[p5sagit/p5-mst-13.2.git] / hv.h

/*    hv.h
 *
 *    Copyright (C) 1991, 1992, 1993, 1996, 1997, 1998, 1999,
 *    2000, 2001, 2002, 2005, 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.
 *
 */

/* entry in hash value chain */
struct he {
    /* Keep hent_next first in this structure, because sv_free_arenas take
       advantage of this to share code between the he arenas and the SV
       body arenas  */
    HE		*hent_next;	/* next entry in chain */
    HEK		*hent_hek;	/* hash key */
    union {
	SV	*hent_val;	/* scalar value that was hashed */
	Size_t	hent_refcount;	/* references for this shared hash key */
    } he_valu;
};

/* hash key -- defined separately for use as shared pointer */
struct hek {
    U32		hek_hash;	/* hash of key */
    I32		hek_len;	/* length of hash key */
    char	hek_key[1];	/* variable-length hash key */
    /* the hash-key is \0-terminated */
    /* after the \0 there is a byte for flags, such as whether the key
       is UTF-8 */
};

struct shared_he {
    struct he shared_he_he;
    struct hek shared_he_hek;
};

/* Subject to change.
   Don't access this directly.
*/
struct xpvhv_aux {
    HEK		*xhv_name;	/* name, if a symbol table */
    AV		*xhv_backreferences; /* back references for weak references */
    HE		*xhv_eiter;	/* current entry of iterator */
    I32		xhv_riter;	/* current root of iterator */
};

/* hash structure: */
/* This structure must match the beginning of struct xpvmg in sv.h. */
struct xpvhv {
    union {
	NV	xnv_nv;		/* numeric value, if any */
	HV *	xgv_stash;
    }		xnv_u;
    STRLEN	xhv_fill;	/* how full xhv_array currently is */
    STRLEN	xhv_max;	/* subscript of last element of xhv_array */
    union {
	IV	xivu_iv;	/* integer value or pv offset */
	UV	xivu_uv;
	void *	xivu_p1;
    }		xiv_u;
    union {
	MAGIC*	xmg_magic;	/* linked list of magicalness */
	HV*	xmg_ourstash;	/* Stash for our (when SvPAD_OUR is true) */
    } xmg_u;
    HV*		xmg_stash;	/* class package */
};

#define xhv_keys xiv_u.xivu_iv

#if 0
typedef struct xpvhv xpvhv_allocated;
#else
typedef struct {
    STRLEN	xhv_fill;	/* how full xhv_array currently is */
    STRLEN	xhv_max;	/* subscript of last element of xhv_array */
    union {
	IV	xivu_iv;	/* integer value or pv offset */
	UV	xivu_uv;
	void *	xivu_p1;
    }		xiv_u;
    union {
	MAGIC*	xmg_magic;	/* linked list of magicalness */
	HV*	xmg_ourstash;	/* Stash for our (when SvPAD_OUR is true) */
    } xmg_u;
    HV*		xmg_stash;	/* class package */
} xpvhv_allocated;
#endif

/* hash a key */
/* FYI: This is the "One-at-a-Time" algorithm by Bob Jenkins
 * from requirements by Colin Plumb.
 * (http://burtleburtle.net/bob/hash/doobs.html) */
/* The use of a temporary pointer and the casting games
 * is needed to serve the dual purposes of
 * (a) the hashed data being interpreted as "unsigned char" (new since 5.8,
 *     a "char" can be either signed or unsigned, depending on the compiler)
 * (b) catering for old code that uses a "char"
 *
 * The "hash seed" feature was added in Perl 5.8.1 to perturb the results
 * to avoid "algorithmic complexity attacks".
 *
 * If USE_HASH_SEED is defined, hash randomisation is done by default
 * If USE_HASH_SEED_EXPLICIT is defined, hash randomisation is done
 * only if the environment variable PERL_HASH_SEED is set.
 * For maximal control, one can define PERL_HASH_SEED.
 * (see also perl.c:perl_parse()).
 */
#ifndef PERL_HASH_SEED
#   if defined(USE_HASH_SEED) || defined(USE_HASH_SEED_EXPLICIT)
#       define PERL_HASH_SEED	PL_hash_seed
#   else
#       define PERL_HASH_SEED	0
#   endif
#endif
#define PERL_HASH(hash,str,len) \
     STMT_START	{ \
	register const char * const s_PeRlHaSh_tmp = str; \
	register const unsigned char *s_PeRlHaSh = (const unsigned char *)s_PeRlHaSh_tmp; \
	register I32 i_PeRlHaSh = len; \
	register U32 hash_PeRlHaSh = PERL_HASH_SEED; \
	while (i_PeRlHaSh--) { \
	    hash_PeRlHaSh += *s_PeRlHaSh++; \
	    hash_PeRlHaSh += (hash_PeRlHaSh << 10); \
	    hash_PeRlHaSh ^= (hash_PeRlHaSh >> 6); \
	} \
	hash_PeRlHaSh += (hash_PeRlHaSh << 3); \
	hash_PeRlHaSh ^= (hash_PeRlHaSh >> 11); \
	(hash) = (hash_PeRlHaSh + (hash_PeRlHaSh << 15)); \
    } STMT_END

/* Only hv.c and mod_perl should be doing this.  */
#ifdef PERL_HASH_INTERNAL_ACCESS
#define PERL_HASH_INTERNAL(hash,str,len) \
     STMT_START	{ \
	register const char * const s_PeRlHaSh_tmp = str; \
	register const unsigned char *s_PeRlHaSh = (const unsigned char *)s_PeRlHaSh_tmp; \
	register I32 i_PeRlHaSh = len; \
	register U32 hash_PeRlHaSh = PL_rehash_seed; \
	while (i_PeRlHaSh--) { \
	    hash_PeRlHaSh += *s_PeRlHaSh++; \
	    hash_PeRlHaSh += (hash_PeRlHaSh << 10); \
	    hash_PeRlHaSh ^= (hash_PeRlHaSh >> 6); \
	} \
	hash_PeRlHaSh += (hash_PeRlHaSh << 3); \
	hash_PeRlHaSh ^= (hash_PeRlHaSh >> 11); \
	(hash) = (hash_PeRlHaSh + (hash_PeRlHaSh << 15)); \
    } STMT_END
#endif

/*
=head1 Hash Manipulation Functions

=for apidoc AmU||HEf_SVKEY
This flag, used in the length slot of hash entries and magic structures,
specifies the structure contains an C<SV*> pointer where a C<char*> pointer
is to be expected. (For information only--not to be used).

=head1 Handy Values

=for apidoc AmU||Nullhv
Null HV pointer.

=head1 Hash Manipulation Functions

=for apidoc Am|char*|HvNAME|HV* stash
Returns the package name of a stash, or NULL if C<stash> isn't a stash.
See C<SvSTASH>, C<CvSTASH>.

=for apidoc Am|void*|HeKEY|HE* he
Returns the actual pointer stored in the key slot of the hash entry. The
pointer may be either C<char*> or C<SV*>, depending on the value of
C<HeKLEN()>.  Can be assigned to.  The C<HePV()> or C<HeSVKEY()> macros are
usually preferable for finding the value of a key.

=for apidoc Am|STRLEN|HeKLEN|HE* he
If this is negative, and amounts to C<HEf_SVKEY>, it indicates the entry
holds an C<SV*> key.  Otherwise, holds the actual length of the key.  Can
be assigned to. The C<HePV()> macro is usually preferable for finding key
lengths.

=for apidoc Am|SV*|HeVAL|HE* he
Returns the value slot (type C<SV*>) stored in the hash entry.

=for apidoc Am|U32|HeHASH|HE* he
Returns the computed hash stored in the hash entry.

=for apidoc Am|char*|HePV|HE* he|STRLEN len
Returns the key slot of the hash entry as a C<char*> value, doing any
necessary dereferencing of possibly C<SV*> keys.  The length of the string
is placed in C<len> (this is a macro, so do I<not> use C<&len>).  If you do
not care about what the length of the key is, you may use the global
variable C<PL_na>, though this is rather less efficient than using a local
variable.  Remember though, that hash keys in perl are free to contain
embedded nulls, so using C<strlen()> or similar is not a good way to find
the length of hash keys. This is very similar to the C<SvPV()> macro
described elsewhere in this document.

=for apidoc Am|SV*|HeSVKEY|HE* he
Returns the key as an C<SV*>, or C<NULL> if the hash entry does not
contain an C<SV*> key.

=for apidoc Am|SV*|HeSVKEY_force|HE* he
Returns the key as an C<SV*>.  Will create and return a temporary mortal
C<SV*> if the hash entry contains only a C<char*> key.

=for apidoc Am|SV*|HeSVKEY_set|HE* he|SV* sv
Sets the key to a given C<SV*>, taking care to set the appropriate flags to
indicate the presence of an C<SV*> key, and returns the same
C<SV*>.

=cut
*/

/* these hash entry flags ride on hent_klen (for use only in magic/tied HVs) */
#define HEf_SVKEY	-2	/* hent_key is an SV* */


#define Nullhv Null(HV*)
#define HvARRAY(hv)	((hv)->sv_u.svu_hash)
#define HvFILL(hv)	((XPVHV*)  SvANY(hv))->xhv_fill
#define HvMAX(hv)	((XPVHV*)  SvANY(hv))->xhv_max
/* This quite intentionally does no flag checking first. That's your
   responsibility.  */
#define HvAUX(hv)	((struct xpvhv_aux*)&(HvARRAY(hv)[HvMAX(hv)+1]))
#define HvRITER(hv)	(*Perl_hv_riter_p(aTHX_ (HV*)(hv)))
#define HvEITER(hv)	(*Perl_hv_eiter_p(aTHX_ (HV*)(hv)))
#define HvRITER_set(hv,r)	Perl_hv_riter_set(aTHX_ (HV*)(hv), r)
#define HvEITER_set(hv,e)	Perl_hv_eiter_set(aTHX_ (HV*)(hv), e)
#define HvRITER_get(hv)	(SvOOK(hv) ? HvAUX(hv)->xhv_riter : -1)
#define HvEITER_get(hv)	(SvOOK(hv) ? HvAUX(hv)->xhv_eiter : 0)
#define HvNAME(hv)	HvNAME_get(hv)
/* FIXME - all of these should use a UTF8 aware API, which should also involve
   getting the length. */
/* This macro may go away without notice.  */
#define HvNAME_HEK(hv) (SvOOK(hv) ? HvAUX(hv)->xhv_name : 0)
#define HvNAME_get(hv)	((SvOOK(hv) && (HvAUX(hv)->xhv_name)) \
			 ? HEK_KEY(HvAUX(hv)->xhv_name) : 0)
#define HvNAMELEN_get(hv)	((SvOOK(hv) && (HvAUX(hv)->xhv_name)) \
				 ? HEK_LEN(HvAUX(hv)->xhv_name) : 0)

/* the number of keys (including any placeholers) */
#define XHvTOTALKEYS(xhv)	((xhv)->xhv_keys)

/*
 * HvKEYS gets the number of keys that actually exist(), and is provided
 * for backwards compatibility with old XS code. The core uses HvUSEDKEYS
 * (keys, excluding placeholdes) and HvTOTALKEYS (including placeholders)
 */
#define HvKEYS(hv)		HvUSEDKEYS(hv)
#define HvUSEDKEYS(hv)		(HvTOTALKEYS(hv) - HvPLACEHOLDERS_get(hv))
#define HvTOTALKEYS(hv)		XHvTOTALKEYS((XPVHV*)  SvANY(hv))
#define HvPLACEHOLDERS(hv)	(*Perl_hv_placeholders_p(aTHX_ (HV*)hv))
#define HvPLACEHOLDERS_get(hv)	(SvMAGIC(hv) ? Perl_hv_placeholders_get(aTHX_ (HV*)hv) : 0)
#define HvPLACEHOLDERS_set(hv,p)	Perl_hv_placeholders_set(aTHX_ (HV*)hv, p)

#define HvSHAREKEYS(hv)		(SvFLAGS(hv) & SVphv_SHAREKEYS)
#define HvSHAREKEYS_on(hv)	(SvFLAGS(hv) |= SVphv_SHAREKEYS)
#define HvSHAREKEYS_off(hv)	(SvFLAGS(hv) &= ~SVphv_SHAREKEYS)

/* This is an optimisation flag. It won't be set if all hash keys have a 0
 * flag. Currently the only flags relate to utf8.
 * Hence it won't be set if all keys are 8 bit only. It will be set if any key
 * is utf8 (including 8 bit keys that were entered as utf8, and need upgrading
 * when retrieved during iteration. It may still be set when there are no longer
 * any utf8 keys.
 * See HVhek_ENABLEHVKFLAGS for the trigger.
 */
#define HvHASKFLAGS(hv)		(SvFLAGS(hv) & SVphv_HASKFLAGS)
#define HvHASKFLAGS_on(hv)	(SvFLAGS(hv) |= SVphv_HASKFLAGS)
#define HvHASKFLAGS_off(hv)	(SvFLAGS(hv) &= ~SVphv_HASKFLAGS)

#define HvLAZYDEL(hv)		(SvFLAGS(hv) & SVphv_LAZYDEL)
#define HvLAZYDEL_on(hv)	(SvFLAGS(hv) |= SVphv_LAZYDEL)
#define HvLAZYDEL_off(hv)	(SvFLAGS(hv) &= ~SVphv_LAZYDEL)

#define HvREHASH(hv)		(SvFLAGS(hv) & SVphv_REHASH)
#define HvREHASH_on(hv)		(SvFLAGS(hv) |= SVphv_REHASH)
#define HvREHASH_off(hv)	(SvFLAGS(hv) &= ~SVphv_REHASH)

/* Maybe amagical: */
/* #define HV_AMAGICmb(hv)      (SvFLAGS(hv) & (SVpgv_badAM | SVpgv_AM)) */

#define HV_AMAGIC(hv)        (SvFLAGS(hv) &   SVpgv_AM)
#define HV_AMAGIC_on(hv)     (SvFLAGS(hv) |=  SVpgv_AM)
#define HV_AMAGIC_off(hv)    (SvFLAGS(hv) &= ~SVpgv_AM)

/*
#define HV_AMAGICbad(hv)     (SvFLAGS(hv) & SVpgv_badAM)
#define HV_badAMAGIC_on(hv)  (SvFLAGS(hv) |= SVpgv_badAM)
#define HV_badAMAGIC_off(hv) (SvFLAGS(hv) &= ~SVpgv_badAM)
*/

#define Nullhe Null(HE*)
#define HeNEXT(he)		(he)->hent_next
#define HeKEY_hek(he)		(he)->hent_hek
#define HeKEY(he)		HEK_KEY(HeKEY_hek(he))
#define HeKEY_sv(he)		(*(SV**)HeKEY(he))
#define HeKLEN(he)		HEK_LEN(HeKEY_hek(he))
#define HeKUTF8(he)  HEK_UTF8(HeKEY_hek(he))
#define HeKWASUTF8(he)  HEK_WASUTF8(HeKEY_hek(he))
#define HeKREHASH(he)  HEK_REHASH(HeKEY_hek(he))
#define HeKLEN_UTF8(he)  (HeKUTF8(he) ? -HeKLEN(he) : HeKLEN(he))
#define HeKFLAGS(he)  HEK_FLAGS(HeKEY_hek(he))
#define HeVAL(he)		(he)->he_valu.hent_val
#define HeHASH(he)		HEK_HASH(HeKEY_hek(he))
#define HePV(he,lp)		((HeKLEN(he) == HEf_SVKEY) ?		\
				 SvPV(HeKEY_sv(he),lp) :		\
				 (((lp = HeKLEN(he)) >= 0) ?		\
				  HeKEY(he) : NULL))

#define HeSVKEY(he)		((HeKEY(he) && 				\
				  HeKLEN(he) == HEf_SVKEY) ?		\
				 HeKEY_sv(he) : NULL)

#define HeSVKEY_force(he)	(HeKEY(he) ?				\
				 ((HeKLEN(he) == HEf_SVKEY) ?		\
				  HeKEY_sv(he) :			\
				  sv_2mortal(newSVpvn(HeKEY(he),	\
						     HeKLEN(he)))) :	\
				 &PL_sv_undef)
#define HeSVKEY_set(he,sv)	((HeKLEN(he) = HEf_SVKEY), (HeKEY_sv(he) = sv))

#define Nullhek Null(HEK*)
#define HEK_BASESIZE		STRUCT_OFFSET(HEK, hek_key[0])
#define HEK_HASH(hek)		(hek)->hek_hash
#define HEK_LEN(hek)		(hek)->hek_len
#define HEK_KEY(hek)		(hek)->hek_key
#define HEK_FLAGS(hek)	(*((unsigned char *)(HEK_KEY(hek))+HEK_LEN(hek)+1))

#define HVhek_UTF8	0x01 /* Key is utf8 encoded. */
#define HVhek_WASUTF8	0x02 /* Key is bytes here, but was supplied as utf8. */
#define HVhek_REHASH	0x04 /* This key is in an hv using a custom HASH . */
#define HVhek_FREEKEY	0x100 /* Internal flag to say key is malloc()ed.  */
#define HVhek_PLACEHOLD	0x200 /* Internal flag to create placeholder.
                               * (may change, but Storable is a core module) */
#define HVhek_MASK	0xFF

/* Which flags enable HvHASKFLAGS? Somewhat a hack on a hack, as
   HVhek_REHASH is only needed because the rehash flag has to be duplicated
   into all keys as hv_iternext has no access to the hash flags. At this
   point Storable's tests get upset, because sometimes hashes are "keyed"
   and sometimes not, depending on the order of data insertion, and whether
   it triggered rehashing. So currently HVhek_REHAS is exempt.
*/
   
#define HVhek_ENABLEHVKFLAGS	(HVhek_MASK - HVhek_REHASH)

#define HEK_UTF8(hek)		(HEK_FLAGS(hek) & HVhek_UTF8)
#define HEK_UTF8_on(hek)	(HEK_FLAGS(hek) |= HVhek_UTF8)
#define HEK_UTF8_off(hek)	(HEK_FLAGS(hek) &= ~HVhek_UTF8)
#define HEK_WASUTF8(hek)	(HEK_FLAGS(hek) & HVhek_WASUTF8)
#define HEK_WASUTF8_on(hek)	(HEK_FLAGS(hek) |= HVhek_WASUTF8)
#define HEK_WASUTF8_off(hek)	(HEK_FLAGS(hek) &= ~HVhek_WASUTF8)
#define HEK_REHASH(hek)		(HEK_FLAGS(hek) & HVhek_REHASH)
#define HEK_REHASH_on(hek)	(HEK_FLAGS(hek) |= HVhek_REHASH)

/* calculate HV array allocation */
#ifndef PERL_USE_LARGE_HV_ALLOC
/* Default to allocating the correct size - default to assuming that malloc()
   is not broken and is efficient at allocating blocks sized at powers-of-two.
*/   
#  define PERL_HV_ARRAY_ALLOC_BYTES(size) ((size) * sizeof(HE*))
#else
#  define MALLOC_OVERHEAD 16
#  define PERL_HV_ARRAY_ALLOC_BYTES(size) \
			(((size) < 64)					\
			 ? (size) * sizeof(HE*)				\
			 : (size) * sizeof(HE*) * 2 - MALLOC_OVERHEAD)
#endif

/* Flags for hv_iternext_flags.  */
#define HV_ITERNEXT_WANTPLACEHOLDERS	0x01	/* Don't skip placeholders.  */

#define hv_iternext(hv)	hv_iternext_flags(hv, 0)
#define hv_magic(hv, gv, how) sv_magic((SV*)(hv), (SV*)(gv), how, NULL, 0)

/* available as a function in hv.c */
#define Perl_sharepvn(sv, len, hash) HEK_KEY(share_hek(sv, len, hash))
#define sharepvn(sv, len, hash)	     Perl_sharepvn(sv, len, hash)

#define share_hek_hek(hek)						\
    (++(((struct shared_he *)(((char *)hek)				\
			      - STRUCT_OFFSET(struct shared_he,		\
					      shared_he_hek)))		\
	->shared_he_he.he_valu.hent_refcount),				\
     hek)

/*
 * Local variables:
 * c-indentation-style: bsd
 * c-basic-offset: 4
 * indent-tabs-mode: t
 * End:
 *
 * ex: set ts=8 sts=4 sw=4 noet:
 */
Commit	Line	Data
a0d0e21e	1	/* hv.h
79072805	2	*
4bb101f2	3	* Copyright (C) 1991, 1992, 1993, 1996, 1997, 1998, 1999,
1d325971	4	* 2000, 2001, 2002, 2005, by Larry Wall and others
79072805	5	*
	6	* You may distribute under the terms of either the GNU General Public
	7	* License or the Artistic License, as specified in the README file.
	8	*
79072805	9	*/
79072805	10
5cbe4eec	11	/* entry in hash value chain */
79072805	12	struct he {
6f51351d	13	/* Keep hent_next first in this structure, because sv_free_arenas take
	14	advantage of this to share code between the he arenas and the SV
	15	body arenas */
5cbe4eec	16	HE hent_next; / next entry in chain */
5cbe4eec	17	HEK hent_hek; / hash key */
de616631	18	union {
	19	SV hent_val; / scalar value that was hashed */
	20	Size_t hent_refcount; /* references for this shared hash key */
	21	} he_valu;
bbce6d69	22	};
bbce6d69	23
5cbe4eec	24	/* hash key -- defined separately for use as shared pointer */
ff68c719	25	struct hek {
5cbe4eec	26	U32 hek_hash; /* hash of key */
	27	I32 hek_len; /* length of hash key */
	28	char hek_key[1]; /* variable-length hash key */
e05949c7	29	/* the hash-key is \0-terminated */
1e54db1a	30	/* after the \0 there is a byte for flags, such as whether the key
1e54db1a	31	is UTF-8 */
79072805	32	};
79072805	33
cbae3960	34	struct shared_he {
	35	struct he shared_he_he;
	36	struct hek shared_he_hek;
	37	};
bfcb3514	38
	39	/* Subject to change.
	40	Don't access this directly.
	41	*/
	42	struct xpvhv_aux {
7423f6db	43	HEK xhv_name; / name, if a symbol table */
86f55936	44	AV xhv_backreferences; / back references for weak references */
bfcb3514	45	HE xhv_eiter; / current entry of iterator */
	46	I32 xhv_riter; /* current root of iterator */
	47	};
	48
5cbe4eec	49	/* hash structure: */
6ee623d5	50	/* This structure must match the beginning of struct xpvmg in sv.h. */
79072805	51	struct xpvhv {
58430790	52	union {
	53	NV xnv_nv; /* numeric value, if any */
	54	HV * xgv_stash;
	55	} xnv_u;
463ee0b2	56	STRLEN xhv_fill; /* how full xhv_array currently is */
463ee0b2	57	STRLEN xhv_max; /* subscript of last element of xhv_array */
e4305a63	58	union {
311a25d9	59	IV xivu_iv; /* integer value or pv offset */
	60	UV xivu_uv;
	61	void * xivu_p1;
	62	} xiv_u;
e736a858	63	union {
	64	MAGIC* xmg_magic; /* linked list of magicalness */
	65	HV* xmg_ourstash; /* Stash for our (when SvPAD_OUR is true) */
	66	} xmg_u;
79072805	67	HV* xmg_stash; /* class package */
79072805	68	};
79072805	69
311a25d9	70	#define xhv_keys xiv_u.xivu_iv
e4305a63	71
c5f87f51	72	#if 0
59813432	73	typedef struct xpvhv xpvhv_allocated;
9f1501b2	74	#else
	75	typedef struct {
	76	STRLEN xhv_fill; /* how full xhv_array currently is */
	77	STRLEN xhv_max; /* subscript of last element of xhv_array */
	78	union {
311a25d9	79	IV xivu_iv; /* integer value or pv offset */
	80	UV xivu_uv;
	81	void * xivu_p1;
	82	} xiv_u;
e736a858	83	union {
	84	MAGIC* xmg_magic; /* linked list of magicalness */
	85	HV* xmg_ourstash; /* Stash for our (when SvPAD_OUR is true) */
	86	} xmg_u;
9f1501b2	87	HV* xmg_stash; /* class package */
	88	} xpvhv_allocated;
	89	#endif
e4305a63	90
5cbe4eec	91	/* hash a key */
5afd6d42	92	/* FYI: This is the "One-at-a-Time" algorithm by Bob Jenkins
	93	* from requirements by Colin Plumb.
	94	* (http://burtleburtle.net/bob/hash/doobs.html) */
	95	/* The use of a temporary pointer and the casting games
	96	* is needed to serve the dual purposes of
	97	* (a) the hashed data being interpreted as "unsigned char" (new since 5.8,
df805c3e	98	* a "char" can be either signed or unsigned, depending on the compiler)
5afd6d42	99	* (b) catering for old code that uses a "char"
830b38bd	100	*
504f80c1	101	* The "hash seed" feature was added in Perl 5.8.1 to perturb the results
504f80c1	102	* to avoid "algorithmic complexity attacks".
830b38bd	103	*
	104	* If USE_HASH_SEED is defined, hash randomisation is done by default
	105	* If USE_HASH_SEED_EXPLICIT is defined, hash randomisation is done
	106	* only if the environment variable PERL_HASH_SEED is set.
	107	* For maximal control, one can define PERL_HASH_SEED.
90ce55a1	108	* (see also perl.c:perl_parse()).
5afd6d42	109	*/
830b38bd	110	#ifndef PERL_HASH_SEED
	111	# if defined(USE_HASH_SEED) \|\| defined(USE_HASH_SEED_EXPLICIT)
	112	# define PERL_HASH_SEED PL_hash_seed
	113	# else
	114	# define PERL_HASH_SEED 0
	115	# endif
504f80c1	116	#endif
bf6bd887	117	#define PERL_HASH(hash,str,len) \
bf6bd887	118	STMT_START { \
0bd48802	119	register const char * const s_PeRlHaSh_tmp = str; \
5afd6d42	120	register const unsigned char s_PeRlHaSh = (const unsigned char )s_PeRlHaSh_tmp; \
bf6bd887	121	register I32 i_PeRlHaSh = len; \
504f80c1	122	register U32 hash_PeRlHaSh = PERL_HASH_SEED; \
a6fe520e	123	while (i_PeRlHaSh--) { \
	124	hash_PeRlHaSh += *s_PeRlHaSh++; \
	125	hash_PeRlHaSh += (hash_PeRlHaSh << 10); \
	126	hash_PeRlHaSh ^= (hash_PeRlHaSh >> 6); \
	127	} \
	128	hash_PeRlHaSh += (hash_PeRlHaSh << 3); \
	129	hash_PeRlHaSh ^= (hash_PeRlHaSh >> 11); \
4fee5c71	130	(hash) = (hash_PeRlHaSh + (hash_PeRlHaSh << 15)); \
bf6bd887	131	} STMT_END
bf6bd887	132
3d78eb94	133	/* Only hv.c and mod_perl should be doing this. */
3d78eb94	134	#ifdef PERL_HASH_INTERNAL_ACCESS
4b5190b5	135	#define PERL_HASH_INTERNAL(hash,str,len) \
4b5190b5	136	STMT_START { \
0bd48802	137	register const char * const s_PeRlHaSh_tmp = str; \
4b5190b5	138	register const unsigned char s_PeRlHaSh = (const unsigned char )s_PeRlHaSh_tmp; \
4b5190b5	139	register I32 i_PeRlHaSh = len; \
008fb0c0	140	register U32 hash_PeRlHaSh = PL_rehash_seed; \
4b5190b5	141	while (i_PeRlHaSh--) { \
	142	hash_PeRlHaSh += *s_PeRlHaSh++; \
	143	hash_PeRlHaSh += (hash_PeRlHaSh << 10); \
	144	hash_PeRlHaSh ^= (hash_PeRlHaSh >> 6); \
	145	} \
	146	hash_PeRlHaSh += (hash_PeRlHaSh << 3); \
	147	hash_PeRlHaSh ^= (hash_PeRlHaSh >> 11); \
	148	(hash) = (hash_PeRlHaSh + (hash_PeRlHaSh << 15)); \
	149	} STMT_END
	150	#endif
	151
954c1994	152	/*
ccfc67b7	153	=head1 Hash Manipulation Functions
ccfc67b7	154
954c1994	155	=for apidoc AmU\|\|HEf_SVKEY
954c1994	156	This flag, used in the length slot of hash entries and magic structures,
d1be9408	157	specifies the structure contains an C<SV> pointer where a C<char> pointer
954c1994	158	is to be expected. (For information only--not to be used).
954c1994	159
ccfc67b7	160	=head1 Handy Values
ccfc67b7	161
954c1994	162	=for apidoc AmU\|\|Nullhv
	163	Null HV pointer.
	164
ccfc67b7	165	=head1 Hash Manipulation Functions
ccfc67b7	166
954c1994	167	=for apidoc Am\|char\|HvNAME\|HV stash
9282b5fd	168	Returns the package name of a stash, or NULL if C<stash> isn't a stash.
9282b5fd	169	See C<SvSTASH>, C<CvSTASH>.
954c1994	170
	171	=for apidoc Am\|void\|HeKEY\|HE he
	172	Returns the actual pointer stored in the key slot of the hash entry. The
	173	pointer may be either C<char> or C<SV>, depending on the value of
	174	C<HeKLEN()>. Can be assigned to. The C<HePV()> or C<HeSVKEY()> macros are
	175	usually preferable for finding the value of a key.
	176
	177	=for apidoc Am\|STRLEN\|HeKLEN\|HE* he
	178	If this is negative, and amounts to C<HEf_SVKEY>, it indicates the entry
	179	holds an C<SV*> key. Otherwise, holds the actual length of the key. Can
	180	be assigned to. The C<HePV()> macro is usually preferable for finding key
	181	lengths.
	182
	183	=for apidoc Am\|SV\|HeVAL\|HE he
	184	Returns the value slot (type C<SV*>) stored in the hash entry.
	185
	186	=for apidoc Am\|U32\|HeHASH\|HE* he
	187	Returns the computed hash stored in the hash entry.
	188
	189	=for apidoc Am\|char\|HePV\|HE he\|STRLEN len
	190	Returns the key slot of the hash entry as a C<char*> value, doing any
	191	necessary dereferencing of possibly C<SV*> keys. The length of the string
	192	is placed in C<len> (this is a macro, so do I<not> use C<&len>). If you do
	193	not care about what the length of the key is, you may use the global
	194	variable C<PL_na>, though this is rather less efficient than using a local
	195	variable. Remember though, that hash keys in perl are free to contain
	196	embedded nulls, so using C<strlen()> or similar is not a good way to find
	197	the length of hash keys. This is very similar to the C<SvPV()> macro
	198	described elsewhere in this document.
	199
	200	=for apidoc Am\|SV\|HeSVKEY\|HE he
a0714e2c	201	Returns the key as an C<SV*>, or C<NULL> if the hash entry does not
954c1994	202	contain an C<SV*> key.
	203
	204	=for apidoc Am\|SV\|HeSVKEY_force\|HE he
	205	Returns the key as an C<SV*>. Will create and return a temporary mortal
	206	C<SV> if the hash entry contains only a C<char> key.
	207
	208	=for apidoc Am\|SV\|HeSVKEY_set\|HE he\|SV* sv
	209	Sets the key to a given C<SV*>, taking care to set the appropriate flags to
	210	indicate the presence of an C<SV*> key, and returns the same
	211	C<SV*>.
	212
	213	=cut
	214	*/
bf6bd887	215
bf5b86ae	216	/* these hash entry flags ride on hent_klen (for use only in magic/tied HVs) */
d1be9408	217	#define HEf_SVKEY -2 /* hent_key is an SV* */
bf6bd887	218
bf6bd887	219
79072805	220	#define Nullhv Null(HV*)
43e6e717	221	#define HvARRAY(hv) ((hv)->sv_u.svu_hash)
79072805	222	#define HvFILL(hv) ((XPVHV*) SvANY(hv))->xhv_fill
463ee0b2	223	#define HvMAX(hv) ((XPVHV*) SvANY(hv))->xhv_max
b79f7545	224	/* This quite intentionally does no flag checking first. That's your
	225	responsibility. */
	226	#define HvAUX(hv) ((struct xpvhv_aux*)&(HvARRAY(hv)[HvMAX(hv)+1]))
bfcb3514	227	#define HvRITER(hv) (Perl_hv_riter_p(aTHX_ (HV)(hv)))
	228	#define HvEITER(hv) (Perl_hv_eiter_p(aTHX_ (HV)(hv)))
	229	#define HvRITER_set(hv,r) Perl_hv_riter_set(aTHX_ (HV*)(hv), r)
	230	#define HvEITER_set(hv,e) Perl_hv_eiter_set(aTHX_ (HV*)(hv), e)
b79f7545	231	#define HvRITER_get(hv) (SvOOK(hv) ? HvAUX(hv)->xhv_riter : -1)
b79f7545	232	#define HvEITER_get(hv) (SvOOK(hv) ? HvAUX(hv)->xhv_eiter : 0)
7423f6db	233	#define HvNAME(hv) HvNAME_get(hv)
bfcb3514	234	/* FIXME - all of these should use a UTF8 aware API, which should also involve
bfcb3514	235	getting the length. */
9b9d0b15	236	/* This macro may go away without notice. */
b79f7545	237	#define HvNAME_HEK(hv) (SvOOK(hv) ? HvAUX(hv)->xhv_name : 0)
	238	#define HvNAME_get(hv) ((SvOOK(hv) && (HvAUX(hv)->xhv_name)) \
	239	? HEK_KEY(HvAUX(hv)->xhv_name) : 0)
	240	#define HvNAMELEN_get(hv) ((SvOOK(hv) && (HvAUX(hv)->xhv_name)) \
	241	? HEK_LEN(HvAUX(hv)->xhv_name) : 0)
a0d0e21e	242
8aacddc1	243	/* the number of keys (including any placeholers) */
	244	#define XHvTOTALKEYS(xhv) ((xhv)->xhv_keys)
	245
8aacddc1	246	/*
	247	* HvKEYS gets the number of keys that actually exist(), and is provided
	248	* for backwards compatibility with old XS code. The core uses HvUSEDKEYS
	249	* (keys, excluding placeholdes) and HvTOTALKEYS (including placeholders)
	250	*/
ca732855	251	#define HvKEYS(hv) HvUSEDKEYS(hv)
ca732855	252	#define HvUSEDKEYS(hv) (HvTOTALKEYS(hv) - HvPLACEHOLDERS_get(hv))
8aacddc1	253	#define HvTOTALKEYS(hv) XHvTOTALKEYS((XPVHV*) SvANY(hv))
ca732855	254	#define HvPLACEHOLDERS(hv) (Perl_hv_placeholders_p(aTHX_ (HV)hv))
	255	#define HvPLACEHOLDERS_get(hv) (SvMAGIC(hv) ? Perl_hv_placeholders_get(aTHX_ (HV*)hv) : 0)
	256	#define HvPLACEHOLDERS_set(hv,p) Perl_hv_placeholders_set(aTHX_ (HV*)hv, p)
8aacddc1	257
bf6bd887	258	#define HvSHAREKEYS(hv) (SvFLAGS(hv) & SVphv_SHAREKEYS)
	259	#define HvSHAREKEYS_on(hv) (SvFLAGS(hv) \|= SVphv_SHAREKEYS)
	260	#define HvSHAREKEYS_off(hv) (SvFLAGS(hv) &= ~SVphv_SHAREKEYS)
	261
19692e8d	262	/* This is an optimisation flag. It won't be set if all hash keys have a 0
	263	* flag. Currently the only flags relate to utf8.
	264	* Hence it won't be set if all keys are 8 bit only. It will be set if any key
	265	* is utf8 (including 8 bit keys that were entered as utf8, and need upgrading
	266	* when retrieved during iteration. It may still be set when there are no longer
	267	* any utf8 keys.
fdcd69b6	268	* See HVhek_ENABLEHVKFLAGS for the trigger.
19692e8d	269	*/
	270	#define HvHASKFLAGS(hv) (SvFLAGS(hv) & SVphv_HASKFLAGS)
	271	#define HvHASKFLAGS_on(hv) (SvFLAGS(hv) \|= SVphv_HASKFLAGS)
	272	#define HvHASKFLAGS_off(hv) (SvFLAGS(hv) &= ~SVphv_HASKFLAGS)
574c8022	273
bf5b86ae	274	#define HvLAZYDEL(hv) (SvFLAGS(hv) & SVphv_LAZYDEL)
	275	#define HvLAZYDEL_on(hv) (SvFLAGS(hv) \|= SVphv_LAZYDEL)
	276	#define HvLAZYDEL_off(hv) (SvFLAGS(hv) &= ~SVphv_LAZYDEL)
	277
4b5190b5	278	#define HvREHASH(hv) (SvFLAGS(hv) & SVphv_REHASH)
	279	#define HvREHASH_on(hv) (SvFLAGS(hv) \|= SVphv_REHASH)
	280	#define HvREHASH_off(hv) (SvFLAGS(hv) &= ~SVphv_REHASH)
	281
a0d0e21e	282	/* Maybe amagical: */
	283	/* #define HV_AMAGICmb(hv) (SvFLAGS(hv) & (SVpgv_badAM \| SVpgv_AM)) */
	284
	285	#define HV_AMAGIC(hv) (SvFLAGS(hv) & SVpgv_AM)
	286	#define HV_AMAGIC_on(hv) (SvFLAGS(hv) \|= SVpgv_AM)
	287	#define HV_AMAGIC_off(hv) (SvFLAGS(hv) &= ~SVpgv_AM)
	288
	289	/*
	290	#define HV_AMAGICbad(hv) (SvFLAGS(hv) & SVpgv_badAM)
	291	#define HV_badAMAGIC_on(hv) (SvFLAGS(hv) \|= SVpgv_badAM)
	292	#define HV_badAMAGIC_off(hv) (SvFLAGS(hv) &= ~SVpgv_badAM)
	293	*/
bf6bd887	294
	295	#define Nullhe Null(HE*)
	296	#define HeNEXT(he) (he)->hent_next
ff68c719	297	#define HeKEY_hek(he) (he)->hent_hek
ff68c719	298	#define HeKEY(he) HEK_KEY(HeKEY_hek(he))
bbce6d69	299	#define HeKEY_sv(he) ((SV*)HeKEY(he))
ff68c719	300	#define HeKLEN(he) HEK_LEN(HeKEY_hek(he))
da58a35d	301	#define HeKUTF8(he) HEK_UTF8(HeKEY_hek(he))
19692e8d	302	#define HeKWASUTF8(he) HEK_WASUTF8(HeKEY_hek(he))
4b5190b5	303	#define HeKREHASH(he) HEK_REHASH(HeKEY_hek(he))
da58a35d	304	#define HeKLEN_UTF8(he) (HeKUTF8(he) ? -HeKLEN(he) : HeKLEN(he))
19692e8d	305	#define HeKFLAGS(he) HEK_FLAGS(HeKEY_hek(he))
de616631	306	#define HeVAL(he) (he)->he_valu.hent_val
ff68c719	307	#define HeHASH(he) HEK_HASH(HeKEY_hek(he))
1e422769	308	#define HePV(he,lp) ((HeKLEN(he) == HEf_SVKEY) ? \
	309	SvPV(HeKEY_sv(he),lp) : \
	310	(((lp = HeKLEN(he)) >= 0) ? \
bd61b366	311	HeKEY(he) : NULL))
1e422769	312
bbce6d69	313	#define HeSVKEY(he) ((HeKEY(he) && \
bbce6d69	314	HeKLEN(he) == HEf_SVKEY) ? \
a0714e2c	315	HeKEY_sv(he) : NULL)
bbce6d69	316
	317	#define HeSVKEY_force(he) (HeKEY(he) ? \
	318	((HeKLEN(he) == HEf_SVKEY) ? \
	319	HeKEY_sv(he) : \
79cb57f6	320	sv_2mortal(newSVpvn(HeKEY(he), \
bbce6d69	321	HeKLEN(he)))) : \
3280af22	322	&PL_sv_undef)
1e422769	323	#define HeSVKEY_set(he,sv) ((HeKLEN(he) = HEf_SVKEY), (HeKEY_sv(he) = sv))
bbce6d69	324
ff68c719	325	#define Nullhek Null(HEK*)
71be2cbc	326	#define HEK_BASESIZE STRUCT_OFFSET(HEK, hek_key[0])
ff68c719	327	#define HEK_HASH(hek) (hek)->hek_hash
	328	#define HEK_LEN(hek) (hek)->hek_len
	329	#define HEK_KEY(hek) (hek)->hek_key
19692e8d	330	#define HEK_FLAGS(hek) (((unsigned char )(HEK_KEY(hek))+HEK_LEN(hek)+1))
	331
	332	#define HVhek_UTF8 0x01 /* Key is utf8 encoded. */
	333	#define HVhek_WASUTF8 0x02 /* Key is bytes here, but was supplied as utf8. */
4b5190b5	334	#define HVhek_REHASH 0x04 /* This key is in an hv using a custom HASH . */
19692e8d	335	#define HVhek_FREEKEY 0x100 /* Internal flag to say key is malloc()ed. */
e16e2ff8	336	#define HVhek_PLACEHOLD 0x200 /* Internal flag to create placeholder.
e16e2ff8	337	* (may change, but Storable is a core module) */
19692e8d	338	#define HVhek_MASK 0xFF
19692e8d	339
fdcd69b6	340	/* Which flags enable HvHASKFLAGS? Somewhat a hack on a hack, as
	341	HVhek_REHASH is only needed because the rehash flag has to be duplicated
	342	into all keys as hv_iternext has no access to the hash flags. At this
	343	point Storable's tests get upset, because sometimes hashes are "keyed"
	344	and sometimes not, depending on the order of data insertion, and whether
	345	it triggered rehashing. So currently HVhek_REHAS is exempt.
	346	*/
	347
	348	#define HVhek_ENABLEHVKFLAGS (HVhek_MASK - HVhek_REHASH)
	349
19692e8d	350	#define HEK_UTF8(hek) (HEK_FLAGS(hek) & HVhek_UTF8)
	351	#define HEK_UTF8_on(hek) (HEK_FLAGS(hek) \|= HVhek_UTF8)
	352	#define HEK_UTF8_off(hek) (HEK_FLAGS(hek) &= ~HVhek_UTF8)
	353	#define HEK_WASUTF8(hek) (HEK_FLAGS(hek) & HVhek_WASUTF8)
	354	#define HEK_WASUTF8_on(hek) (HEK_FLAGS(hek) \|= HVhek_WASUTF8)
	355	#define HEK_WASUTF8_off(hek) (HEK_FLAGS(hek) &= ~HVhek_WASUTF8)
4b5190b5	356	#define HEK_REHASH(hek) (HEK_FLAGS(hek) & HVhek_REHASH)
4b5190b5	357	#define HEK_REHASH_on(hek) (HEK_FLAGS(hek) \|= HVhek_REHASH)
d18c6117	358
5cbe4eec	359	/* calculate HV array allocation */
36768cf4	360	#ifndef PERL_USE_LARGE_HV_ALLOC
	361	/* Default to allocating the correct size - default to assuming that malloc()
	362	is not broken and is efficient at allocating blocks sized at powers-of-two.
	363	*/
d18c6117	364	# define PERL_HV_ARRAY_ALLOC_BYTES(size) ((size) * sizeof(HE*))
	365	#else
	366	# define MALLOC_OVERHEAD 16
	367	# define PERL_HV_ARRAY_ALLOC_BYTES(size) \
	368	(((size) < 64) \
	369	? (size) * sizeof(HE*) \
	370	: (size) * sizeof(HE) 2 - MALLOC_OVERHEAD)
	371	#endif
37d85e3a	372
e16e2ff8	373	/* Flags for hv_iternext_flags. */
	374	#define HV_ITERNEXT_WANTPLACEHOLDERS 0x01 /* Don't skip placeholders. */
	375
7a7b9979	376	#define hv_iternext(hv) hv_iternext_flags(hv, 0)
bd61b366	377	#define hv_magic(hv, gv, how) sv_magic((SV)(hv), (SV)(gv), how, NULL, 0)
7a7b9979	378
37d85e3a	379	/* available as a function in hv.c */
	380	#define Perl_sharepvn(sv, len, hash) HEK_KEY(share_hek(sv, len, hash))
	381	#define sharepvn(sv, len, hash) Perl_sharepvn(sv, len, hash)
bfcb3514	382
d1db91c6	383	#define share_hek_hek(hek) \
	384	(++(((struct shared_he )(((char )hek) \
	385	- STRUCT_OFFSET(struct shared_he, \
	386	shared_he_hek))) \
de616631	387	->shared_he_he.he_valu.hent_refcount), \
d1db91c6	388	hek)
d1db91c6	389
bfcb3514	390	/*
	391	* Local variables:
	392	* c-indentation-style: bsd
	393	* c-basic-offset: 4
	394	* indent-tabs-mode: t
	395	* End:
	396	*
	397	* ex: set ts=8 sts=4 sw=4 noet:
	398	*/