| 1 | /* hv.h
|
|---|
| 2 | *
|
|---|
| 3 | * Copyright (C) 1991, 1992, 1993, 1996, 1997, 1998, 1999,
|
|---|
| 4 | * 2000, 2001, 2002, 2005, by Larry Wall and others
|
|---|
| 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 | *
|
|---|
| 9 | */
|
|---|
| 10 |
|
|---|
| 11 | /* typedefs to eliminate some typing */
|
|---|
| 12 | typedef struct he HE;
|
|---|
| 13 | typedef struct hek HEK;
|
|---|
| 14 |
|
|---|
| 15 | /* entry in hash value chain */
|
|---|
| 16 | struct he {
|
|---|
| 17 | HE *hent_next; /* next entry in chain */
|
|---|
| 18 | HEK *hent_hek; /* hash key */
|
|---|
| 19 | SV *hent_val; /* scalar value that was hashed */
|
|---|
| 20 | };
|
|---|
| 21 |
|
|---|
| 22 | /* hash key -- defined separately for use as shared pointer */
|
|---|
| 23 | struct hek {
|
|---|
| 24 | U32 hek_hash; /* hash of key */
|
|---|
| 25 | I32 hek_len; /* length of hash key */
|
|---|
| 26 | char hek_key[1]; /* variable-length hash key */
|
|---|
| 27 | /* the hash-key is \0-terminated */
|
|---|
| 28 | /* after the \0 there is a byte for flags, such as whether the key
|
|---|
| 29 | is UTF-8 */
|
|---|
| 30 | };
|
|---|
| 31 |
|
|---|
| 32 | /* hash structure: */
|
|---|
| 33 | /* This structure must match the beginning of struct xpvmg in sv.h. */
|
|---|
| 34 | struct xpvhv {
|
|---|
| 35 | char * xhv_array; /* pointer to malloced string */
|
|---|
| 36 | STRLEN xhv_fill; /* how full xhv_array currently is */
|
|---|
| 37 | STRLEN xhv_max; /* subscript of last element of xhv_array */
|
|---|
| 38 | IV xhv_keys; /* how many elements in the array */
|
|---|
| 39 | NV xnv_nv; /* numeric value, if any */
|
|---|
| 40 | #define xhv_placeholders xnv_nv
|
|---|
| 41 | MAGIC* xmg_magic; /* magic for scalar array */
|
|---|
| 42 | HV* xmg_stash; /* class package */
|
|---|
| 43 |
|
|---|
| 44 | I32 xhv_riter; /* current root of iterator */
|
|---|
| 45 | HE *xhv_eiter; /* current entry of iterator */
|
|---|
| 46 | PMOP *xhv_pmroot; /* list of pm's for this package */
|
|---|
| 47 | char *xhv_name; /* name, if a symbol table */
|
|---|
| 48 | };
|
|---|
| 49 |
|
|---|
| 50 | /* hash a key */
|
|---|
| 51 | /* FYI: This is the "One-at-a-Time" algorithm by Bob Jenkins
|
|---|
| 52 | * from requirements by Colin Plumb.
|
|---|
| 53 | * (http://burtleburtle.net/bob/hash/doobs.html) */
|
|---|
| 54 | /* The use of a temporary pointer and the casting games
|
|---|
| 55 | * is needed to serve the dual purposes of
|
|---|
| 56 | * (a) the hashed data being interpreted as "unsigned char" (new since 5.8,
|
|---|
| 57 | * a "char" can be either signed or unsigned, depending on the compiler)
|
|---|
| 58 | * (b) catering for old code that uses a "char"
|
|---|
| 59 | *
|
|---|
| 60 | * The "hash seed" feature was added in Perl 5.8.1 to perturb the results
|
|---|
| 61 | * to avoid "algorithmic complexity attacks".
|
|---|
| 62 | *
|
|---|
| 63 | * If USE_HASH_SEED is defined, hash randomisation is done by default
|
|---|
| 64 | * If USE_HASH_SEED_EXPLICIT is defined, hash randomisation is done
|
|---|
| 65 | * only if the environment variable PERL_HASH_SEED is set.
|
|---|
| 66 | * For maximal control, one can define PERL_HASH_SEED.
|
|---|
| 67 | * (see also perl.c:perl_parse()).
|
|---|
| 68 | */
|
|---|
| 69 | #ifndef PERL_HASH_SEED
|
|---|
| 70 | # if defined(USE_HASH_SEED) || defined(USE_HASH_SEED_EXPLICIT)
|
|---|
| 71 | # define PERL_HASH_SEED PL_hash_seed
|
|---|
| 72 | # else
|
|---|
| 73 | # define PERL_HASH_SEED 0
|
|---|
| 74 | # endif
|
|---|
| 75 | #endif
|
|---|
| 76 | #define PERL_HASH(hash,str,len) \
|
|---|
| 77 | STMT_START { \
|
|---|
| 78 | register const char *s_PeRlHaSh_tmp = str; \
|
|---|
| 79 | register const unsigned char *s_PeRlHaSh = (const unsigned char *)s_PeRlHaSh_tmp; \
|
|---|
| 80 | register I32 i_PeRlHaSh = len; \
|
|---|
| 81 | register U32 hash_PeRlHaSh = PERL_HASH_SEED; \
|
|---|
| 82 | while (i_PeRlHaSh--) { \
|
|---|
| 83 | hash_PeRlHaSh += *s_PeRlHaSh++; \
|
|---|
| 84 | hash_PeRlHaSh += (hash_PeRlHaSh << 10); \
|
|---|
| 85 | hash_PeRlHaSh ^= (hash_PeRlHaSh >> 6); \
|
|---|
| 86 | } \
|
|---|
| 87 | hash_PeRlHaSh += (hash_PeRlHaSh << 3); \
|
|---|
| 88 | hash_PeRlHaSh ^= (hash_PeRlHaSh >> 11); \
|
|---|
| 89 | (hash) = (hash_PeRlHaSh + (hash_PeRlHaSh << 15)); \
|
|---|
| 90 | } STMT_END
|
|---|
| 91 |
|
|---|
| 92 | /* Only hv.c and mod_perl should be doing this. */
|
|---|
| 93 | #ifdef PERL_HASH_INTERNAL_ACCESS
|
|---|
| 94 | #define PERL_HASH_INTERNAL(hash,str,len) \
|
|---|
| 95 | STMT_START { \
|
|---|
| 96 | register const char *s_PeRlHaSh_tmp = str; \
|
|---|
| 97 | register const unsigned char *s_PeRlHaSh = (const unsigned char *)s_PeRlHaSh_tmp; \
|
|---|
| 98 | register I32 i_PeRlHaSh = len; \
|
|---|
| 99 | register U32 hash_PeRlHaSh = PL_rehash_seed; \
|
|---|
| 100 | while (i_PeRlHaSh--) { \
|
|---|
| 101 | hash_PeRlHaSh += *s_PeRlHaSh++; \
|
|---|
| 102 | hash_PeRlHaSh += (hash_PeRlHaSh << 10); \
|
|---|
| 103 | hash_PeRlHaSh ^= (hash_PeRlHaSh >> 6); \
|
|---|
| 104 | } \
|
|---|
| 105 | hash_PeRlHaSh += (hash_PeRlHaSh << 3); \
|
|---|
| 106 | hash_PeRlHaSh ^= (hash_PeRlHaSh >> 11); \
|
|---|
| 107 | (hash) = (hash_PeRlHaSh + (hash_PeRlHaSh << 15)); \
|
|---|
| 108 | } STMT_END
|
|---|
| 109 | #endif
|
|---|
| 110 |
|
|---|
| 111 | /*
|
|---|
| 112 | =head1 Hash Manipulation Functions
|
|---|
| 113 |
|
|---|
| 114 | =for apidoc AmU||HEf_SVKEY
|
|---|
| 115 | This flag, used in the length slot of hash entries and magic structures,
|
|---|
| 116 | specifies the structure contains an C<SV*> pointer where a C<char*> pointer
|
|---|
| 117 | is to be expected. (For information only--not to be used).
|
|---|
| 118 |
|
|---|
| 119 | =head1 Handy Values
|
|---|
| 120 |
|
|---|
| 121 | =for apidoc AmU||Nullhv
|
|---|
| 122 | Null HV pointer.
|
|---|
| 123 |
|
|---|
| 124 | =head1 Hash Manipulation Functions
|
|---|
| 125 |
|
|---|
| 126 | =for apidoc Am|char*|HvNAME|HV* stash
|
|---|
| 127 | Returns the package name of a stash, or NULL if C<stash> isn't a stash.
|
|---|
| 128 | See C<SvSTASH>, C<CvSTASH>.
|
|---|
| 129 |
|
|---|
| 130 | =for apidoc Am|void*|HeKEY|HE* he
|
|---|
| 131 | Returns the actual pointer stored in the key slot of the hash entry. The
|
|---|
| 132 | pointer may be either C<char*> or C<SV*>, depending on the value of
|
|---|
| 133 | C<HeKLEN()>. Can be assigned to. The C<HePV()> or C<HeSVKEY()> macros are
|
|---|
| 134 | usually preferable for finding the value of a key.
|
|---|
| 135 |
|
|---|
| 136 | =for apidoc Am|STRLEN|HeKLEN|HE* he
|
|---|
| 137 | If this is negative, and amounts to C<HEf_SVKEY>, it indicates the entry
|
|---|
| 138 | holds an C<SV*> key. Otherwise, holds the actual length of the key. Can
|
|---|
| 139 | be assigned to. The C<HePV()> macro is usually preferable for finding key
|
|---|
| 140 | lengths.
|
|---|
| 141 |
|
|---|
| 142 | =for apidoc Am|SV*|HeVAL|HE* he
|
|---|
| 143 | Returns the value slot (type C<SV*>) stored in the hash entry.
|
|---|
| 144 |
|
|---|
| 145 | =for apidoc Am|U32|HeHASH|HE* he
|
|---|
| 146 | Returns the computed hash stored in the hash entry.
|
|---|
| 147 |
|
|---|
| 148 | =for apidoc Am|char*|HePV|HE* he|STRLEN len
|
|---|
| 149 | Returns the key slot of the hash entry as a C<char*> value, doing any
|
|---|
| 150 | necessary dereferencing of possibly C<SV*> keys. The length of the string
|
|---|
| 151 | is placed in C<len> (this is a macro, so do I<not> use C<&len>). If you do
|
|---|
| 152 | not care about what the length of the key is, you may use the global
|
|---|
| 153 | variable C<PL_na>, though this is rather less efficient than using a local
|
|---|
| 154 | variable. Remember though, that hash keys in perl are free to contain
|
|---|
| 155 | embedded nulls, so using C<strlen()> or similar is not a good way to find
|
|---|
| 156 | the length of hash keys. This is very similar to the C<SvPV()> macro
|
|---|
| 157 | described elsewhere in this document.
|
|---|
| 158 |
|
|---|
| 159 | =for apidoc Am|SV*|HeSVKEY|HE* he
|
|---|
| 160 | Returns the key as an C<SV*>, or C<Nullsv> if the hash entry does not
|
|---|
| 161 | contain an C<SV*> key.
|
|---|
| 162 |
|
|---|
| 163 | =for apidoc Am|SV*|HeSVKEY_force|HE* he
|
|---|
| 164 | Returns the key as an C<SV*>. Will create and return a temporary mortal
|
|---|
| 165 | C<SV*> if the hash entry contains only a C<char*> key.
|
|---|
| 166 |
|
|---|
| 167 | =for apidoc Am|SV*|HeSVKEY_set|HE* he|SV* sv
|
|---|
| 168 | Sets the key to a given C<SV*>, taking care to set the appropriate flags to
|
|---|
| 169 | indicate the presence of an C<SV*> key, and returns the same
|
|---|
| 170 | C<SV*>.
|
|---|
| 171 |
|
|---|
| 172 | =cut
|
|---|
| 173 | */
|
|---|
| 174 |
|
|---|
| 175 | /* these hash entry flags ride on hent_klen (for use only in magic/tied HVs) */
|
|---|
| 176 | #define HEf_SVKEY -2 /* hent_key is an SV* */
|
|---|
| 177 |
|
|---|
| 178 |
|
|---|
| 179 | #define Nullhv Null(HV*)
|
|---|
| 180 | #define HvARRAY(hv) (*(HE***)&((XPVHV*) SvANY(hv))->xhv_array)
|
|---|
| 181 | #define HvFILL(hv) ((XPVHV*) SvANY(hv))->xhv_fill
|
|---|
| 182 | #define HvMAX(hv) ((XPVHV*) SvANY(hv))->xhv_max
|
|---|
| 183 | #define HvRITER(hv) ((XPVHV*) SvANY(hv))->xhv_riter
|
|---|
| 184 | #define HvRITER_get(hv) (0 + ((XPVHV*) SvANY(hv))->xhv_riter)
|
|---|
| 185 | #define HvRITER_set(hv,r) (HvRITER(hv) = (r))
|
|---|
| 186 | #define HvEITER(hv) ((XPVHV*) SvANY(hv))->xhv_eiter
|
|---|
| 187 | #define HvEITER_get(hv) (0 + ((XPVHV*) SvANY(hv))->xhv_eiter)
|
|---|
| 188 | #define HvEITER_set(hv,e) (HvEITER(hv) = (e))
|
|---|
| 189 | #define HvPMROOT(hv) ((XPVHV*) SvANY(hv))->xhv_pmroot
|
|---|
| 190 | #define HvNAME(hv) ((XPVHV*) SvANY(hv))->xhv_name
|
|---|
| 191 | /* FIXME - all of these should use a UTF8 aware API, which should also involve
|
|---|
| 192 | getting the length. */
|
|---|
| 193 | #define HvNAME_get(hv) (0 + ((XPVHV*) SvANY(hv))->xhv_name)
|
|---|
| 194 | #define hv_name_set(hv,name,length,flags) \
|
|---|
| 195 | (HvNAME((hv)) = (name) ? savepvn(name, length) : 0)
|
|---|
| 196 |
|
|---|
| 197 | /* the number of keys (including any placeholers) */
|
|---|
| 198 | #define XHvTOTALKEYS(xhv) ((xhv)->xhv_keys)
|
|---|
| 199 |
|
|---|
| 200 | /* The number of placeholders in the enumerated-keys hash */
|
|---|
| 201 | #define XHvPLACEHOLDERS(xhv) ((xhv)->xhv_placeholders)
|
|---|
| 202 |
|
|---|
| 203 | /* the number of keys that exist() (i.e. excluding placeholders) */
|
|---|
| 204 | #define XHvUSEDKEYS(xhv) (XHvTOTALKEYS(xhv) - (IV)XHvPLACEHOLDERS(xhv))
|
|---|
| 205 |
|
|---|
| 206 | /*
|
|---|
| 207 | * HvKEYS gets the number of keys that actually exist(), and is provided
|
|---|
| 208 | * for backwards compatibility with old XS code. The core uses HvUSEDKEYS
|
|---|
| 209 | * (keys, excluding placeholdes) and HvTOTALKEYS (including placeholders)
|
|---|
| 210 | */
|
|---|
| 211 | #define HvKEYS(hv) XHvUSEDKEYS((XPVHV*) SvANY(hv))
|
|---|
| 212 | #define HvUSEDKEYS(hv) XHvUSEDKEYS((XPVHV*) SvANY(hv))
|
|---|
| 213 | #define HvTOTALKEYS(hv) XHvTOTALKEYS((XPVHV*) SvANY(hv))
|
|---|
| 214 | #define HvPLACEHOLDERS(hv) (XHvPLACEHOLDERS((XPVHV*) SvANY(hv)))
|
|---|
| 215 | #define HvPLACEHOLDERS_get(hv) (0 + XHvPLACEHOLDERS((XPVHV*) SvANY(hv)))
|
|---|
| 216 | #define HvPLACEHOLDERS_set(hv, p) \
|
|---|
| 217 | (XHvPLACEHOLDERS((XPVHV*) SvANY(hv)) = (p))
|
|---|
| 218 |
|
|---|
| 219 | #define HvSHAREKEYS(hv) (SvFLAGS(hv) & SVphv_SHAREKEYS)
|
|---|
|
|---|
