          \
        generic_LC_non_invlist_utf8_safe_(isBLANK_LC, is_HORIZWS_high, p, e)
#define isCNTRL_LC_utf8_safe(p, e)                                          \
            generic_LC_utf8_safe_(isCNTRL_LC, p, e, 0)
#define isDIGIT_LC_utf8_safe(p, e)                                          \
            generic_LC_invlist_utf8_safe_(isDIGIT_LC, CC_DIGIT_, p, e)
#define isGRAPH_LC_utf8_safe(p, e)                                          \
            generic_LC_invlist_utf8_safe_(isGRAPH_LC, CC_GRAPH_, p, e)
#define isIDCONT_LC_utf8_safe(p, e)                                         \
            generic_LC_func_utf8_safe_(isIDCONT_LC,                         \
                                                _is_utf8_perl_idcont, p, e)
#define isIDFIRST_LC_utf8_safe(p, e)                                        \
            generic_LC_func_utf8_safe_(isIDFIRST_LC,                        \
                                               _is_utf8_perl_idstart, p, e)
#define isLOWER_LC_utf8_safe(p, e)                                          \
            generic_LC_invlist_utf8_safe_(isLOWER_LC, CC_LOWER_, p, e)
#define isPRINT_LC_utf8_safe(p, e)                                          \
            generic_LC_invlist_utf8_safe_(isPRINT_LC, CC_PRINT_, p, e)
#define isPSXSPC_LC_utf8_safe(p, e)    isSPACE_LC_utf8_safe(p, e)
#define isPUNCT_LC_utf8_safe(p, e)                                          \
            generic_LC_invlist_utf8_safe_(isPUNCT_LC, CC_PUNCT_, p, e)
#define isSPACE_LC_utf8_safe(p, e)                                          \
    generic_LC_non_invlist_utf8_safe_(isSPACE_LC, is_XPERLSPACE_high, p, e)
#define isUPPER_LC_utf8_safe(p, e)                                          \
            generic_LC_invlist_utf8_safe_(isUPPER_LC, CC_UPPER_, p, e)
#define isWORDCHAR_LC_utf8_safe(p, e)                                       \
            generic_LC_invlist_utf8_safe_(isWORDCHAR_LC, CC_WORDCHAR_, p, e)
#define isXDIGIT_LC_utf8_safe(p, e)                                         \
        generic_LC_non_invlist_utf8_safe_(isXDIGIT_LC, is_XDIGIT_high, p, e)

/* 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 isALNUM_A(c)        isALNUM(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,e)   isWORDCHAR_utf8(p,e)
#define isALNUM_utf8_safe(p,e) isWORDCHAR_utf8_safe(p,e)
#define isALNUM_LC_utf8(p,e)isWORDCHAR_LC_utf8(p,e)
#define isALNUM_LC_utf8_safe(p,e)isWORDCHAR_LC_utf8_safe(p,e)
#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,e)  isALPHANUMERIC_utf8(p,e)
#define isALNUMC_utf8_safe(p,e)  isALPHANUMERIC_utf8_safe(p,e)
#define isALNUMC_LC_utf8_safe(p,e) isALPHANUMERIC_LC_utf8_safe(p,e)

/* On EBCDIC platforms, 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.
 * '?' is special-cased on EBCDIC to APC, which is the control there that is
 * the outlier from the block that contains the other controls, just like
 * toCTRL('?') on ASCII yields DEL, the control that is the outlier from the C0
 * block.  If it weren't special cased, it would yield a non-control.
 * The conversion works both ways, so toCTRL('D') is 4, and toCTRL(4) is D,
 * etc. */
#ifndef EBCDIC
#  define toCTRL(c)    (__ASSERT_(FITS_IN_8_BITS(c)) toUPPER(((U8)(c))) ^ 64)
#else
#  define toCTRL(c)   (__ASSERT_(FITS_IN_8_BITS(c))                     \
                      ((isPRINT_A(c))                                   \
                       ? (UNLIKELY((c) == '?')                          \
                         ? QUESTION_MARK_CTRL                           \
                         : (NATIVE_TO_LATIN1(toUPPER((U8) (c))) ^ 64))  \
                       : (UNLIKELY((c) == QUESTION_MARK_CTRL)           \
                         ? '?'                                          \
                         : (LATIN1_TO_NATIVE(((U8) (c)) ^ 64)))))
#endif

/*
=for apidoc Ay||line_t
The typedef to use to declare variables that are to hold line numbers.

=cut

  Line numbers are unsigned, 32 bits.
*/
typedef U32 line_t;
#define LINE_Tf  U32uf
#define NOLINE ((line_t) 4294967295UL)  /* = FFFFFFFF */

/* Helpful alias for version prescan */
#define is_LAX_VERSION(a,b) \
        (a != Perl_prescan_version(aTHX_ a, FALSE, b, NULL, NULL, NULL, NULL))

#define is_STRICT_VERSION(a,b) \
        (a != Perl_prescan_version(aTHX_ a, TRUE, b, NULL, NULL, NULL, NULL))

#define BADVERSION(a,b,c) \
        if (b) { \
            *b = c; \
        } \
        return a;

/* Converts a character KNOWN to represent a hexadecimal digit (0-9, A-F, or
 * a-f) to its numeric value without using any branches.  The input is
 * validated only by an assert() in DEBUGGING builds.
 *
 * It works by right shifting and isolating the bit that is 0 for the digits,
 * and 1 for at least the alphas A-F, a-f.  The bit is shifted to the ones
 * position, and then to the eights position.  Both are added together to form
 * 0 if the input is '0'-'9' and to form 9 if alpha.  This is added to the
 * final four bits of the input to form the correct value. */
#define XDIGIT_VALUE(c) (__ASSERT_(isXDIGIT(c))                             \
           ((NATIVE_TO_LATIN1(c) >> 6) & 1)  /* 1 if alpha; 0 if not */     \
         + ((NATIVE_TO_LATIN1(c) >> 3) & 8)  /* 8 if alpha; 0 if not */     \
         + ((c) & 0xF))   /* 0-9 if input valid hex digit */

/* The argument is a string pointer, which is advanced. */
#define READ_XDIGIT(s)  ((s)++, XDIGIT_VALUE(*((s) - 1)))

/* Converts a character known to represent an octal digit (0-7) to its numeric
 * value.  The input is validated only by an assert() in DEBUGGING builds.  In
 * both ASCII and EBCDIC the last 3 bits of the octal digits range from 0-7. */
#define OCTAL_VALUE(c) (__ASSERT_(isOCTAL(c)) (7 & (c)))

/* Efficiently returns a boolean as to if two native characters are equivalent
 * case-insensitively.  At least one of the characters must be one of [A-Za-z];
 * the ALPHA in the name is to remind you of that.  This is asserted() in
 * DEBUGGING builds.  Because [A-Za-z] are invariant under UTF-8, this macro
 * works (on valid input) for both non- and UTF-8-encoded bytes.
 *
 * When one of the inputs is a compile-time constant and gets folded by the
 * compiler, this reduces to an AND and a TEST.  On both EBCDIC and ASCII
 * machines, 'A' and 'a' differ by a single bit; the same with the upper and
 * lower case of all other ASCII-range alphabetics.  On ASCII platforms, they
 * are 32 apart; on EBCDIC, they are 64.  At compile time, this uses an
 * exclusive 'or' to find that bit and then inverts it to form a mask, with
 * just a single 0, in the bit position where the upper- and lowercase differ.
 * */
#define isALPHA_FOLD_EQ(c1, c2)                                         \
                      (__ASSERT_(isALPHA_A(c1) || isALPHA_A(c2))        \
                      ((c1) & ~('A' ^ 'a')) ==  ((c2) & ~('A' ^ 'a')))
#define isALPHA_FOLD_NE(c1, c2) (! isALPHA_FOLD_EQ((c1), (c2)))

/*
=for apidoc_section $memory

=for apidoc Am|void|Newx|void* ptr|int nitems|type
=for apidoc_item |void*|safemalloc|size_t size

The XSUB-writer's interface to the C C<malloc> function.

Memory obtained by this should B<ONLY> be freed with L</"Safefree">.

In 5.9.3, Newx() and friends replace the older New() API, and drops
the first parameter, I<x>, a debug aid which allowed callers to identify
themselves.  This aid has been superseded by a new build option,
PERL_MEM_LOG (see L<perlhacktips/PERL_MEM_LOG>).  The older API is still
there for use in XS modules supporting older perls.

=for apidoc Am|void|Newxc|void* ptr|int nitems|type|cast
The XSUB-writer's interface to the C C<malloc> function, with
cast.  See also C<L</Newx>>.

Memory obtained by this should B<ONLY> be freed with L</"Safefree">.

=for apidoc Am|void|Newxz|void* ptr|int nitems|type
=for apidoc_item |void*|safecalloc|size_t nitems|size_t item_size

The XSUB-writer's interface to the C C<malloc> function.  The allocated
memory is zeroed with C<memzero>.  See also C<L</Newx>>.

Memory obtained by this should B<ONLY> be freed with L</"Safefree">.

=for apidoc Am|void|Renew|void* ptr|int nitems|type
=for apidoc_item |void*|saferealloc|void *ptr|size_t size

The XSUB-writer's interface to the C C<realloc> function.

Memory obtained by this should B<ONLY> be freed with L</"Safefree">.

=for apidoc Am|void|Renewc|void* ptr|int nitems|type|cast
The XSUB-writer's interface to the C C<realloc> function, with
cast.

Memory obtained by this should B<ONLY> be freed with L</"Safefree">.

=for apidoc Am|void|Safefree|void* ptr
The XSUB-writer's interface to the C C<free> function.

This should B<ONLY> be used on memory obtained using L</"Newx"> and friends.

=for apidoc_section $string
=for apidoc    Am|void  |Move |void* src|void* dest|int nitems|type
=for apidoc_item |void *|MoveD|void* src|void* dest|int nitems|type
The XSUB-writer's interface to the C C<memmove> function.  The C<src> is the
source, C<dest> is the destination, C<nitems> is the number of items, and
C<type> is the type.  Can do overlapping moves.  See also C<L</Copy>>.

C<MoveD> is like C<Move> but returns C<dest>.  Useful
for encouraging compilers to tail-call
optimise.

=for apidoc    Am|void  |Copy |void* src|void* dest|int nitems|type
=for apidoc_item |void *|CopyD|void* src|void* dest|int nitems|type
The XSUB-writer's interface to the C C<memcpy> function.  The C<src> is the
source, C<dest> is the destination, C<nitems> is the number of items, and
C<type> is the type.  May fail on overlapping copies.  See also C<L</Move>>.

C<CopyD> is like C<Copy> but returns C<dest>.  Useful
for encouraging compilers to tail-call
optimise.

=for apidoc    Am|void  |Zero |void* dest|int nitems|type
=for apidoc_item |void *|ZeroD|void* dest|int nitems|type

The XSUB-writer's interface to the C C<memzero> function.  The C<dest> is the
destination, C<nitems> is the number of items, and C<type> is the type.

C<ZeroD> is like C<Zero> but returns C<dest>.  Useful
for encouraging compilers to tail-call
optimise.

=for apidoc_section $utility
=for apidoc Amu|void|StructCopy|type *src|type *dest|type
This is an architecture-independent macro to copy one structure to another.

=for apidoc Am|void|PoisonWith|void* dest|int nitems|type|U8 byte

Fill up memory with a byte pattern (a byte repeated over and over
again) that hopefully catches attempts to access uninitialized memory.

=for apidoc Am|void|PoisonNew|void* dest|int nitems|type

PoisonWith(0xAB) for catching access to allocated but uninitialized memory.

=for apidoc Am|void|PoisonFree|void* dest|int nitems|type

PoisonWith(0xEF) for catching access to freed memory.

=for apidoc Am|void|Poison|void* dest|int nitems|type

PoisonWith(0xEF) for catching access to freed memory.

=cut */

/* Maintained for backwards-compatibility only. Use newSV() instead. */
#ifndef PERL_CORE
#define NEWSV(x,len)	newSV(len)
#endif

#define MEM_SIZE_MAX ((MEM_SIZE)-1)

#define _PERL_STRLEN_ROUNDUP_UNCHECKED(n) (((n) - 1 + PERL_STRLEN_ROUNDUP_QUANTUM) & ~((MEM_SIZE)PERL_STRLEN_ROUNDUP_QUANTUM - 1))

#ifdef PERL_MALLOC_WRAP

/* This expression will be constant-folded at compile time.  It checks
 * whether or not the type of the count n is so small (e.g. U8 or U16, or
 * U32 on 64-bit systems) that there's no way a wrap-around could occur.
 * As well as avoiding the need for a run-time check in some cases, it's
 * designed to avoid compiler warnings like:
 *     comparison is always false due to limited range of data type
 * It's mathematically equivalent to
 *    max(n) * sizeof(t) > MEM_SIZE_MAX
 */

#  define _MEM_WRAP_NEEDS_RUNTIME_CHECK(n,t) \
    (  sizeof(MEM_SIZE) < sizeof(n) \
    || sizeof(t) > ((MEM_SIZE)1 << 8*(sizeof(MEM_SIZE) - sizeof(n))))

/* This is written in a slightly odd way to avoid various spurious
 * compiler warnings. We *want* to write the expression as
 *    _MEM_WRAP_NEEDS_RUNTIME_CHECK(n,t) && (n > C)
 * (for some compile-time constant C), but even when the LHS
 * constant-folds to false at compile-time, g++ insists on emitting
 * warnings about the RHS (e.g. "comparison is always false"), so instead
 * we write it as
 *
 *    (cond ? n : X) > C
 *
 * where X is a constant with X > C always false. Choosing a value for X
 * is tricky. If 0, some compilers will complain about 0 > C always being
 * false; if 1, Coverity complains when n happens to be the constant value
 * '1', that cond ? 1 : 1 has the same value on both branches; so use C
 * for X and hope that nothing else whines.
 */

#  define _MEM_WRAP_WILL_WRAP(n,t) \
      ((_MEM_WRAP_NEEDS_RUNTIME_CHECK(n,t) ? (MEM_SIZE)(n) : \
            MEM_SIZE_MAX/sizeof(t)) > MEM_SIZE_MAX/sizeof(t))

#  define MEM_WRAP_CHECK(n,t) \
        (void)(UNLIKELY(_MEM_WRAP_WILL_WRAP(n,t)) \
        && (croak_memory_wrap(),0))

#  define MEM_WRAP_CHECK_1(n,t,a) \
        (void)(UNLIKELY(_MEM_WRAP_WILL_WRAP(n,t)) \
        && (Perl_croak_nocontext("%s",(a)),0))

/* "a" arg must be a string literal */
#  define MEM_WRAP_CHECK_s(n,t,a) \
        (   (void) (UNLIKELY(_MEM_WRAP_WILL_WRAP(n,t))          \
         && (Perl_croak_nocontext(ASSERT_IS_LITERAL(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) ? (croak_memory_wrap(),0) : 0), _PERL_STRLEN_ROUNDUP_UNCHECKED(n))
#else

#  define MEM_WRAP_CHECK(n,t)
#  define MEM_WRAP_CHECK_1(n,t,a)
#  define MEM_WRAP_CHECK_s(n,t,a)
#  define MEM_WRAP_CHECK_(n,t)

#  define PERL_STRLEN_ROUNDUP(n) _PERL_STRLEN_ROUNDUP_UNCHECKED(n)

#endif

#ifdef PERL_MEM_LOG
/*
 * If PERL_MEM_LOG is defined, all Newx()s, Renew()s, and Safefree()s
 * go through functions, which are handy for debugging breakpoints, but
 * which more importantly get the immediate calling environment (file and
 * line number, and C function name if available) passed in.  This info can
 * then be used for logging the calls, for which one gets a sample
 * implementation unless -DPERL_MEM_LOG_NOIMPL is also defined.
 *
 * Known problems:
 * - not all memory allocs get logged, only those
 *   that go through Newx() and derivatives (while all
 *   Safefrees do get logged)
 * - __FILE__ and __LINE__ do not work everywhere
 * - __func__ or __FUNCTION__ even less so
 * - I think more goes on after the perlio frees but
 *   the thing is that STDERR gets closed (as do all
 *   the file descriptors)
 * - no deeper calling stack than the caller of the Newx()
 *   or the kind, but do I look like a C reflection/introspection
 *   utility to you?
 * - the function prototypes for the logging functions
 *   probably should maybe be somewhere else than handy.h
 * - one could consider inlining (macrofying) the logging
 *   for speed, but I am too lazy
 * - one could imagine recording the allocations in a hash,
 *   (keyed by the allocation address?), and maintain that
 *   through reallocs and frees, but how to do that without
 *   any News() happening...?
 * - lots of -Ddefines to get useful/controllable output
 * - lots of ENV reads
 */

# ifdef PERL_CORE
#  ifndef PERL_MEM_LOG_NOIMPL
enum mem_log_type {
  MLT_ALLOC,
  MLT_REALLOC,
  MLT_FREE,
  MLT_NEW_SV,
  MLT_DEL_SV
};
#  endif
# endif

#endif

#ifdef PERL_MEM_LOG
#define MEM_LOG_ALLOC(n,t,a)     Perl_mem_log_alloc(n,sizeof(t),STRINGIFY(t),a,__FILE__,__LINE__,FUNCTION__)
#define MEM_LOG_REALLOC(n,t,v,a) Perl_mem_log_realloc(n,sizeof(t),STRINGIFY(t),v,a,__FILE__,__LINE__,FUNCTION__)
#define MEM_LOG_FREE(a)          Perl_mem_log_free(a,__FILE__,__LINE__,FUNCTION__)
#endif

#ifndef MEM_LOG_ALLOC
#define MEM_LOG_ALLOC(n,t,a)     (a)
#endif
#ifndef MEM_LOG_REALLOC
#define MEM_LOG_REALLOC(n,t,v,a) (a)
#endif
#ifndef MEM_LOG_FREE
#define MEM_LOG_FREE(a)          (a)
#endif

#define Newx(v,n,t)	(v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t))))))
#define Newxc(v,n,t,c)	(v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_ALLOC(n,t,safemalloc((MEM_SIZE)((n)*sizeof(t))))))
#define Newxz(v,n,t)	(v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_ALLOC(n,t,safecalloc((n),sizeof(t)))))

#ifndef PERL_CORE
/* pre 5.9.x compatibility */
#define New(x,v,n,t)	Newx(v,n,t)
#define Newc(x,v,n,t,c)	Newxc(v,n,t,c)
#define Newz(x,v,n,t)	Newxz(v,n,t)
#endif

#define Renew(v,n,t) \
          (v = (MEM_WRAP_CHECK_(n,t) (t*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t))))))
#define Renewc(v,n,t,c) \
          (v = (MEM_WRAP_CHECK_(n,t) (c*)MEM_LOG_REALLOC(n,t,v,saferealloc((Malloc_t)(v),(MEM_SIZE)((n)*sizeof(t))))))

#ifdef PERL_POISON
#define Safefree(d) \
  ((d) ? (void)(safefree(MEM_LOG_FREE((Malloc_t)(d))), Poison(&(d), 1, Malloc_t)) : (void) 0)
#else
#define Safefree(d)	safefree(MEM_LOG_FREE((Malloc_t)(d)))
#endif

/* assert that a valid ptr has been supplied - use this instead of assert(ptr)  *
 * as it handles cases like constant string arguments without throwing warnings *
 * the cast is required, as is the inequality check, to avoid warnings          */
#define perl_assert_ptr(p) assert( ((void*)(p)) != 0 )


#define Move(s,d,n,t)	(MEM_WRAP_CHECK_(n,t) perl_assert_ptr(d), perl_assert_ptr(s), (void)memmove((char*)(d),(const char*)(s), (n) * sizeof(t)))
#define Copy(s,d,n,t)	(MEM_WRAP_CHECK_(n,t) perl_assert_ptr(d), perl_assert_ptr(s), (void)memcpy((char*)(d),(const char*)(s), (n) * sizeof(t)))
#define Zero(d,n,t)	(MEM_WRAP_CHECK_(n,t) perl_assert_ptr(d), (void)memzero((char*)(d), (n) * sizeof(t)))

/* Like above, but returns a pointer to 'd' */
#define MoveD(s,d,n,t)	(MEM_WRAP_CHECK_(n,t) perl_assert_ptr(d), perl_assert_ptr(s), memmove((char*)(d),(const char*)(s), (n) * sizeof(t)))
#define CopyD(s,d,n,t)	(MEM_WRAP_CHECK_(n,t) perl_assert_ptr(d), perl_assert_ptr(s), memcpy((char*)(d),(const char*)(s), (n) * sizeof(t)))
#define ZeroD(d,n,t)	(MEM_WRAP_CHECK_(n,t) perl_assert_ptr(d), memzero((char*)(d), (n) * sizeof(t)))

#define PoisonWith(d,n,t,b)	(MEM_WRAP_CHECK_(n,t) (void)memset((char*)(d), (U8)(b), (n) * sizeof(t)))
#define PoisonNew(d,n,t)	PoisonWith(d,n,t,0xAB)
#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

/* Shallow copy */
#define StructCopy(s,d,t) (*((t*)(d)) = *((t*)(s)))

/*
=for apidoc_section $utility

=for apidoc Am|STRLEN|C_ARRAY_LENGTH|void *a

Returns the number of elements in the input C array (so you want your
zero-based indices to be less than but not equal to).

=for apidoc Am|void *|C_ARRAY_END|void *a

Returns a pointer to one element past the final element of the input C array.

=cut

C_ARRAY_END is one past the last: half-open/half-closed range, not
last-inclusive range.
*/
#define C_ARRAY_LENGTH(a)	(sizeof(a)/sizeof((a)[0]))
#define C_ARRAY_END(a)		((a) + C_ARRAY_LENGTH(a))

#if defined(PERL_CORE) || defined(PERL_EXT_RE_BUILD)
/* strlen() of a literal string constant.  Restricting this to core, in part
 * because it can generate compiler warnings about comparing unlike signs */
#  define STRLENs(s)  (sizeof("" s "") - 1)
#endif

#ifdef NEED_VA_COPY
# ifdef va_copy
#  define Perl_va_copy(s, d) va_copy(d, s)
# elif defined(__va_copy)
#  define Perl_va_copy(s, d) __va_copy(d, s)
# else
#  define Perl_va_copy(s, d) Copy(s, d, 1, va_list)
# endif
#endif

/* convenience debug macros */
#ifdef USE_ITHREADS
#define pTHX_FORMAT  "Perl interpreter: 0x%p"
#define pTHX__FORMAT ", Perl interpreter: 0x%p"
#define pTHX_VALUE_   (void *)my_perl,
#define pTHX_VALUE    (void *)my_perl
#define pTHX__VALUE_ ,(void *)my_perl,
#define pTHX__VALUE  ,(void *)my_perl
#else
#define pTHX_FORMAT
#define pTHX__FORMAT
#define pTHX_VALUE_
#define pTHX_VALUE
#define pTHX__VALUE_
#define pTHX__VALUE
#endif /* USE_ITHREADS */

/* Perl_deprecate was not part of the public API, and did not have a deprecate()
   shortcut macro defined without -DPERL_CORE. Neither codesearch.google.com nor
   CPAN::Unpack show any users outside the core.  */
#ifdef PERL_CORE
#  define deprecate(s) Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED),    \
                                            "Use of " s " is deprecated")
#  define deprecate_disappears_in(when,message) \
              Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED),    \
                               message " is deprecated, and will disappear in Perl " when)
#  define deprecate_fatal_in(when,message) \
              Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED),    \
                               message " is deprecated, and will become fatal in Perl " when)
#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)
#  elif 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 /* Uid_t_size */

#  if Gid_t_size > IVSIZE
#    define sv_setgid(sv, gid)       sv_setnv((sv), (NV)(gid))
#    define SvGID(sv)                SvNV(sv)
#  elif 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 /* Gid_t_size */

#endif

/* These are simple Marsaglia XOR-SHIFT RNG's for 64 and 32 bits. These
 * RNG's are of reasonable quality, very fast, and have the interesting
 * property that provided 'x' is non-zero they create a cycle of 2^32-1
 * or 2^64-1 "random" like numbers, with the exception of 0. Thus they
 * are very useful when you want an integer to "dance" in a random way,
 * but you also never want it to become 0 and thus false.
 *
 * Obviously they leave x unchanged if it starts out as 0.
 *
 * We have two variants just because that can be helpful in certain
 * places. There is no advantage to either, they are equally bad as each
 * other as far RNG's go. Sufficiently random for many purposes, but
 * insufficiently random for serious use as they fail important tests in
 * the Test01 BigCrush RNG test suite by L’Ecuyer and Simard. (Note
 * that Drand48 also fails BigCrush). The main point is they produce
 * different sequences and in places where we want some randomlike
 * behavior they are cheap and easy.
 *
 * Marsaglia was one of the early researchers into RNG testing and wrote
 * the Diehard RNG test suite, which after his death become the
 * Dieharder RNG suite, and was generally supplanted by the Test01 suite
 * by L'Ecruyer and associates.
 *
 * There are dozens of shift parameters that create a pseudo random ring
 * of integers 1..2^N-1, if you need a different sequence just read the
 * paper and select a set of parameters. In fact, simply reversing the
 * shift order from L/R/L to R/L/R should result in another valid
 * example, but read the paper before you do that.
 *
 * PDF of the original paper:
 *  https://www.jstatsoft.org/article/download/v008i14/916
 * Wikipedia:
 *  https://en.wikipedia.org/wiki/Xorshift
 * Criticism:
 *  https://www.iro.umontreal.ca/~lecuyer/myftp/papers/xorshift.pdf
 * Test01:
 *  http://simul.iro.umontreal.ca/testu01/tu01.html
 * Diehard:
 *  https://en.wikipedia.org/wiki/Diehard_tests
 * Dieharder:
 *  https://webhome.phy.duke.edu/~rgb/General/rand_rate/rand_rate.abs
 *
 */

/* 32 bit version */
#define PERL_XORSHIFT32_A(x)    \
STMT_START {                    \
    (x) ^= ((x) << 13);         \
    (x) ^= ((x) >> 17);         \
    (x) ^= ((x) << 5);          \
} STMT_END

/* 64 bit version */
#define PERL_XORSHIFT64_A(x)    \
STMT_START {                    \
    (x) ^= ((x) << 13);         \
    (x) ^= ((x) >> 7);          \
    (x) ^= ((x) << 17);         \
} STMT_END

/* 32 bit version */
#define PERL_XORSHIFT32_B(x)    \
STMT_START {                    \
    (x) ^= ((x) << 5);          \
    (x) ^= ((x) >> 27);         \
    (x) ^= ((x) << 8);          \
} STMT_END

/* 64 bit version - currently this is unused,
 * it is provided here to complement the 32 bit _B
 * variant which IS used. */
#define PERL_XORSHIFT64_B(x)    \
STMT_START {                    \
    (x) ^= ((x) << 15);         \
    (x) ^= ((x) >> 49);         \
    (x) ^= ((x) << 26);         \
} STMT_END


#endif  /* PERL_HANDY_H_ */

/*
 * ex: set ts=8 sts=4 sw=4 et:
 */