diff options
Diffstat (limited to 'hash.c')
-rw-r--r-- | hash.c | 2855 |
1 files changed, 1537 insertions, 1318 deletions
@@ -28,13 +28,16 @@ #include "internal.h" #include "internal/array.h" #include "internal/bignum.h" +#include "internal/basic_operators.h" #include "internal/class.h" #include "internal/cont.h" #include "internal/error.h" #include "internal/hash.h" #include "internal/object.h" #include "internal/proc.h" +#include "internal/st.h" #include "internal/symbol.h" +#include "internal/thread.h" #include "internal/time.h" #include "internal/vm.h" #include "probes.h" @@ -42,14 +45,33 @@ #include "ruby/util.h" #include "ruby_assert.h" #include "symbol.h" -#include "transient_heap.h" +#include "ruby/thread_native.h" +#include "ruby/ractor.h" +#include "vm_sync.h" + +/* Flags of RHash + * + * 1: RHASH_PASS_AS_KEYWORDS + * The hash is flagged as Ruby 2 keywords hash. + * 2: RHASH_PROC_DEFAULT + * The hash has a default proc (rather than a default value). + * 3: RHASH_ST_TABLE_FLAG + * The hash uses a ST table (rather than an AR table). + * 4-7: RHASH_AR_TABLE_SIZE_MASK + * The size of the AR table. + * 8-11: RHASH_AR_TABLE_BOUND_MASK + * The bounds of the AR table. + * 13-19: RHASH_LEV_MASK + * The iterational level of the hash. Used to prevent modifications + * to the hash during interation. + */ #ifndef HASH_DEBUG #define HASH_DEBUG 0 #endif #if HASH_DEBUG -#include "gc.h" +#include "internal/gc.h" #endif #define SET_DEFAULT(hash, ifnone) ( \ @@ -89,9 +111,11 @@ rb_hash_freeze(VALUE hash) VALUE rb_cHash; static VALUE envtbl; -static ID id_hash, id_default, id_flatten_bang; +static ID id_hash, id_flatten_bang; static ID id_hash_iter_lev; +#define id_default idDefault + VALUE rb_hash_set_ifnone(VALUE hash, VALUE ifnone) { @@ -99,17 +123,17 @@ rb_hash_set_ifnone(VALUE hash, VALUE ifnone) return hash; } -static int +int rb_any_cmp(VALUE a, VALUE b) { if (a == b) return 0; if (RB_TYPE_P(a, T_STRING) && RBASIC(a)->klass == rb_cString && - RB_TYPE_P(b, T_STRING) && RBASIC(b)->klass == rb_cString) { - return rb_str_hash_cmp(a, b); + RB_TYPE_P(b, T_STRING) && RBASIC(b)->klass == rb_cString) { + return rb_str_hash_cmp(a, b); } - if (a == Qundef || b == Qundef) return -1; + if (UNDEF_P(a) || UNDEF_P(b)) return -1; if (SYMBOL_P(a) && SYMBOL_P(b)) { - return a != b; + return a != b; } return !rb_eql(a, b); @@ -152,7 +176,7 @@ any_hash(VALUE a, st_index_t (*other_func)(VALUE)) switch (TYPE(a)) { case T_SYMBOL: - if (STATIC_SYM_P(a)) { + if (STATIC_SYM_P(a)) { hnum = a >> (RUBY_SPECIAL_SHIFT + ID_SCOPE_SHIFT); hnum = rb_hash_start(hnum); } @@ -164,35 +188,51 @@ any_hash(VALUE a, st_index_t (*other_func)(VALUE)) case T_TRUE: case T_FALSE: case T_NIL: - hnum = rb_objid_hash((st_index_t)a); + hnum = rb_objid_hash((st_index_t)a); break; case T_STRING: - hnum = rb_str_hash(a); + hnum = rb_str_hash(a); break; case T_BIGNUM: - hval = rb_big_hash(a); - hnum = FIX2LONG(hval); + hval = rb_big_hash(a); + hnum = FIX2LONG(hval); break; case T_FLOAT: /* prevent pathological behavior: [Bug #10761] */ - hnum = rb_dbl_long_hash(rb_float_value(a)); + hnum = rb_dbl_long_hash(rb_float_value(a)); break; default: - hnum = other_func(a); + hnum = other_func(a); } if ((SIGNED_VALUE)hnum > 0) - hnum &= FIXNUM_MAX; + hnum &= FIXNUM_MAX; else - hnum |= FIXNUM_MIN; + hnum |= FIXNUM_MIN; return (long)hnum; } +VALUE rb_obj_hash(VALUE obj); +VALUE rb_vm_call0(rb_execution_context_t *ec, VALUE recv, ID id, int argc, const VALUE *argv, const rb_callable_method_entry_t *cme, int kw_splat); + static st_index_t obj_any_hash(VALUE obj) { - VALUE hval = rb_check_funcall_basic_kw(obj, id_hash, rb_mKernel, 0, 0, 0); + VALUE hval = Qundef; + VALUE klass = CLASS_OF(obj); + if (klass) { + const rb_callable_method_entry_t *cme = rb_callable_method_entry(klass, id_hash); + if (cme && METHOD_ENTRY_BASIC(cme)) { + // Optimize away the frame push overhead if it's the default Kernel#hash + if (cme->def->type == VM_METHOD_TYPE_CFUNC && cme->def->body.cfunc.func == (rb_cfunc_t)rb_obj_hash) { + hval = rb_obj_hash(obj); + } + else if (RBASIC_CLASS(cme->defined_class) == rb_mKernel) { + hval = rb_vm_call0(GET_EC(), obj, id_hash, 0, 0, cme, 0); + } + } + } - if (hval == Qundef) { - hval = rb_exec_recursive_outer(hash_recursive, obj, 0); + if (UNDEF_P(hval)) { + hval = rb_exec_recursive_outer_mid(hash_recursive, obj, 0, id_hash); } while (!FIXNUM_P(hval)) { @@ -214,7 +254,7 @@ obj_any_hash(VALUE obj) return FIX2LONG(hval); } -static st_index_t +st_index_t rb_any_hash(VALUE a) { return any_hash(a, obj_any_hash); @@ -300,6 +340,19 @@ objid_hash(VALUE obj) * * Certain core classes such as Integer use built-in hash calculations and * do not call the #hash method when used as a hash key. + * + * When implementing your own #hash based on multiple values, the best + * practice is to combine the class and any values using the hash code of an + * array: + * + * For example: + * + * def hash + * [self.class, a, b, c].hash + * end + * + * The reason for this is that the Array#hash method already has logic for + * safely and efficiently combining multiple hash values. *-- * \private *++ @@ -341,16 +394,17 @@ const struct st_hash_type rb_hashtype_ident = { rb_ident_hash, }; +#define RHASH_IDENTHASH_P(hash) (RHASH_TYPE(hash) == &identhash) +#define RHASH_STRING_KEY_P(hash, key) (!RHASH_IDENTHASH_P(hash) && (rb_obj_class(key) == rb_cString)) + typedef st_index_t st_hash_t; /* * RHASH_AR_TABLE_P(h): - * * as.ar == NULL or - * as.ar points ar_table. - * * as.ar is allocated by transient heap or xmalloc. + * RHASH_AR_TABLE points to ar_table. * * !RHASH_AR_TABLE_P(h): - * * as.st points st_table. + * RHASH_ST_TABLE points st_table. */ #define RHASH_AR_TABLE_MAX_BOUND RHASH_AR_TABLE_MAX_SIZE @@ -358,22 +412,6 @@ typedef st_index_t st_hash_t; #define RHASH_AR_TABLE_REF(hash, n) (&RHASH_AR_TABLE(hash)->pairs[n]) #define RHASH_AR_CLEARED_HINT 0xff -typedef struct ar_table_pair_struct { - VALUE key; - VALUE val; -} ar_table_pair; - -typedef struct ar_table_struct { - /* 64bit CPU: 8B * 2 * 8 = 128B */ - ar_table_pair pairs[RHASH_AR_TABLE_MAX_SIZE]; -} ar_table; - -size_t -rb_hash_ar_table_size(void) -{ - return sizeof(ar_table); -} - static inline st_hash_t ar_do_hash(st_data_t key) { @@ -389,13 +427,13 @@ ar_do_hash_hint(st_hash_t hash_value) static inline ar_hint_t ar_hint(VALUE hash, unsigned int index) { - return RHASH(hash)->ar_hint.ary[index]; + return RHASH_AR_TABLE(hash)->ar_hint.ary[index]; } static inline void ar_hint_set_hint(VALUE hash, unsigned int index, ar_hint_t hint) { - RHASH(hash)->ar_hint.ary[index] = hint; + RHASH_AR_TABLE(hash)->ar_hint.ary[index] = hint; } static inline void @@ -420,7 +458,7 @@ ar_cleared_entry(VALUE hash, unsigned int index) * so you need to check key == Qundef */ ar_table_pair *pair = RHASH_AR_TABLE_REF(hash, index); - return pair->key == Qundef; + return UNDEF_P(pair->key); } else { return FALSE; @@ -443,14 +481,20 @@ ar_set_entry(VALUE hash, unsigned int index, st_data_t key, st_data_t val, st_ha ((unsigned int)((RBASIC(h)->flags >> RHASH_AR_TABLE_BOUND_SHIFT) & \ (RHASH_AR_TABLE_BOUND_MASK >> RHASH_AR_TABLE_BOUND_SHIFT))) -#define RHASH_AR_TABLE_BOUND(h) (HASH_ASSERT(RHASH_AR_TABLE_P(h)), \ - RHASH_AR_TABLE_BOUND_RAW(h)) - #define RHASH_ST_TABLE_SET(h, s) rb_hash_st_table_set(h, s) #define RHASH_TYPE(hash) (RHASH_AR_TABLE_P(hash) ? &objhash : RHASH_ST_TABLE(hash)->type) #define HASH_ASSERT(expr) RUBY_ASSERT_MESG_WHEN(HASH_DEBUG, expr, #expr) +static inline unsigned int +RHASH_AR_TABLE_BOUND(VALUE h) +{ + HASH_ASSERT(RHASH_AR_TABLE_P(h)); + const unsigned int bound = RHASH_AR_TABLE_BOUND_RAW(h); + HASH_ASSERT(bound <= RHASH_AR_TABLE_MAX_SIZE); + return bound; +} + #if HASH_DEBUG #define hash_verify(hash) hash_verify_(hash, __FILE__, __LINE__) @@ -460,10 +504,10 @@ rb_hash_dump(VALUE hash) rb_obj_info_dump(hash); if (RHASH_AR_TABLE_P(hash)) { - unsigned i, n = 0, bound = RHASH_AR_TABLE_BOUND(hash); + unsigned i, bound = RHASH_AR_TABLE_BOUND(hash); fprintf(stderr, " size:%u bound:%u\n", - RHASH_AR_TABLE_SIZE(hash), RHASH_AR_TABLE_BOUND(hash)); + RHASH_AR_TABLE_SIZE(hash), bound); for (i=0; i<bound; i++) { st_data_t k, v; @@ -477,7 +521,6 @@ rb_hash_dump(VALUE hash) rb_raw_obj_info(b1, 0x100, k), rb_raw_obj_info(b2, 0x100, v), ar_hint(hash, i)); - n++; } else { fprintf(stderr, " %d empty\n", i); @@ -500,8 +543,8 @@ hash_verify_(VALUE hash, const char *file, int line) ar_table_pair *pair = RHASH_AR_TABLE_REF(hash, i); k = pair->key; v = pair->val; - HASH_ASSERT(k != Qundef); - HASH_ASSERT(v != Qundef); + HASH_ASSERT(!UNDEF_P(k)); + HASH_ASSERT(!UNDEF_P(v)); n++; } } @@ -515,13 +558,6 @@ hash_verify_(VALUE hash, const char *file, int line) HASH_ASSERT(RHASH_AR_TABLE_BOUND_RAW(hash) == 0); } -#if USE_TRANSIENT_HEAP - if (RHASH_TRANSIENT_P(hash)) { - volatile st_data_t MAYBE_UNUSED(key) = RHASH_AR_TABLE_REF(hash, 0)->key; /* read */ - HASH_ASSERT(RHASH_AR_TABLE(hash) != NULL); - HASH_ASSERT(rb_transient_heap_managed_ptr_p(RHASH_AR_TABLE(hash))); - } -#endif return hash; } @@ -530,69 +566,30 @@ hash_verify_(VALUE hash, const char *file, int line) #endif static inline int -RHASH_TABLE_NULL_P(VALUE hash) -{ - if (RHASH(hash)->as.ar == NULL) { - HASH_ASSERT(RHASH_AR_TABLE_P(hash)); - return TRUE; - } - else { - return FALSE; - } -} - -static inline int RHASH_TABLE_EMPTY_P(VALUE hash) { return RHASH_SIZE(hash) == 0; } -int -rb_hash_ar_table_p(VALUE hash) -{ - if (FL_TEST_RAW((hash), RHASH_ST_TABLE_FLAG)) { - HASH_ASSERT(RHASH(hash)->as.st != NULL); - return FALSE; - } - else { - return TRUE; - } -} - -ar_table * -rb_hash_ar_table(VALUE hash) -{ - HASH_ASSERT(RHASH_AR_TABLE_P(hash)); - return RHASH(hash)->as.ar; -} +#define RHASH_SET_ST_FLAG(h) FL_SET_RAW(h, RHASH_ST_TABLE_FLAG) +#define RHASH_UNSET_ST_FLAG(h) FL_UNSET_RAW(h, RHASH_ST_TABLE_FLAG) -st_table * -rb_hash_st_table(VALUE hash) +static void +hash_st_table_init(VALUE hash, const struct st_hash_type *type, st_index_t size) { - HASH_ASSERT(!RHASH_AR_TABLE_P(hash)); - return RHASH(hash)->as.st; + st_init_existing_table_with_size(RHASH_ST_TABLE(hash), type, size); + RHASH_SET_ST_FLAG(hash); } void rb_hash_st_table_set(VALUE hash, st_table *st) { HASH_ASSERT(st != NULL); - FL_SET_RAW((hash), RHASH_ST_TABLE_FLAG); - RHASH(hash)->as.st = st; -} + RHASH_SET_ST_FLAG(hash); -static void -hash_ar_table_set(VALUE hash, ar_table *ar) -{ - HASH_ASSERT(RHASH_AR_TABLE_P(hash)); - HASH_ASSERT((RHASH_TRANSIENT_P(hash) && ar == NULL) ? FALSE : TRUE); - RHASH(hash)->as.ar = ar; - hash_verify(hash); + *RHASH_ST_TABLE(hash) = *st; } -#define RHASH_SET_ST_FLAG(h) FL_SET_RAW(h, RHASH_ST_TABLE_FLAG) -#define RHASH_UNSET_ST_FLAG(h) FL_UNSET_RAW(h, RHASH_ST_TABLE_FLAG) - static inline void RHASH_AR_TABLE_BOUND_SET(VALUE h, st_index_t n) { @@ -647,27 +644,7 @@ RHASH_AR_TABLE_CLEAR(VALUE h) RBASIC(h)->flags &= ~RHASH_AR_TABLE_SIZE_MASK; RBASIC(h)->flags &= ~RHASH_AR_TABLE_BOUND_MASK; - hash_ar_table_set(h, NULL); -} - -static ar_table* -ar_alloc_table(VALUE hash) -{ - ar_table *tab = (ar_table*)rb_transient_heap_alloc(hash, sizeof(ar_table)); - - if (tab != NULL) { - RHASH_SET_TRANSIENT_FLAG(hash); - } - else { - RHASH_UNSET_TRANSIENT_FLAG(hash); - tab = (ar_table*)ruby_xmalloc(sizeof(ar_table)); - } - - RHASH_AR_TABLE_SIZE_SET(hash, 0); - RHASH_AR_TABLE_BOUND_SET(hash, 0); - hash_ar_table_set(hash, tab); - - return tab; + memset(RHASH_AR_TABLE(h), 0, sizeof(ar_table)); } NOINLINE(static int ar_equal(VALUE x, VALUE y)); @@ -682,7 +659,7 @@ static unsigned ar_find_entry_hint(VALUE hash, ar_hint_t hint, st_data_t key) { unsigned i, bound = RHASH_AR_TABLE_BOUND(hash); - const ar_hint_t *hints = RHASH(hash)->ar_hint.ary; + const ar_hint_t *hints = RHASH_AR_TABLE(hash)->ar_hint.ary; /* if table is NULL, then bound also should be 0 */ @@ -729,92 +706,83 @@ ar_find_entry(VALUE hash, st_hash_t hash_value, st_data_t key) } static inline void -ar_free_and_clear_table(VALUE hash) +hash_ar_free_and_clear_table(VALUE hash) { - ar_table *tab = RHASH_AR_TABLE(hash); + RHASH_AR_TABLE_CLEAR(hash); - if (tab) { - if (RHASH_TRANSIENT_P(hash)) { - RHASH_UNSET_TRANSIENT_FLAG(hash); - } - else { - ruby_xfree(RHASH_AR_TABLE(hash)); - } - RHASH_AR_TABLE_CLEAR(hash); - } HASH_ASSERT(RHASH_AR_TABLE_SIZE(hash) == 0); HASH_ASSERT(RHASH_AR_TABLE_BOUND(hash) == 0); - HASH_ASSERT(RHASH_TRANSIENT_P(hash) == 0); } -static void -ar_try_convert_table(VALUE hash) -{ - if (!RHASH_AR_TABLE_P(hash)) return; - - const unsigned size = RHASH_AR_TABLE_SIZE(hash); +void rb_st_add_direct_with_hash(st_table *tab, st_data_t key, st_data_t value, st_hash_t hash); // st.c - st_table *new_tab; - st_index_t i; +enum ar_each_key_type { + ar_each_key_copy, + ar_each_key_cmp, + ar_each_key_insert, +}; - if (size < RHASH_AR_TABLE_MAX_SIZE) { - return; +static inline int +ar_each_key(ar_table *ar, int max, enum ar_each_key_type type, st_data_t *dst_keys, st_table *new_tab, st_hash_t *hashes) +{ + for (int i = 0; i < max; i++) { + ar_table_pair *pair = &ar->pairs[i]; + + switch (type) { + case ar_each_key_copy: + dst_keys[i] = pair->key; + break; + case ar_each_key_cmp: + if (dst_keys[i] != pair->key) return 1; + break; + case ar_each_key_insert: + if (UNDEF_P(pair->key)) continue; // deleted entry + rb_st_add_direct_with_hash(new_tab, pair->key, pair->val, hashes[i]); + break; + } } - new_tab = st_init_table_with_size(&objhash, size * 2); - - for (i = 0; i < RHASH_AR_TABLE_MAX_BOUND; i++) { - ar_table_pair *pair = RHASH_AR_TABLE_REF(hash, i); - st_add_direct(new_tab, pair->key, pair->val); - } - ar_free_and_clear_table(hash); - RHASH_ST_TABLE_SET(hash, new_tab); - return; + return 0; } static st_table * ar_force_convert_table(VALUE hash, const char *file, int line) { - st_table *new_tab; - if (RHASH_ST_TABLE_P(hash)) { return RHASH_ST_TABLE(hash); } - - if (RHASH_AR_TABLE(hash)) { - unsigned i, bound = RHASH_AR_TABLE_BOUND(hash); - -#if defined(RHASH_CONVERT_TABLE_DEBUG) && RHASH_CONVERT_TABLE_DEBUG - rb_obj_info_dump(hash); - fprintf(stderr, "force_convert: %s:%d\n", file, line); - RB_DEBUG_COUNTER_INC(obj_hash_force_convert); -#endif - - new_tab = st_init_table_with_size(&objhash, RHASH_AR_TABLE_SIZE(hash)); - - for (i = 0; i < bound; i++) { - if (ar_cleared_entry(hash, i)) continue; - - ar_table_pair *pair = RHASH_AR_TABLE_REF(hash, i); - st_add_direct(new_tab, pair->key, pair->val); - } - ar_free_and_clear_table(hash); - } else { - new_tab = st_init_table(&objhash); - } - RHASH_ST_TABLE_SET(hash, new_tab); - - return new_tab; -} + ar_table *ar = RHASH_AR_TABLE(hash); + st_hash_t hashes[RHASH_AR_TABLE_MAX_SIZE]; + unsigned int bound, size; + + // prepare hash values + do { + st_data_t keys[RHASH_AR_TABLE_MAX_SIZE]; + bound = RHASH_AR_TABLE_BOUND(hash); + size = RHASH_AR_TABLE_SIZE(hash); + ar_each_key(ar, bound, ar_each_key_copy, keys, NULL, NULL); + + for (unsigned int i = 0; i < bound; i++) { + // do_hash calls #hash method and it can modify hash object + hashes[i] = UNDEF_P(keys[i]) ? 0 : ar_do_hash(keys[i]); + } -static ar_table * -hash_ar_table(VALUE hash) -{ - if (RHASH_TABLE_NULL_P(hash)) { - ar_alloc_table(hash); + // check if modified + if (UNLIKELY(!RHASH_AR_TABLE_P(hash))) return RHASH_ST_TABLE(hash); + if (UNLIKELY(RHASH_AR_TABLE_BOUND(hash) != bound)) continue; + if (UNLIKELY(ar_each_key(ar, bound, ar_each_key_cmp, keys, NULL, NULL))) continue; + } while (0); + + // make st + st_table tab; + st_table *new_tab = &tab; + st_init_existing_table_with_size(new_tab, &objhash, size); + ar_each_key(ar, bound, ar_each_key_insert, NULL, new_tab, hashes); + hash_ar_free_and_clear_table(hash); + RHASH_ST_TABLE_SET(hash, new_tab); + return RHASH_ST_TABLE(hash); } - return RHASH_AR_TABLE(hash); } static int @@ -867,7 +835,6 @@ ar_add_direct_with_hash(VALUE hash, st_data_t key, st_data_t val, st_hash_t hash else { if (UNLIKELY(bin >= RHASH_AR_TABLE_MAX_BOUND)) { bin = ar_compact_table(hash); - hash_ar_table(hash); } HASH_ASSERT(bin < RHASH_AR_TABLE_MAX_BOUND); @@ -878,6 +845,14 @@ ar_add_direct_with_hash(VALUE hash, st_data_t key, st_data_t val, st_hash_t hash } } +static void +ensure_ar_table(VALUE hash) +{ + if (!RHASH_AR_TABLE_P(hash)) { + rb_raise(rb_eRuntimeError, "hash representation was changed during iteration"); + } +} + static int ar_general_foreach(VALUE hash, st_foreach_check_callback_func *func, st_update_callback_func *replace, st_data_t arg) { @@ -888,7 +863,10 @@ ar_general_foreach(VALUE hash, st_foreach_check_callback_func *func, st_update_c if (ar_cleared_entry(hash, i)) continue; ar_table_pair *pair = RHASH_AR_TABLE_REF(hash, i); - enum st_retval retval = (*func)(pair->key, pair->val, arg, 0); + st_data_t key = (st_data_t)pair->key; + st_data_t val = (st_data_t)pair->val; + enum st_retval retval = (*func)(key, val, arg, 0); + ensure_ar_table(hash); /* pair may be not valid here because of theap */ switch (retval) { @@ -899,14 +877,12 @@ ar_general_foreach(VALUE hash, st_foreach_check_callback_func *func, st_update_c return 0; case ST_REPLACE: if (replace) { - VALUE key = pair->key; - VALUE val = pair->val; retval = (*replace)(&key, &val, arg, TRUE); // TODO: pair should be same as pair before. - ar_table_pair *pair = RHASH_AR_TABLE_REF(hash, i); - pair->key = key; - pair->val = val; + pair = RHASH_AR_TABLE_REF(hash, i); + pair->key = (VALUE)key; + pair->val = (VALUE)val; } break; case ST_DELETE: @@ -946,7 +922,7 @@ ar_foreach(VALUE hash, st_foreach_callback_func *func, st_data_t arg) static int ar_foreach_check(VALUE hash, st_foreach_check_callback_func *func, st_data_t arg, - st_data_t never) + st_data_t never) { if (RHASH_AR_TABLE_SIZE(hash) > 0) { unsigned i, ret = 0, bound = RHASH_AR_TABLE_BOUND(hash); @@ -963,17 +939,18 @@ ar_foreach_check(VALUE hash, st_foreach_check_callback_func *func, st_data_t arg hint = ar_hint(hash, i); retval = (*func)(key, pair->val, arg, 0); + ensure_ar_table(hash); hash_verify(hash); switch (retval) { case ST_CHECK: { - pair = RHASH_AR_TABLE_REF(hash, i); - if (pair->key == never) break; - ret = ar_find_entry_hint(hash, hint, key); - if (ret == RHASH_AR_TABLE_MAX_BOUND) { - retval = (*func)(0, 0, arg, 1); - return 2; - } + pair = RHASH_AR_TABLE_REF(hash, i); + if (pair->key == never) break; + ret = ar_find_entry_hint(hash, hint, key); + if (ret == RHASH_AR_TABLE_MAX_BOUND) { + retval = (*func)(0, 0, arg, 1); + return 2; + } } case ST_CONTINUE: break; @@ -981,11 +958,11 @@ ar_foreach_check(VALUE hash, st_foreach_check_callback_func *func, st_data_t arg case ST_REPLACE: return 0; case ST_DELETE: { - if (!ar_cleared_entry(hash, i)) { - ar_clear_entry(hash, i); - RHASH_AR_TABLE_SIZE_DEC(hash); - } - break; + if (!ar_cleared_entry(hash, i)) { + ar_clear_entry(hash, i); + RHASH_AR_TABLE_SIZE_DEC(hash); + } + break; } } } @@ -1012,7 +989,6 @@ ar_update(VALUE hash, st_data_t key, existing = (bin != RHASH_AR_TABLE_MAX_BOUND) ? TRUE : FALSE; } else { - hash_ar_table(hash); /* allocate ltbl if needed */ existing = FALSE; } @@ -1024,6 +1000,7 @@ ar_update(VALUE hash, st_data_t key, old_key = key; retval = (*func)(&key, &value, arg, existing); /* pair can be invalid here because of theap */ + ensure_ar_table(hash); switch (retval) { case ST_CONTINUE: @@ -1061,8 +1038,6 @@ ar_insert(VALUE hash, st_data_t key, st_data_t value) return -1; } - hash_ar_table(hash); /* prepare ltbl */ - bin = ar_find_entry(hash, hash_value, key); if (bin == RHASH_AR_TABLE_MAX_BOUND) { if (RHASH_AR_TABLE_SIZE(hash) >= RHASH_AR_TABLE_MAX_SIZE) { @@ -1070,7 +1045,6 @@ ar_insert(VALUE hash, st_data_t key, st_data_t value) } else if (bin >= RHASH_AR_TABLE_MAX_BOUND) { bin = ar_compact_table(hash); - hash_ar_table(hash); } HASH_ASSERT(bin < RHASH_AR_TABLE_MAX_BOUND); @@ -1205,44 +1179,16 @@ static ar_table* ar_copy(VALUE hash1, VALUE hash2) { ar_table *old_tab = RHASH_AR_TABLE(hash2); + ar_table *new_tab = RHASH_AR_TABLE(hash1); - if (old_tab != NULL) { - ar_table *new_tab = RHASH_AR_TABLE(hash1); - if (new_tab == NULL) { - new_tab = (ar_table*) rb_transient_heap_alloc(hash1, sizeof(ar_table)); - if (new_tab != NULL) { - RHASH_SET_TRANSIENT_FLAG(hash1); - } - else { - RHASH_UNSET_TRANSIENT_FLAG(hash1); - new_tab = (ar_table*)ruby_xmalloc(sizeof(ar_table)); - } - } - *new_tab = *old_tab; - RHASH(hash1)->ar_hint.word = RHASH(hash2)->ar_hint.word; - RHASH_AR_TABLE_BOUND_SET(hash1, RHASH_AR_TABLE_BOUND(hash2)); - RHASH_AR_TABLE_SIZE_SET(hash1, RHASH_AR_TABLE_SIZE(hash2)); - hash_ar_table_set(hash1, new_tab); - - rb_gc_writebarrier_remember(hash1); - return new_tab; - } - else { - RHASH_AR_TABLE_BOUND_SET(hash1, RHASH_AR_TABLE_BOUND(hash2)); - RHASH_AR_TABLE_SIZE_SET(hash1, RHASH_AR_TABLE_SIZE(hash2)); - - if (RHASH_TRANSIENT_P(hash1)) { - RHASH_UNSET_TRANSIENT_FLAG(hash1); - } - else if (RHASH_AR_TABLE(hash1)) { - ruby_xfree(RHASH_AR_TABLE(hash1)); - } + *new_tab = *old_tab; + RHASH_AR_TABLE(hash1)->ar_hint.word = RHASH_AR_TABLE(hash2)->ar_hint.word; + RHASH_AR_TABLE_BOUND_SET(hash1, RHASH_AR_TABLE_BOUND(hash2)); + RHASH_AR_TABLE_SIZE_SET(hash1, RHASH_AR_TABLE_SIZE(hash2)); - hash_ar_table_set(hash1, NULL); + rb_gc_writebarrier_remember(hash1); - rb_gc_writebarrier_remember(hash1); - return old_tab; - } + return new_tab; } static void @@ -1258,32 +1204,32 @@ ar_clear(VALUE hash) } } -#if USE_TRANSIENT_HEAP -void -rb_hash_transient_heap_evacuate(VALUE hash, int promote) +static void +hash_st_free(VALUE hash) { - if (RHASH_TRANSIENT_P(hash)) { - ar_table *new_tab; - ar_table *old_tab = RHASH_AR_TABLE(hash); + HASH_ASSERT(RHASH_ST_TABLE_P(hash)); - if (UNLIKELY(old_tab == NULL)) { - return; - } - HASH_ASSERT(old_tab != NULL); - if (! promote) { - new_tab = rb_transient_heap_alloc(hash, sizeof(ar_table)); - if (new_tab == NULL) promote = true; - } - if (promote) { - new_tab = ruby_xmalloc(sizeof(ar_table)); - RHASH_UNSET_TRANSIENT_FLAG(hash); - } - *new_tab = *old_tab; - hash_ar_table_set(hash, new_tab); + st_table *tab = RHASH_ST_TABLE(hash); + + xfree(tab->bins); + xfree(tab->entries); +} + +static void +hash_st_free_and_clear_table(VALUE hash) +{ + hash_st_free(hash); + + RHASH_ST_CLEAR(hash); +} + +void +rb_hash_free(VALUE hash) +{ + if (RHASH_ST_TABLE_P(hash)) { + hash_st_free(hash); } - hash_verify(hash); } -#endif typedef int st_foreach_func(st_data_t, st_data_t, st_data_t); @@ -1302,7 +1248,7 @@ foreach_safe_i(st_data_t key, st_data_t value, st_data_t args, int error) if (error) return ST_STOP; status = (*arg->func)(key, value, arg->arg); if (status == ST_CONTINUE) { - return ST_CHECK; + return ST_CHECK; } return status; } @@ -1316,7 +1262,7 @@ st_foreach_safe(st_table *table, st_foreach_func *func, st_data_t a) arg.func = (st_foreach_func *)func; arg.arg = a; if (st_foreach_check(table, foreach_safe_i, (st_data_t)&arg, 0)) { - rb_raise(rb_eRuntimeError, "hash modified during iteration"); + rb_raise(rb_eRuntimeError, "hash modified during iteration"); } } @@ -1329,15 +1275,8 @@ struct hash_foreach_arg { }; static int -hash_ar_foreach_iter(st_data_t key, st_data_t value, st_data_t argp, int error) +hash_iter_status_check(int status) { - struct hash_foreach_arg *arg = (struct hash_foreach_arg *)argp; - int status; - - if (error) return ST_STOP; - status = (*arg->func)((VALUE)key, (VALUE)value, arg->arg); - /* TODO: rehash check? rb_raise(rb_eRuntimeError, "rehash occurred during iteration"); */ - switch (status) { case ST_DELETE: return ST_DELETE; @@ -1346,99 +1285,113 @@ hash_ar_foreach_iter(st_data_t key, st_data_t value, st_data_t argp, int error) case ST_STOP: return ST_STOP; } + return ST_CHECK; } static int +hash_ar_foreach_iter(st_data_t key, st_data_t value, st_data_t argp, int error) +{ + struct hash_foreach_arg *arg = (struct hash_foreach_arg *)argp; + + if (error) return ST_STOP; + + int status = (*arg->func)((VALUE)key, (VALUE)value, arg->arg); + /* TODO: rehash check? rb_raise(rb_eRuntimeError, "rehash occurred during iteration"); */ + + return hash_iter_status_check(status); +} + +static int hash_foreach_iter(st_data_t key, st_data_t value, st_data_t argp, int error) { struct hash_foreach_arg *arg = (struct hash_foreach_arg *)argp; - int status; - st_table *tbl; if (error) return ST_STOP; - tbl = RHASH_ST_TABLE(arg->hash); - status = (*arg->func)((VALUE)key, (VALUE)value, arg->arg); + + st_table *tbl = RHASH_ST_TABLE(arg->hash); + int status = (*arg->func)((VALUE)key, (VALUE)value, arg->arg); + if (RHASH_ST_TABLE(arg->hash) != tbl) { - rb_raise(rb_eRuntimeError, "rehash occurred during iteration"); - } - switch (status) { - case ST_DELETE: - return ST_DELETE; - case ST_CONTINUE: - break; - case ST_STOP: - return ST_STOP; + rb_raise(rb_eRuntimeError, "rehash occurred during iteration"); } - return ST_CHECK; + + return hash_iter_status_check(status); } -static int +static unsigned long iter_lev_in_ivar(VALUE hash) { VALUE levval = rb_ivar_get(hash, id_hash_iter_lev); HASH_ASSERT(FIXNUM_P(levval)); - return FIX2INT(levval); + long lev = FIX2LONG(levval); + HASH_ASSERT(lev >= 0); + return (unsigned long)lev; } void rb_ivar_set_internal(VALUE obj, ID id, VALUE val); static void -iter_lev_in_ivar_set(VALUE hash, int lev) +iter_lev_in_ivar_set(VALUE hash, unsigned long lev) { - rb_ivar_set_internal(hash, id_hash_iter_lev, INT2FIX(lev)); + HASH_ASSERT(lev >= RHASH_LEV_MAX); + HASH_ASSERT(POSFIXABLE(lev)); /* POSFIXABLE means fitting to long */ + rb_ivar_set_internal(hash, id_hash_iter_lev, LONG2FIX((long)lev)); } -static int +static inline unsigned long iter_lev_in_flags(VALUE hash) { - unsigned int u = (unsigned int)((RBASIC(hash)->flags >> RHASH_LEV_SHIFT) & RHASH_LEV_MAX); - return (int)u; + return (unsigned long)((RBASIC(hash)->flags >> RHASH_LEV_SHIFT) & RHASH_LEV_MAX); } -static int -RHASH_ITER_LEV(VALUE hash) +static inline void +iter_lev_in_flags_set(VALUE hash, unsigned long lev) { - int lev = iter_lev_in_flags(hash); + HASH_ASSERT(lev <= RHASH_LEV_MAX); + RBASIC(hash)->flags = ((RBASIC(hash)->flags & ~RHASH_LEV_MASK) | ((VALUE)lev << RHASH_LEV_SHIFT)); +} - if (lev == RHASH_LEV_MAX) { - return iter_lev_in_ivar(hash); - } - else { - return lev; - } +static inline bool +hash_iterating_p(VALUE hash) +{ + return iter_lev_in_flags(hash) > 0; } static void hash_iter_lev_inc(VALUE hash) { - int lev = iter_lev_in_flags(hash); + unsigned long lev = iter_lev_in_flags(hash); if (lev == RHASH_LEV_MAX) { - lev = iter_lev_in_ivar(hash); - iter_lev_in_ivar_set(hash, lev+1); + lev = iter_lev_in_ivar(hash) + 1; + if (!POSFIXABLE(lev)) { /* paranoiac check */ + rb_raise(rb_eRuntimeError, "too much nested iterations"); + } } else { lev += 1; - RBASIC(hash)->flags = ((RBASIC(hash)->flags & ~RHASH_LEV_MASK) | ((VALUE)lev << RHASH_LEV_SHIFT)); - if (lev == RHASH_LEV_MAX) { - iter_lev_in_ivar_set(hash, lev); - } + iter_lev_in_flags_set(hash, lev); + if (lev < RHASH_LEV_MAX) return; } + iter_lev_in_ivar_set(hash, lev); } static void hash_iter_lev_dec(VALUE hash) { - int lev = iter_lev_in_flags(hash); + unsigned long lev = iter_lev_in_flags(hash); if (lev == RHASH_LEV_MAX) { lev = iter_lev_in_ivar(hash); - HASH_ASSERT(lev > 0); - iter_lev_in_ivar_set(hash, lev-1); + if (lev > RHASH_LEV_MAX) { + iter_lev_in_ivar_set(hash, lev-1); + return; + } + rb_attr_delete(hash, id_hash_iter_lev); } - else { - HASH_ASSERT(lev > 0); - RBASIC(hash)->flags = ((RBASIC(hash)->flags & ~RHASH_LEV_MASK) | ((lev-1) << RHASH_LEV_SHIFT)); + else if (lev == 0) { + rb_raise(rb_eRuntimeError, "iteration level underflow"); } + iter_lev_in_flags_set(hash, lev - 1); } static VALUE @@ -1516,11 +1469,23 @@ rb_hash_foreach(VALUE hash, rb_foreach_func *func, VALUE farg) hash_verify(hash); } +void rb_st_compact_table(st_table *tab); + +static void +compact_after_delete(VALUE hash) +{ + if (!hash_iterating_p(hash) && RHASH_ST_TABLE_P(hash)) { + rb_st_compact_table(RHASH_ST_TABLE(hash)); + } +} + static VALUE -hash_alloc_flags(VALUE klass, VALUE flags, VALUE ifnone) +hash_alloc_flags(VALUE klass, VALUE flags, VALUE ifnone, bool st) { const VALUE wb = (RGENGC_WB_PROTECTED_HASH ? FL_WB_PROTECTED : 0); - NEWOBJ_OF(hash, struct RHash, klass, T_HASH | wb | flags); + const size_t size = sizeof(struct RHash) + (st ? sizeof(st_table) : sizeof(ar_table)); + + NEWOBJ_OF(hash, struct RHash, klass, T_HASH | wb | flags, size, 0); RHASH_SET_IFNONE((VALUE)hash, ifnone); @@ -1530,7 +1495,8 @@ hash_alloc_flags(VALUE klass, VALUE flags, VALUE ifnone) static VALUE hash_alloc(VALUE klass) { - return hash_alloc_flags(klass, 0, Qnil); + /* Allocate to be able to fit both st_table and ar_table. */ + return hash_alloc_flags(klass, 0, Qnil, sizeof(st_table) > sizeof(ar_table)); } static VALUE @@ -1556,30 +1522,54 @@ copy_compare_by_id(VALUE hash, VALUE basis) return hash; } -MJIT_FUNC_EXPORTED VALUE +VALUE rb_hash_new_with_size(st_index_t size) { - VALUE ret = rb_hash_new(); - if (size == 0) { - /* do nothing */ - } - else if (size <= RHASH_AR_TABLE_MAX_SIZE) { - ar_alloc_table(ret); - } - else { - RHASH_ST_TABLE_SET(ret, st_init_table_with_size(&objhash, size)); + bool st = size > RHASH_AR_TABLE_MAX_SIZE; + VALUE ret = hash_alloc_flags(rb_cHash, 0, Qnil, st); + + if (st) { + hash_st_table_init(ret, &objhash, size); } + return ret; } +VALUE +rb_hash_new_capa(long capa) +{ + return rb_hash_new_with_size((st_index_t)capa); +} + static VALUE hash_copy(VALUE ret, VALUE hash) { - if (!RHASH_EMPTY_P(hash)) { - if (RHASH_AR_TABLE_P(hash)) + if (RHASH_AR_TABLE_P(hash)) { + if (RHASH_AR_TABLE_P(ret)) { ar_copy(ret, hash); - else if (RHASH_ST_TABLE_P(hash)) - RHASH_ST_TABLE_SET(ret, st_copy(RHASH_ST_TABLE(hash))); + } + else { + st_table *tab = RHASH_ST_TABLE(ret); + st_init_existing_table_with_size(tab, &objhash, RHASH_AR_TABLE_SIZE(hash)); + + int bound = RHASH_AR_TABLE_BOUND(hash); + for (int i = 0; i < bound; i++) { + if (ar_cleared_entry(hash, i)) continue; + + ar_table_pair *pair = RHASH_AR_TABLE_REF(hash, i); + st_add_direct(tab, pair->key, pair->val); + RB_OBJ_WRITTEN(ret, Qundef, pair->key); + RB_OBJ_WRITTEN(ret, Qundef, pair->val); + } + } + } + else { + HASH_ASSERT(sizeof(st_table) <= sizeof(ar_table)); + + RHASH_SET_ST_FLAG(ret); + st_replace(RHASH_ST_TABLE(ret), RHASH_ST_TABLE(hash)); + + rb_gc_writebarrier_remember(ret); } return ret; } @@ -1587,13 +1577,21 @@ hash_copy(VALUE ret, VALUE hash) static VALUE hash_dup_with_compare_by_id(VALUE hash) { - return hash_copy(copy_compare_by_id(rb_hash_new(), hash), hash); + VALUE dup = hash_alloc_flags(rb_cHash, 0, Qnil, RHASH_ST_TABLE_P(hash)); + if (RHASH_ST_TABLE_P(hash)) { + RHASH_SET_ST_FLAG(dup); + } + else { + RHASH_UNSET_ST_FLAG(dup); + } + + return hash_copy(dup, hash); } static VALUE hash_dup(VALUE hash, VALUE klass, VALUE flags) { - return hash_copy(hash_alloc_flags(klass, flags, RHASH_IFNONE(hash)), + return hash_copy(hash_alloc_flags(klass, flags, RHASH_IFNONE(hash), !RHASH_EMPTY_P(hash) && RHASH_ST_TABLE_P(hash)), hash); } @@ -1608,7 +1606,7 @@ rb_hash_dup(VALUE hash) return ret; } -MJIT_FUNC_EXPORTED VALUE +VALUE rb_hash_resurrect(VALUE hash) { VALUE ret = hash_dup(hash, rb_cHash, 0); @@ -1621,7 +1619,7 @@ rb_hash_modify_check(VALUE hash) rb_check_frozen(hash); } -MJIT_FUNC_EXPORTED struct st_table * +struct st_table * rb_hash_tbl_raw(VALUE hash, const char *file, int line) { return ar_force_convert_table(hash, file, line); @@ -1670,6 +1668,8 @@ struct update_arg { st_data_t arg; st_update_callback_func *func; VALUE hash; + VALUE key; + VALUE value; }; typedef int (*tbl_update_func)(st_data_t *, st_data_t *, st_data_t, int); @@ -1680,7 +1680,7 @@ rb_hash_stlike_update(VALUE hash, st_data_t key, st_update_callback_func *func, if (RHASH_AR_TABLE_P(hash)) { int result = ar_update(hash, key, func, arg); if (result == -1) { - ar_try_convert_table(hash); + ar_force_convert_table(hash, __FILE__, __LINE__); } else { return result; @@ -1702,11 +1702,11 @@ tbl_update_modify(st_data_t *key, st_data_t *val, st_data_t arg, int existing) default: break; case ST_CONTINUE: - if (!existing || *key != old_key || *val != old_value) + if (!existing || *key != old_key || *val != old_value) { rb_hash_modify(hash); - /* write barrier */ - RB_OBJ_WRITTEN(hash, Qundef, *key); - RB_OBJ_WRITTEN(hash, Qundef, *val); + p->key = *key; + p->value = *val; + } break; case ST_DELETE: if (existing) @@ -1724,30 +1724,38 @@ tbl_update(VALUE hash, VALUE key, tbl_update_func func, st_data_t optional_arg) .arg = optional_arg, .func = func, .hash = hash, + .key = key, + .value = (VALUE)optional_arg, }; - return rb_hash_stlike_update(hash, key, tbl_update_modify, (st_data_t)&arg); + int ret = rb_hash_stlike_update(hash, key, tbl_update_modify, (st_data_t)&arg); + + /* write barrier */ + RB_OBJ_WRITTEN(hash, Qundef, arg.key); + RB_OBJ_WRITTEN(hash, Qundef, arg.value); + + return ret; } -#define UPDATE_CALLBACK(iter_lev, func) ((iter_lev) > 0 ? func##_noinsert : func##_insert) +#define UPDATE_CALLBACK(iter_p, func) ((iter_p) ? func##_noinsert : func##_insert) -#define RHASH_UPDATE_ITER(h, iter_lev, key, func, a) do { \ - tbl_update((h), (key), UPDATE_CALLBACK((iter_lev), func), (st_data_t)(a)); \ +#define RHASH_UPDATE_ITER(h, iter_p, key, func, a) do { \ + tbl_update((h), (key), UPDATE_CALLBACK(iter_p, func), (st_data_t)(a)); \ } while (0) #define RHASH_UPDATE(hash, key, func, arg) \ - RHASH_UPDATE_ITER(hash, RHASH_ITER_LEV(hash), key, func, arg) + RHASH_UPDATE_ITER(hash, hash_iterating_p(hash), key, func, arg) static void set_proc_default(VALUE hash, VALUE proc) { if (rb_proc_lambda_p(proc)) { - int n = rb_proc_arity(proc); + int n = rb_proc_arity(proc); - if (n != 2 && (n >= 0 || n < -3)) { - if (n < 0) n = -n-1; - rb_raise(rb_eTypeError, "default_proc takes two arguments (2 for %d)", n); - } + if (n != 2 && (n >= 0 || n < -3)) { + if (n < 0) n = -n-1; + rb_raise(rb_eTypeError, "default_proc takes two arguments (2 for %d)", n); + } } FL_SET_RAW(hash, RHASH_PROC_DEFAULT); @@ -1759,10 +1767,10 @@ set_proc_default(VALUE hash, VALUE proc) * Hash.new(default_value = nil) -> new_hash * Hash.new {|hash, key| ... } -> new_hash * - * Returns a new empty \Hash object. + * Returns a new empty +Hash+ object. * * The initial default value and initial default proc for the new hash - * depend on which form above was used. See {Default Values}[#class-Hash-label-Default+Values]. + * depend on which form above was used. See {Default Values}[rdoc-ref:Hash@Default+Values]. * * If neither an argument nor a block given, * initializes both the default value and the default proc to <tt>nil</tt>: @@ -1788,23 +1796,29 @@ set_proc_default(VALUE hash, VALUE proc) static VALUE rb_hash_initialize(int argc, VALUE *argv, VALUE hash) { - VALUE ifnone; - rb_hash_modify(hash); + if (rb_block_given_p()) { - rb_check_arity(argc, 0, 0); - ifnone = rb_block_proc(); - SET_PROC_DEFAULT(hash, ifnone); + rb_check_arity(argc, 0, 0); + SET_PROC_DEFAULT(hash, rb_block_proc()); } else { - rb_check_arity(argc, 0, 1); - ifnone = argc == 0 ? Qnil : argv[0]; - RHASH_SET_IFNONE(hash, ifnone); + rb_check_arity(argc, 0, 1); + + VALUE options, ifnone; + rb_scan_args(argc, argv, "01:", &ifnone, &options); + if (NIL_P(ifnone) && !NIL_P(options)) { + ifnone = options; + rb_warn_deprecated_to_remove("3.4", "Calling Hash.new with keyword arguments", "Hash.new({ key: value })"); + } + RHASH_SET_IFNONE(hash, ifnone); } return hash; } +static VALUE rb_hash_to_a(VALUE hash); + /* * call-seq: * Hash[] -> new_empty_hash @@ -1812,26 +1826,26 @@ rb_hash_initialize(int argc, VALUE *argv, VALUE hash) * Hash[ [*2_element_arrays] ] -> new_hash * Hash[*objects] -> new_hash * - * Returns a new \Hash object populated with the given objects, if any. + * Returns a new +Hash+ object populated with the given objects, if any. * See Hash::new. * - * With no argument, returns a new empty \Hash. + * With no argument, returns a new empty +Hash+. * - * When the single given argument is a \Hash, returns a new \Hash - * populated with the entries from the given \Hash, excluding the + * When the single given argument is a +Hash+, returns a new +Hash+ + * populated with the entries from the given +Hash+, excluding the * default value or proc. * * h = {foo: 0, bar: 1, baz: 2} * Hash[h] # => {:foo=>0, :bar=>1, :baz=>2} * - * When the single given argument is an \Array of 2-element Arrays, - * returns a new \Hash object wherein each 2-element array forms a + * When the single given argument is an Array of 2-element Arrays, + * returns a new +Hash+ object wherein each 2-element array forms a * key-value entry: * * Hash[ [ [:foo, 0], [:bar, 1] ] ] # => {:foo=>0, :bar=>1} * * When the argument count is an even number; - * returns a new \Hash object wherein each successive pair of arguments + * returns a new +Hash+ object wherein each successive pair of arguments * has become a key-value entry: * * Hash[:foo, 0, :bar, 1] # => {:foo=>0, :bar=>1} @@ -1847,42 +1861,53 @@ rb_hash_s_create(int argc, VALUE *argv, VALUE klass) if (argc == 1) { tmp = rb_hash_s_try_convert(Qnil, argv[0]); - if (!NIL_P(tmp)) { - hash = hash_alloc(klass); - hash_copy(hash, tmp); - return hash; - } - - tmp = rb_check_array_type(argv[0]); - if (!NIL_P(tmp)) { - long i; - - hash = hash_alloc(klass); - for (i = 0; i < RARRAY_LEN(tmp); ++i) { - VALUE e = RARRAY_AREF(tmp, i); - VALUE v = rb_check_array_type(e); - VALUE key, val = Qnil; - - if (NIL_P(v)) { - rb_raise(rb_eArgError, "wrong element type %s at %ld (expected array)", - rb_builtin_class_name(e), i); - } - switch (RARRAY_LEN(v)) { - default: - rb_raise(rb_eArgError, "invalid number of elements (%ld for 1..2)", - RARRAY_LEN(v)); - case 2: - val = RARRAY_AREF(v, 1); - case 1: - key = RARRAY_AREF(v, 0); - rb_hash_aset(hash, key, val); - } - } - return hash; - } + if (!NIL_P(tmp)) { + if (!RHASH_EMPTY_P(tmp) && rb_hash_compare_by_id_p(tmp)) { + /* hash_copy for non-empty hash will copy compare_by_identity + flag, but we don't want it copied. Work around by + converting hash to flattened array and using that. */ + tmp = rb_hash_to_a(tmp); + } + else { + hash = hash_alloc(klass); + if (!RHASH_EMPTY_P(tmp)) + hash_copy(hash, tmp); + return hash; + } + } + else { + tmp = rb_check_array_type(argv[0]); + } + + if (!NIL_P(tmp)) { + long i; + + hash = hash_alloc(klass); + for (i = 0; i < RARRAY_LEN(tmp); ++i) { + VALUE e = RARRAY_AREF(tmp, i); + VALUE v = rb_check_array_type(e); + VALUE key, val = Qnil; + + if (NIL_P(v)) { + rb_raise(rb_eArgError, "wrong element type %s at %ld (expected array)", + rb_builtin_class_name(e), i); + } + switch (RARRAY_LEN(v)) { + default: + rb_raise(rb_eArgError, "invalid number of elements (%ld for 1..2)", + RARRAY_LEN(v)); + case 2: + val = RARRAY_AREF(v, 1); + case 1: + key = RARRAY_AREF(v, 0); + rb_hash_aset(hash, key, val); + } + } + return hash; + } } if (argc % 2 != 0) { - rb_raise(rb_eArgError, "odd number of arguments for Hash"); + rb_raise(rb_eArgError, "odd number of arguments for Hash"); } hash = hash_alloc(klass); @@ -1891,7 +1916,7 @@ rb_hash_s_create(int argc, VALUE *argv, VALUE klass) return hash; } -MJIT_FUNC_EXPORTED VALUE +VALUE rb_to_hash_type(VALUE hash) { return rb_convert_type_with_id(hash, T_HASH, "Hash", idTo_hash); @@ -1908,14 +1933,14 @@ rb_check_hash_type(VALUE hash) * call-seq: * Hash.try_convert(obj) -> obj, new_hash, or nil * - * If +obj+ is a \Hash object, returns +obj+. + * If +obj+ is a +Hash+ object, returns +obj+. * * Otherwise if +obj+ responds to <tt>:to_hash</tt>, * calls <tt>obj.to_hash</tt> and returns the result. * * Returns +nil+ if +obj+ does not respond to <tt>:to_hash</tt> * - * Raises an exception unless <tt>obj.to_hash</tt> returns a \Hash object. + * Raises an exception unless <tt>obj.to_hash</tt> returns a +Hash+ object. */ static VALUE rb_hash_s_try_convert(VALUE dummy, VALUE hash) @@ -1964,9 +1989,12 @@ static VALUE rb_hash_s_ruby2_keywords_hash(VALUE dummy, VALUE hash) { Check_Type(hash, T_HASH); - hash = rb_hash_dup(hash); - RHASH(hash)->basic.flags |= RHASH_PASS_AS_KEYWORDS; - return hash; + VALUE tmp = rb_hash_dup(hash); + if (RHASH_EMPTY_P(hash) && rb_hash_compare_by_id_p(hash)) { + rb_hash_compare_by_id(tmp); + } + RHASH(tmp)->basic.flags |= RHASH_PASS_AS_KEYWORDS; + return tmp; } struct rehash_arg { @@ -1995,7 +2023,7 @@ rb_hash_rehash_i(VALUE key, VALUE value, VALUE arg) * * The hash table becomes invalid if the hash value of a key * has changed after the entry was created. - * See {Modifying an Active Hash Key}[#class-Hash-label-Modifying+an+Active+Hash+Key]. + * See {Modifying an Active Hash Key}[rdoc-ref:Hash@Modifying+an+Active+Hash+Key]. */ VALUE @@ -2004,25 +2032,27 @@ rb_hash_rehash(VALUE hash) VALUE tmp; st_table *tbl; - if (RHASH_ITER_LEV(hash) > 0) { - rb_raise(rb_eRuntimeError, "rehash during iteration"); + if (hash_iterating_p(hash)) { + rb_raise(rb_eRuntimeError, "rehash during iteration"); } rb_hash_modify_check(hash); if (RHASH_AR_TABLE_P(hash)) { tmp = hash_alloc(0); - ar_alloc_table(tmp); rb_hash_foreach(hash, rb_hash_rehash_i, (VALUE)tmp); - ar_free_and_clear_table(hash); + + hash_ar_free_and_clear_table(hash); ar_copy(hash, tmp); - ar_free_and_clear_table(tmp); } else if (RHASH_ST_TABLE_P(hash)) { st_table *old_tab = RHASH_ST_TABLE(hash); tmp = hash_alloc(0); - tbl = st_init_table_with_size(old_tab->type, old_tab->num_entries); - RHASH_ST_TABLE_SET(tmp, tbl); + + hash_st_table_init(tmp, old_tab->type, old_tab->num_entries); + tbl = RHASH_ST_TABLE(tmp); + rb_hash_foreach(hash, rb_hash_rehash_i, (VALUE)tmp); - st_free_table(old_tab); + + hash_st_free(hash); RHASH_ST_TABLE_SET(hash, tbl); RHASH_ST_CLEAR(tmp); } @@ -2037,17 +2067,31 @@ call_default_proc(VALUE proc, VALUE hash, VALUE key) return rb_proc_call_with_block(proc, 2, args, Qnil); } +static bool +rb_hash_default_unredefined(VALUE hash) +{ + VALUE klass = RBASIC_CLASS(hash); + if (LIKELY(klass == rb_cHash)) { + return !!BASIC_OP_UNREDEFINED_P(BOP_DEFAULT, HASH_REDEFINED_OP_FLAG); + } + else { + return LIKELY(rb_method_basic_definition_p(klass, id_default)); + } +} + VALUE rb_hash_default_value(VALUE hash, VALUE key) { - if (LIKELY(rb_method_basic_definition_p(CLASS_OF(hash), id_default))) { - VALUE ifnone = RHASH_IFNONE(hash); - if (!FL_TEST(hash, RHASH_PROC_DEFAULT)) return ifnone; - if (key == Qundef) return Qnil; + RUBY_ASSERT(RB_TYPE_P(hash, T_HASH)); + + if (LIKELY(rb_hash_default_unredefined(hash))) { + VALUE ifnone = RHASH_IFNONE(hash); + if (LIKELY(!FL_TEST_RAW(hash, RHASH_PROC_DEFAULT))) return ifnone; + if (UNDEF_P(key)) return Qnil; return call_default_proc(ifnone, hash, key); } else { - return rb_funcall(hash, id_default, 1, key); + return rb_funcall(hash, id_default, 1, key); } } @@ -2060,11 +2104,15 @@ hash_stlike_lookup(VALUE hash, st_data_t key, st_data_t *pval) return ar_lookup(hash, key, pval); } else { + extern st_index_t rb_iseq_cdhash_hash(VALUE); + RUBY_ASSERT(RHASH_ST_TABLE(hash)->type->hash == rb_any_hash || + RHASH_ST_TABLE(hash)->type->hash == rb_ident_hash || + RHASH_ST_TABLE(hash)->type->hash == rb_iseq_cdhash_hash); return st_lookup(RHASH_ST_TABLE(hash), key, pval); } } -MJIT_FUNC_EXPORTED int +int rb_hash_stlike_lookup(VALUE hash, st_data_t key, st_data_t *pval) { return hash_stlike_lookup(hash, key, pval); @@ -2079,7 +2127,7 @@ rb_hash_stlike_lookup(VALUE hash, st_data_t key, st_data_t *pval) * h[:foo] # => 0 * * If +key+ is not found, returns a default value - * (see {Default Values}[#class-Hash-label-Default+Values]): + * (see {Default Values}[rdoc-ref:Hash@Default+Values]): * h = {foo: 0, bar: 1, baz: 2} * h[:nosuch] # => nil */ @@ -2151,7 +2199,7 @@ rb_hash_fetch_m(int argc, VALUE *argv, VALUE hash) block_given = rb_block_given_p(); if (block_given && argc == 2) { - rb_warn("block supersedes default value argument"); + rb_warn("block supersedes default value argument"); } if (hash_stlike_lookup(hash, key, &val)) { @@ -2188,7 +2236,7 @@ rb_hash_fetch(VALUE hash, VALUE key) * * Returns the default value for the given +key+. * The returned value will be determined either by the default proc or by the default value. - * See {Default Values}[#class-Hash-label-Default+Values]. + * See {Default Values}[rdoc-ref:Hash@Default+Values]. * * With no argument, returns the current default value: * h = {} @@ -2209,8 +2257,8 @@ rb_hash_default(int argc, VALUE *argv, VALUE hash) rb_check_arity(argc, 0, 1); ifnone = RHASH_IFNONE(hash); if (FL_TEST(hash, RHASH_PROC_DEFAULT)) { - if (argc == 0) return Qnil; - return call_default_proc(ifnone, hash, argv[0]); + if (argc == 0) return Qnil; + return call_default_proc(ifnone, hash, argv[0]); } return ifnone; } @@ -2225,7 +2273,7 @@ rb_hash_default(int argc, VALUE *argv, VALUE hash) * h.default = false # => false * h.default # => false * - * See {Default Values}[#class-Hash-label-Default+Values]. + * See {Default Values}[rdoc-ref:Hash@Default+Values]. */ static VALUE @@ -2241,7 +2289,7 @@ rb_hash_set_default(VALUE hash, VALUE ifnone) * hash.default_proc -> proc or nil * * Returns the default proc for +self+ - * (see {Default Values}[#class-Hash-label-Default+Values]): + * (see {Default Values}[rdoc-ref:Hash@Default+Values]): * h = {} * h.default_proc # => nil * h.default_proc = proc {|hash, key| "Default value for #{key}" } @@ -2252,7 +2300,7 @@ static VALUE rb_hash_default_proc(VALUE hash) { if (FL_TEST(hash, RHASH_PROC_DEFAULT)) { - return RHASH_IFNONE(hash); + return RHASH_IFNONE(hash); } return Qnil; } @@ -2261,8 +2309,8 @@ rb_hash_default_proc(VALUE hash) * call-seq: * hash.default_proc = proc -> proc * - * Sets the default proc for +self+ to +proc+: - * (see {Default Values}[#class-Hash-label-Default+Values]): + * Sets the default proc for +self+ to +proc+ + * (see {Default Values}[rdoc-ref:Hash@Default+Values]): * h = {} * h.default_proc # => nil * h.default_proc = proc { |hash, key| "Default value for #{key}" } @@ -2278,14 +2326,14 @@ rb_hash_set_default_proc(VALUE hash, VALUE proc) rb_hash_modify_check(hash); if (NIL_P(proc)) { - SET_DEFAULT(hash, proc); - return proc; + SET_DEFAULT(hash, proc); + return proc; } b = rb_check_convert_type_with_id(proc, T_DATA, "Proc", idTo_proc); if (NIL_P(b) || !rb_obj_is_proc(b)) { - rb_raise(rb_eTypeError, - "wrong default_proc type %s (expected Proc)", - rb_obj_classname(proc)); + rb_raise(rb_eTypeError, + "wrong default_proc type %s (expected Proc)", + rb_obj_classname(proc)); } proc = b; SET_PROC_DEFAULT(hash, proc); @@ -2298,8 +2346,8 @@ key_i(VALUE key, VALUE value, VALUE arg) VALUE *args = (VALUE *)arg; if (rb_equal(value, args[0])) { - args[1] = key; - return ST_STOP; + args[1] = key; + return ST_STOP; } return ST_CONTINUE; } @@ -2309,12 +2357,12 @@ key_i(VALUE key, VALUE value, VALUE arg) * hash.key(value) -> key or nil * * Returns the key for the first-found entry with the given +value+ - * (see {Entry Order}[#class-Hash-label-Entry+Order]): + * (see {Entry Order}[rdoc-ref:Hash@Entry+Order]): * h = {foo: 0, bar: 2, baz: 2} * h.key(0) # => :foo * h.key(2) # => :bar * - * Returns +nil+ if so such value is found. + * Returns +nil+ if no such value is found. */ static VALUE @@ -2369,11 +2417,11 @@ rb_hash_delete(VALUE hash, VALUE key) { VALUE deleted_value = rb_hash_delete_entry(hash, key); - if (deleted_value != Qundef) { /* likely pass */ - return deleted_value; + if (!UNDEF_P(deleted_value)) { /* likely pass */ + return deleted_value; } else { - return Qnil; + return Qnil; } } @@ -2412,16 +2460,17 @@ rb_hash_delete_m(VALUE hash, VALUE key) rb_hash_modify_check(hash); val = rb_hash_delete_entry(hash, key); - if (val != Qundef) { - return val; + if (!UNDEF_P(val)) { + compact_after_delete(hash); + return val; } else { - if (rb_block_given_p()) { - return rb_yield(key); - } - else { - return Qnil; - } + if (rb_block_given_p()) { + return rb_yield(key); + } + else { + return Qnil; + } } } @@ -2442,17 +2491,16 @@ shift_i_safe(VALUE key, VALUE value, VALUE arg) /* * call-seq: - * hash.shift -> [key, value] or default_value + * hash.shift -> [key, value] or nil * * Removes the first hash entry - * (see {Entry Order}[#class-Hash-label-Entry+Order]); - * returns a 2-element \Array containing the removed key and value: + * (see {Entry Order}[rdoc-ref:Hash@Entry+Order]); + * returns a 2-element Array containing the removed key and value: * h = {foo: 0, bar: 1, baz: 2} * h.shift # => [:foo, 0] * h # => {:bar=>1, :baz=>2} * - * Returns the default value if the hash is empty - * (see {Default Values}[#class-Hash-label-Default+Values]). + * Returns nil if the hash is empty. */ static VALUE @@ -2462,15 +2510,15 @@ rb_hash_shift(VALUE hash) rb_hash_modify_check(hash); if (RHASH_AR_TABLE_P(hash)) { - var.key = Qundef; - if (RHASH_ITER_LEV(hash) == 0) { + var.key = Qundef; + if (!hash_iterating_p(hash)) { if (ar_shift(hash, &var.key, &var.val)) { - return rb_assoc_new(var.key, var.val); - } - } - else { + return rb_assoc_new(var.key, var.val); + } + } + else { rb_hash_foreach(hash, shift_i_safe, (VALUE)&var); - if (var.key != Qundef) { + if (!UNDEF_P(var.key)) { rb_hash_delete_entry(hash, var.key); return rb_assoc_new(var.key, var.val); } @@ -2478,28 +2526,28 @@ rb_hash_shift(VALUE hash) } if (RHASH_ST_TABLE_P(hash)) { var.key = Qundef; - if (RHASH_ITER_LEV(hash) == 0) { + if (!hash_iterating_p(hash)) { if (st_shift(RHASH_ST_TABLE(hash), &var.key, &var.val)) { return rb_assoc_new(var.key, var.val); } } else { - rb_hash_foreach(hash, shift_i_safe, (VALUE)&var); - if (var.key != Qundef) { - rb_hash_delete_entry(hash, var.key); - return rb_assoc_new(var.key, var.val); - } - } + rb_hash_foreach(hash, shift_i_safe, (VALUE)&var); + if (!UNDEF_P(var.key)) { + rb_hash_delete_entry(hash, var.key); + return rb_assoc_new(var.key, var.val); + } + } } - return rb_hash_default_value(hash, Qnil); + return Qnil; } static int delete_if_i(VALUE key, VALUE value, VALUE hash) { if (RTEST(rb_yield_values(2, key, value))) { - rb_hash_modify(hash); - return ST_DELETE; + rb_hash_modify(hash); + return ST_DELETE; } return ST_CONTINUE; } @@ -2521,7 +2569,7 @@ hash_enum_size(VALUE hash, VALUE args, VALUE eobj) * h = {foo: 0, bar: 1, baz: 2} * h.delete_if {|key, value| value > 0 } # => {:foo=>0} * - * If no block given, returns a new \Enumerator: + * If no block given, returns a new Enumerator: * h = {foo: 0, bar: 1, baz: 2} * e = h.delete_if # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:delete_if> * e.each { |key, value| value > 0 } # => {:foo=>0} @@ -2534,6 +2582,7 @@ rb_hash_delete_if(VALUE hash) rb_hash_modify_check(hash); if (!RHASH_TABLE_EMPTY_P(hash)) { rb_hash_foreach(hash, delete_if_i, hash); + compact_after_delete(hash); } return hash; } @@ -2550,7 +2599,7 @@ rb_hash_delete_if(VALUE hash) * * Returns +nil+ if no entries are removed. * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.reject! # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:reject!> * e.each {|key, value| key.start_with?('b') } # => {:foo=>0} @@ -2575,13 +2624,13 @@ rb_hash_reject_bang(VALUE hash) * hash.reject {|key, value| ... } -> new_hash * hash.reject -> new_enumerator * - * Returns a new \Hash object whose entries are all those + * Returns a new +Hash+ object whose entries are all those * from +self+ for which the block returns +false+ or +nil+: * h = {foo: 0, bar: 1, baz: 2} * h1 = h.reject {|key, value| key.start_with?('b') } * h1 # => {:foo=>0} * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.reject # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:reject> * h1 = e.each {|key, value| key.start_with?('b') } @@ -2596,7 +2645,8 @@ rb_hash_reject(VALUE hash) RETURN_SIZED_ENUMERATOR(hash, 0, 0, hash_enum_size); result = hash_dup_with_compare_by_id(hash); if (!RHASH_EMPTY_P(hash)) { - rb_hash_foreach(result, delete_if_i, result); + rb_hash_foreach(result, delete_if_i, result); + compact_after_delete(result); } return result; } @@ -2605,7 +2655,7 @@ rb_hash_reject(VALUE hash) * call-seq: * hash.slice(*keys) -> new_hash * - * Returns a new \Hash object containing the entries for the given +keys+: + * Returns a new +Hash+ object containing the entries for the given +keys+: * h = {foo: 0, bar: 1, baz: 2} * h.slice(:baz, :foo) # => {:baz=>2, :foo=>0} * @@ -2624,10 +2674,10 @@ rb_hash_slice(int argc, VALUE *argv, VALUE hash) result = copy_compare_by_id(rb_hash_new_with_size(argc), hash); for (i = 0; i < argc; i++) { - key = argv[i]; - value = rb_hash_lookup2(hash, key, Qundef); - if (value != Qundef) - rb_hash_aset(result, key, value); + key = argv[i]; + value = rb_hash_lookup2(hash, key, Qundef); + if (!UNDEF_P(value)) + rb_hash_aset(result, key, value); } return result; @@ -2637,7 +2687,7 @@ rb_hash_slice(int argc, VALUE *argv, VALUE hash) * call-seq: * hsh.except(*keys) -> a_hash * - * Returns a new \Hash excluding entries for the given +keys+: + * Returns a new +Hash+ excluding entries for the given +keys+: * h = { a: 100, b: 200, c: 300 } * h.except(:a) #=> {:b=>200, :c=>300} * @@ -2656,6 +2706,7 @@ rb_hash_except(int argc, VALUE *argv, VALUE hash) key = argv[i]; rb_hash_delete(result, key); } + compact_after_delete(result); return result; } @@ -2664,11 +2715,11 @@ rb_hash_except(int argc, VALUE *argv, VALUE hash) * call-seq: * hash.values_at(*keys) -> new_array * - * Returns a new \Array containing values for the given +keys+: + * Returns a new Array containing values for the given +keys+: * h = {foo: 0, bar: 1, baz: 2} * h.values_at(:baz, :foo) # => [2, 0] * - * The {default values}[#class-Hash-label-Default+Values] are returned + * The {default values}[rdoc-ref:Hash@Default+Values] are returned * for any keys that are not found: * h.values_at(:hello, :foo) # => [nil, 0] */ @@ -2680,7 +2731,7 @@ rb_hash_values_at(int argc, VALUE *argv, VALUE hash) long i; for (i=0; i<argc; i++) { - rb_ary_push(result, rb_hash_aref(hash, argv[i])); + rb_ary_push(result, rb_hash_aref(hash, argv[i])); } return result; } @@ -2690,11 +2741,11 @@ rb_hash_values_at(int argc, VALUE *argv, VALUE hash) * hash.fetch_values(*keys) -> new_array * hash.fetch_values(*keys) {|key| ... } -> new_array * - * Returns a new \Array containing the values associated with the given keys *keys: + * Returns a new Array containing the values associated with the given keys *keys: * h = {foo: 0, bar: 1, baz: 2} * h.fetch_values(:baz, :foo) # => [2, 0] * - * Returns a new empty \Array if no arguments given. + * Returns a new empty Array if no arguments given. * * When a block is given, calls the block with each missing key, * treating the block's return value as the value for that key: @@ -2712,7 +2763,7 @@ rb_hash_fetch_values(int argc, VALUE *argv, VALUE hash) long i; for (i=0; i<argc; i++) { - rb_ary_push(result, rb_hash_fetch(hash, argv[i])); + rb_ary_push(result, rb_hash_fetch(hash, argv[i])); } return result; } @@ -2721,8 +2772,8 @@ static int keep_if_i(VALUE key, VALUE value, VALUE hash) { if (!RTEST(rb_yield_values(2, key, value))) { - rb_hash_modify(hash); - return ST_DELETE; + rb_hash_modify(hash); + return ST_DELETE; } return ST_CONTINUE; } @@ -2732,13 +2783,11 @@ keep_if_i(VALUE key, VALUE value, VALUE hash) * hash.select {|key, value| ... } -> new_hash * hash.select -> new_enumerator * - * Hash#filter is an alias for Hash#select. - * - * Returns a new \Hash object whose entries are those for which the block returns a truthy value: + * Returns a new +Hash+ object whose entries are those for which the block returns a truthy value: * h = {foo: 0, bar: 1, baz: 2} * h.select {|key, value| value < 2 } # => {:foo=>0, :bar=>1} * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.select # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:select> * e.each {|key, value| value < 2 } # => {:foo=>0, :bar=>1} @@ -2752,7 +2801,8 @@ rb_hash_select(VALUE hash) RETURN_SIZED_ENUMERATOR(hash, 0, 0, hash_enum_size); result = hash_dup_with_compare_by_id(hash); if (!RHASH_EMPTY_P(hash)) { - rb_hash_foreach(result, keep_if_i, result); + rb_hash_foreach(result, keep_if_i, result); + compact_after_delete(result); } return result; } @@ -2762,15 +2812,13 @@ rb_hash_select(VALUE hash) * hash.select! {|key, value| ... } -> self or nil * hash.select! -> new_enumerator * - * Hash#filter! is an alias for Hash#select!. - * * Returns +self+, whose entries are those for which the block returns a truthy value: * h = {foo: 0, bar: 1, baz: 2} * h.select! {|key, value| value < 2 } => {:foo=>0, :bar=>1} * * Returns +nil+ if no entries were removed. * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.select! # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:select!> * e.each { |key, value| value < 2 } # => {:foo=>0, :bar=>1} @@ -2801,7 +2849,7 @@ rb_hash_select_bang(VALUE hash) * h = {foo: 0, bar: 1, baz: 2} * h.keep_if { |key, value| key.start_with?('b') } # => {:bar=>1, :baz=>2} * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.keep_if # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:keep_if> * e.each { |key, value| key.start_with?('b') } # => {:bar=>1, :baz=>2} @@ -2836,7 +2884,7 @@ rb_hash_clear(VALUE hash) { rb_hash_modify_check(hash); - if (RHASH_ITER_LEV(hash) > 0) { + if (hash_iterating_p(hash)) { rb_hash_foreach(hash, clear_i, 0); } else if (RHASH_AR_TABLE_P(hash)) { @@ -2844,6 +2892,7 @@ rb_hash_clear(VALUE hash) } else { st_clear(RHASH_ST_TABLE(hash)); + compact_after_delete(hash); } return hash; @@ -2863,7 +2912,7 @@ rb_hash_key_str(VALUE key) return rb_fstring(key); } else { - return rb_str_new_frozen(key); + return rb_str_new_frozen(key); } } @@ -2871,7 +2920,7 @@ static int hash_aset_str(st_data_t *key, st_data_t *val, struct update_arg *arg, int existing) { if (!existing && !RB_OBJ_FROZEN(*key)) { - *key = rb_hash_key_str(*key); + *key = rb_hash_key_str(*key); } return hash_aset(key, val, arg, existing); } @@ -2884,13 +2933,11 @@ NOINSERT_UPDATE_CALLBACK(hash_aset_str) * hash[key] = value -> value * hash.store(key, value) * - * Hash#store is an alias for Hash#[]=. - * Associates the given +value+ with the given +key+; returns +value+. * * If the given +key+ exists, replaces its value with the given +value+; * the ordering is not affected - * (see {Entry Order}[#class-Hash-label-Entry+Order]): + * (see {Entry Order}[rdoc-ref:Hash@Entry+Order]): * h = {foo: 0, bar: 1} * h[:foo] = 2 # => 2 * h.store(:bar, 3) # => 3 @@ -2898,7 +2945,7 @@ NOINSERT_UPDATE_CALLBACK(hash_aset_str) * * If +key+ does not exist, adds the +key+ and +value+; * the new entry is last in the order - * (see {Entry Order}[#class-Hash-label-Entry+Order]): + * (see {Entry Order}[rdoc-ref:Hash@Entry+Order]): * h = {foo: 0, bar: 1} * h[:baz] = 2 # => 2 * h.store(:bat, 3) # => 3 @@ -2908,20 +2955,15 @@ NOINSERT_UPDATE_CALLBACK(hash_aset_str) VALUE rb_hash_aset(VALUE hash, VALUE key, VALUE val) { - int iter_lev = RHASH_ITER_LEV(hash); + bool iter_p = hash_iterating_p(hash); rb_hash_modify(hash); - if (RHASH_TABLE_NULL_P(hash)) { - if (iter_lev > 0) no_new_key(); - ar_alloc_table(hash); - } - - if (RHASH_TYPE(hash) == &identhash || rb_obj_class(key) != rb_cString) { - RHASH_UPDATE_ITER(hash, iter_lev, key, hash_aset, val); + if (!RHASH_STRING_KEY_P(hash, key)) { + RHASH_UPDATE_ITER(hash, iter_p, key, hash_aset, val); } else { - RHASH_UPDATE_ITER(hash, iter_lev, key, hash_aset_str, val); + RHASH_UPDATE_ITER(hash, iter_p, key, hash_aset_str, val); } return val; } @@ -2941,7 +2983,7 @@ rb_hash_replace(VALUE hash, VALUE hash2) { rb_hash_modify_check(hash); if (hash == hash2) return hash; - if (RHASH_ITER_LEV(hash) > 0) { + if (hash_iterating_p(hash)) { rb_raise(rb_eRuntimeError, "can't replace hash during iteration"); } hash2 = to_hash(hash2); @@ -2949,19 +2991,13 @@ rb_hash_replace(VALUE hash, VALUE hash2) COPY_DEFAULT(hash, hash2); if (RHASH_AR_TABLE_P(hash)) { - ar_free_and_clear_table(hash); + hash_ar_free_and_clear_table(hash); } else { - st_free_table(RHASH_ST_TABLE(hash)); - RHASH_ST_CLEAR(hash); - } - hash_copy(hash, hash2); - if (RHASH_EMPTY_P(hash2) && RHASH_ST_TABLE_P(hash2)) { - /* ident hash */ - RHASH_ST_TABLE_SET(hash, st_init_table_with_size(RHASH_TYPE(hash2), 0)); + hash_st_free_and_clear_table(hash); } - rb_gc_writebarrier_remember(hash); + hash_copy(hash, hash2); return hash; } @@ -2972,9 +3008,9 @@ rb_hash_replace(VALUE hash, VALUE hash2) * hash.size -> integer * * Returns the count of entries in +self+: + * * {foo: 0, bar: 1, baz: 2}.length # => 3 * - * Hash#length is an alias for Hash#size. */ VALUE @@ -2998,7 +3034,7 @@ rb_hash_size_num(VALUE hash) * {foo: 0, bar: 1, baz: 2}.empty? # => false */ -static VALUE +VALUE rb_hash_empty_p(VALUE hash) { return RBOOL(RHASH_EMPTY_P(hash)); @@ -3024,7 +3060,7 @@ each_value_i(VALUE key, VALUE value, VALUE _) * 1 * 2 * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.each_value # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:each_value> * h1 = e.each {|value| puts value } @@ -3063,7 +3099,7 @@ each_key_i(VALUE key, VALUE value, VALUE _) * bar * baz * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.each_key # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:each_key> * h1 = e.each {|key| puts key } @@ -3105,8 +3141,6 @@ each_pair_i_fast(VALUE key, VALUE value, VALUE _) * hash.each -> new_enumerator * hash.each_pair -> new_enumerator * - * Hash#each is an alias for Hash#each_pair. - * Calls the given block with each key-value pair; returns +self+: * h = {foo: 0, bar: 1, baz: 2} * h.each_pair {|key, value| puts "#{key}: #{value}"} # => {:foo=>0, :bar=>1, :baz=>2} @@ -3115,7 +3149,7 @@ each_pair_i_fast(VALUE key, VALUE value, VALUE _) * bar: 1 * baz: 2 * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.each_pair # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:each_pair> * h1 = e.each {|key, value| puts "#{key}: #{value}"} @@ -3131,9 +3165,9 @@ rb_hash_each_pair(VALUE hash) { RETURN_SIZED_ENUMERATOR(hash, 0, 0, hash_enum_size); if (rb_block_pair_yield_optimizable()) - rb_hash_foreach(hash, each_pair_i_fast, 0); + rb_hash_foreach(hash, each_pair_i_fast, 0); else - rb_hash_foreach(hash, each_pair_i, 0); + rb_hash_foreach(hash, each_pair_i, 0); return hash; } @@ -3149,7 +3183,7 @@ transform_keys_hash_i(VALUE key, VALUE value, VALUE transarg) struct transform_keys_args *p = (void *)transarg; VALUE trans = p->trans, result = p->result; VALUE new_key = rb_hash_lookup2(trans, key, Qundef); - if (new_key == Qundef) { + if (UNDEF_P(new_key)) { if (p->block_given) new_key = rb_yield(key); else @@ -3174,7 +3208,7 @@ transform_keys_i(VALUE key, VALUE value, VALUE result) * hash.transform_keys(hash2) {|other_key| ...} -> new_hash * hash.transform_keys -> new_enumerator * - * Returns a new \Hash object; each entry has: + * Returns a new +Hash+ object; each entry has: * * A key provided by the block. * * The value from +self+. * @@ -3198,7 +3232,7 @@ transform_keys_i(VALUE key, VALUE value, VALUE result) * h1 = h.transform_keys {|key| :bat } * h1 # => {:bat=>2} * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.transform_keys # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:transform_keys> * h1 = e.each { |key| key.to_s } @@ -3262,7 +3296,7 @@ rb_hash_transform_keys_bang(int argc, VALUE *argv, VALUE hash) if (!RHASH_TABLE_EMPTY_P(hash)) { long i; VALUE new_keys = hash_alloc(0); - VALUE pairs = rb_ary_tmp_new(RHASH_SIZE(hash) * 2); + VALUE pairs = rb_ary_hidden_new(RHASH_SIZE(hash) * 2); rb_hash_foreach(hash, flatten_i, pairs); for (i = 0; i < RARRAY_LEN(pairs); i += 2) { VALUE key = RARRAY_AREF(pairs, i), new_key, val; @@ -3270,7 +3304,7 @@ rb_hash_transform_keys_bang(int argc, VALUE *argv, VALUE hash) if (!trans) { new_key = rb_yield(key); } - else if ((new_key = rb_hash_lookup2(trans, key, Qundef)) != Qundef) { + else if (!UNDEF_P(new_key = rb_hash_lookup2(trans, key, Qundef))) { /* use the transformed key */ } else if (block_given) { @@ -3289,6 +3323,7 @@ rb_hash_transform_keys_bang(int argc, VALUE *argv, VALUE hash) rb_ary_clear(pairs); rb_hash_clear(new_keys); } + compact_after_delete(hash); return hash; } @@ -3313,7 +3348,7 @@ transform_values_foreach_replace(st_data_t *key, st_data_t *value, st_data_t arg * hash.transform_values {|value| ... } -> new_hash * hash.transform_values -> new_enumerator * - * Returns a new \Hash object; each entry has: + * Returns a new +Hash+ object; each entry has: * * A key from +self+. * * A value provided by the block. * @@ -3322,7 +3357,7 @@ transform_values_foreach_replace(st_data_t *key, st_data_t *value, st_data_t arg * h1 = h.transform_values {|value| value * 100} * h1 # => {:foo=>0, :bar=>100, :baz=>200} * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.transform_values # => #<Enumerator: {:foo=>0, :bar=>1, :baz=>2}:transform_values> * h1 = e.each { |value| value * 100} @@ -3339,6 +3374,7 @@ rb_hash_transform_values(VALUE hash) if (!RHASH_EMPTY_P(hash)) { rb_hash_stlike_foreach_with_replace(result, transform_values_foreach_func, transform_values_foreach_replace, result); + compact_after_delete(result); } return result; @@ -3353,7 +3389,7 @@ rb_hash_transform_values(VALUE hash) * h = {foo: 0, bar: 1, baz: 2} * h.transform_values! {|value| value * 100} # => {:foo=>0, :bar=>100, :baz=>200} * - * Returns a new \Enumerator if no block given: + * Returns a new Enumerator if no block given: * h = {foo: 0, bar: 1, baz: 2} * e = h.transform_values! # => #<Enumerator: {:foo=>0, :bar=>100, :baz=>200}:transform_values!> * h1 = e.each {|value| value * 100} @@ -3383,8 +3419,8 @@ to_a_i(VALUE key, VALUE value, VALUE ary) * call-seq: * hash.to_a -> new_array * - * Returns a new \Array of 2-element \Array objects; - * each nested \Array contains a key-value pair from +self+: + * Returns a new Array of 2-element Array objects; + * each nested Array contains a key-value pair from +self+: * h = {foo: 0, bar: 1, baz: 2} * h.to_a # => [[:foo, 0], [:bar, 1], [:baz, 2]] */ @@ -3407,10 +3443,10 @@ inspect_i(VALUE key, VALUE value, VALUE str) str2 = rb_inspect(key); if (RSTRING_LEN(str) > 1) { - rb_str_buf_cat_ascii(str, ", "); + rb_str_buf_cat_ascii(str, ", "); } else { - rb_enc_copy(str, str2); + rb_enc_copy(str, str2); } rb_str_buf_append(str, str2); rb_str_buf_cat_ascii(str, "=>"); @@ -3437,18 +3473,18 @@ inspect_hash(VALUE hash, VALUE dummy, int recur) * call-seq: * hash.inspect -> new_string * - * Returns a new \String containing the hash entries: + * Returns a new String containing the hash entries: + * h = {foo: 0, bar: 1, baz: 2} * h.inspect # => "{:foo=>0, :bar=>1, :baz=>2}" * - * Hash#to_s is an alias for Hash#inspect. */ static VALUE rb_hash_inspect(VALUE hash) { if (RHASH_EMPTY_P(hash)) - return rb_usascii_str_new2("{}"); + return rb_usascii_str_new2("{}"); return rb_exec_recursive(inspect_hash, hash, 0); } @@ -3502,15 +3538,15 @@ rb_hash_to_h_block(VALUE hash) * hash.to_h -> self or new_hash * hash.to_h {|key, value| ... } -> new_hash * - * For an instance of \Hash, returns +self+. + * For an instance of +Hash+, returns +self+. * - * For a subclass of \Hash, returns a new \Hash + * For a subclass of +Hash+, returns a new +Hash+ * containing the content of +self+. * - * When a block is given, returns a new \Hash object + * When a block is given, returns a new +Hash+ object * whose content is based on the block; - * the block should return a 2-element \Array object - * specifying the key-value pair to be included in the returned \Array: + * the block should return a 2-element Array object + * specifying the key-value pair to be included in the returned Array: * h = {foo: 0, bar: 1, baz: 2} * h1 = h.to_h {|key, value| [value, key] } * h1 # => {0=>:foo, 1=>:bar, 2=>:baz} @@ -3523,7 +3559,7 @@ rb_hash_to_h(VALUE hash) return rb_hash_to_h_block(hash); } if (rb_obj_class(hash) != rb_cHash) { - const VALUE flags = RBASIC(hash)->flags; + const VALUE flags = RBASIC(hash)->flags; hash = hash_dup(hash, rb_cHash, flags & RHASH_PROC_DEFAULT); } return hash; @@ -3540,12 +3576,12 @@ keys_i(VALUE key, VALUE value, VALUE ary) * call-seq: * hash.keys -> new_array * - * Returns a new \Array containing all keys in +self+: + * Returns a new Array containing all keys in +self+: * h = {foo: 0, bar: 1, baz: 2} * h.keys # => [:foo, :bar, :baz] */ -MJIT_FUNC_EXPORTED VALUE +VALUE rb_hash_keys(VALUE hash) { st_index_t size = RHASH_SIZE(hash); @@ -3554,7 +3590,7 @@ rb_hash_keys(VALUE hash) if (size == 0) return keys; if (ST_DATA_COMPATIBLE_P(VALUE)) { - RARRAY_PTR_USE_TRANSIENT(keys, ptr, { + RARRAY_PTR_USE(keys, ptr, { if (RHASH_AR_TABLE_P(hash)) { size = ar_keys(hash, ptr, size); } @@ -3564,10 +3600,10 @@ rb_hash_keys(VALUE hash) } }); rb_gc_writebarrier_remember(keys); - rb_ary_set_len(keys, size); + rb_ary_set_len(keys, size); } else { - rb_hash_foreach(hash, keys_i, keys); + rb_hash_foreach(hash, keys_i, keys); } return keys; @@ -3584,7 +3620,7 @@ values_i(VALUE key, VALUE value, VALUE ary) * call-seq: * hash.values -> new_array * - * Returns a new \Array containing all values in +self+: + * Returns a new Array containing all values in +self+: * h = {foo: 0, bar: 1, baz: 2} * h.values # => [0, 1, 2] */ @@ -3601,22 +3637,22 @@ rb_hash_values(VALUE hash) if (ST_DATA_COMPATIBLE_P(VALUE)) { if (RHASH_AR_TABLE_P(hash)) { rb_gc_writebarrier_remember(values); - RARRAY_PTR_USE_TRANSIENT(values, ptr, { + RARRAY_PTR_USE(values, ptr, { size = ar_values(hash, ptr, size); }); } else if (RHASH_ST_TABLE_P(hash)) { st_table *table = RHASH_ST_TABLE(hash); rb_gc_writebarrier_remember(values); - RARRAY_PTR_USE_TRANSIENT(values, ptr, { + RARRAY_PTR_USE(values, ptr, { size = st_values(table, ptr, size); }); } - rb_ary_set_len(values, size); + rb_ary_set_len(values, size); } else { - rb_hash_foreach(hash, values_i, values); + rb_hash_foreach(hash, values_i, values); } return values; @@ -3628,13 +3664,11 @@ rb_hash_values(VALUE hash) * hash.has_key?(key) -> true or false * hash.key?(key) -> true or false * hash.member?(key) -> true or false - - * Methods #has_key?, #key?, and #member? are aliases for \#include?. * * Returns +true+ if +key+ is a key in +self+, otherwise +false+. */ -MJIT_FUNC_EXPORTED VALUE +VALUE rb_hash_has_key(VALUE hash, VALUE key) { return RBOOL(hash_stlike_lookup(hash, key, NULL)); @@ -3646,8 +3680,8 @@ rb_hash_search_value(VALUE key, VALUE value, VALUE arg) VALUE *data = (VALUE *)arg; if (rb_equal(value, data[1])) { - data[0] = Qtrue; - return ST_STOP; + data[0] = Qtrue; + return ST_STOP; } return ST_CONTINUE; } @@ -3657,8 +3691,6 @@ rb_hash_search_value(VALUE key, VALUE value, VALUE arg) * hash.has_value?(value) -> true or false * hash.value?(value) -> true or false * - * Method #value? is an alias for \#has_value?. - * * Returns +true+ if +value+ is a value in +self+, otherwise +false+. */ @@ -3718,23 +3750,23 @@ hash_equal(VALUE hash1, VALUE hash2, int eql) if (hash1 == hash2) return Qtrue; if (!RB_TYPE_P(hash2, T_HASH)) { - if (!rb_respond_to(hash2, idTo_hash)) { - return Qfalse; - } - if (eql) { - if (rb_eql(hash2, hash1)) { - return Qtrue; - } - else { - return Qfalse; - } - } - else { - return rb_equal(hash2, hash1); - } + if (!rb_respond_to(hash2, idTo_hash)) { + return Qfalse; + } + if (eql) { + if (rb_eql(hash2, hash1)) { + return Qtrue; + } + else { + return Qfalse; + } + } + else { + return rb_equal(hash2, hash1); + } } if (RHASH_SIZE(hash1) != RHASH_SIZE(hash2)) - return Qfalse; + return Qfalse; if (!RHASH_TABLE_EMPTY_P(hash1) && !RHASH_TABLE_EMPTY_P(hash2)) { if (RHASH_TYPE(hash1) != RHASH_TYPE(hash2)) { return Qfalse; @@ -3749,7 +3781,7 @@ hash_equal(VALUE hash1, VALUE hash2, int eql) #if 0 if (!(rb_equal(RHASH_IFNONE(hash1), RHASH_IFNONE(hash2)) && FL_TEST(hash1, RHASH_PROC_DEFAULT) == FL_TEST(hash2, RHASH_PROC_DEFAULT))) - return Qfalse; + return Qfalse; #endif return Qtrue; } @@ -3759,7 +3791,7 @@ hash_equal(VALUE hash1, VALUE hash2, int eql) * hash == object -> true or false * * Returns +true+ if all of the following are true: - * * +object+ is a \Hash object. + * * +object+ is a +Hash+ object. * * +hash+ and +object+ have the same keys (regardless of order). * * For each key +key+, <tt>hash[key] == object[key]</tt>. * @@ -3781,16 +3813,15 @@ rb_hash_equal(VALUE hash1, VALUE hash2) /* * call-seq: - * hash.eql? object -> true or false + * hash.eql?(object) -> true or false * * Returns +true+ if all of the following are true: - * * +object+ is a \Hash object. + * * +object+ is a +Hash+ object. * * +hash+ and +object+ have the same keys (regardless of order). - * * For each key +key+, <tt>h[key] eql? object[key]</tt>. + * * For each key +key+, <tt>h[key].eql?(object[key])</tt>. * * Otherwise, returns +false+. * - * Equal: * h1 = {foo: 0, bar: 1, baz: 2} * h2 = {foo: 0, bar: 1, baz: 2} * h1.eql? h2 # => true @@ -3820,10 +3851,10 @@ hash_i(VALUE key, VALUE val, VALUE arg) * call-seq: * hash.hash -> an_integer * - * Returns the \Integer hash-code for the hash. + * Returns the Integer hash-code for the hash. * - * Two \Hash objects have the same hash-code if their content is the same - * (regardless or order): + * Two +Hash+ objects have the same hash-code if their content is the same + * (regardless of order): * h1 = {foo: 0, bar: 1, baz: 2} * h2 = {baz: 2, bar: 1, foo: 0} * h2.hash == h1.hash # => true @@ -3837,7 +3868,7 @@ rb_hash_hash(VALUE hash) st_index_t hval = rb_hash_start(size); hval = rb_hash_uint(hval, (st_index_t)rb_hash_hash); if (size) { - rb_hash_foreach(hash, hash_i, (VALUE)&hval); + rb_hash_foreach(hash, hash_i, (VALUE)&hval); } hval = rb_hash_end(hval); return ST2FIX(hval); @@ -3854,13 +3885,13 @@ rb_hash_invert_i(VALUE key, VALUE value, VALUE hash) * call-seq: * hash.invert -> new_hash * - * Returns a new \Hash object with the each key-value pair inverted: + * Returns a new +Hash+ object with the each key-value pair inverted: * h = {foo: 0, bar: 1, baz: 2} * h1 = h.invert * h1 # => {0=>:foo, 1=>:bar, 2=>:baz} * * Overwrites any repeated new keys: - * (see {Entry Order}[#class-Hash-label-Entry+Order]): + * (see {Entry Order}[rdoc-ref:Hash@Entry+Order]): * h = {foo: 0, bar: 0, baz: 0} * h.invert # => {0=>:baz} */ @@ -3875,18 +3906,9 @@ rb_hash_invert(VALUE hash) } static int -rb_hash_update_callback(st_data_t *key, st_data_t *value, struct update_arg *arg, int existing) -{ - *value = arg->arg; - return ST_CONTINUE; -} - -NOINSERT_UPDATE_CALLBACK(rb_hash_update_callback) - -static int rb_hash_update_i(VALUE key, VALUE value, VALUE hash) { - RHASH_UPDATE(hash, key, rb_hash_update_callback, value); + rb_hash_aset(hash, key, value); return ST_CONTINUE; } @@ -3898,6 +3920,9 @@ rb_hash_update_block_callback(st_data_t *key, st_data_t *value, struct update_ar if (existing) { newvalue = (st_data_t)rb_yield_values(3, (VALUE)*key, (VALUE)*value, (VALUE)newvalue); } + else if (RHASH_STRING_KEY_P(arg->hash, *key) && !RB_OBJ_FROZEN(*key)) { + *key = rb_hash_key_str(*key); + } *value = newvalue; return ST_CONTINUE; } @@ -3919,9 +3944,7 @@ rb_hash_update_block_i(VALUE key, VALUE value, VALUE hash) * * Merges each of +other_hashes+ into +self+; returns +self+. * - * Each argument in +other_hashes+ must be a \Hash. - * - * \Method #update is an alias for \#merge!. + * Each argument in +other_hashes+ must be a +Hash+. * * With arguments and no block: * * Returns +self+, after the given hashes are merged into it. @@ -3993,7 +4016,7 @@ rb_hash_update_func_callback(st_data_t *key, st_data_t *value, struct update_arg VALUE newvalue = uf_arg->value; if (existing) { - newvalue = (*uf_arg->func)((VALUE)*key, (VALUE)*value, newvalue); + newvalue = (*uf_arg->func)((VALUE)*key, (VALUE)*value, newvalue); } *value = newvalue; return ST_CONTINUE; @@ -4018,13 +4041,13 @@ rb_hash_update_by(VALUE hash1, VALUE hash2, rb_hash_update_func *func) rb_hash_modify(hash1); hash2 = to_hash(hash2); if (func) { - struct update_func_arg arg; - arg.hash = hash1; - arg.func = func; - rb_hash_foreach(hash2, rb_hash_update_func_i, (VALUE)&arg); + struct update_func_arg arg; + arg.hash = hash1; + arg.func = func; + rb_hash_foreach(hash2, rb_hash_update_func_i, (VALUE)&arg); } else { - rb_hash_foreach(hash2, rb_hash_update_i, hash1); + rb_hash_foreach(hash2, rb_hash_update_i, hash1); } return hash1; } @@ -4035,16 +4058,16 @@ rb_hash_update_by(VALUE hash1, VALUE hash2, rb_hash_update_func *func) * hash.merge(*other_hashes) -> new_hash * hash.merge(*other_hashes) { |key, old_value, new_value| ... } -> new_hash * - * Returns the new \Hash formed by merging each of +other_hashes+ + * Returns the new +Hash+ formed by merging each of +other_hashes+ * into a copy of +self+. * - * Each argument in +other_hashes+ must be a \Hash. + * Each argument in +other_hashes+ must be a +Hash+. * * --- * * With arguments and no block: - * * Returns the new \Hash object formed by merging each successive - * \Hash in +other_hashes+ into +self+. + * * Returns the new +Hash+ object formed by merging each successive + * +Hash+ in +other_hashes+ into +self+. * * Each new-key entry is added at the end. * * Each duplicate-key entry's value overwrites the previous value. * @@ -4055,7 +4078,7 @@ rb_hash_update_by(VALUE hash1, VALUE hash2, rb_hash_update_func *func) * h.merge(h1, h2) # => {:foo=>0, :bar=>4, :baz=>2, :bat=>6, :bam=>5} * * With arguments and a block: - * * Returns a new \Hash object that is the merge of +self+ and each given hash. + * * Returns a new +Hash+ object that is the merge of +self+ and each given hash. * * The given hashes are merged left to right. * * Each new-key entry is added at the end. * * For each duplicate key: @@ -4092,24 +4115,17 @@ assoc_cmp(VALUE a, VALUE b) return !RTEST(rb_equal(a, b)); } -static VALUE -lookup2_call(VALUE arg) -{ - VALUE *args = (VALUE *)arg; - return rb_hash_lookup2(args[0], args[1], Qundef); -} - -struct reset_hash_type_arg { - VALUE hash; - const struct st_hash_type *orighash; +struct assoc_arg { + st_table *tbl; + st_data_t key; }; static VALUE -reset_hash_type(VALUE arg) +assoc_lookup(VALUE arg) { - struct reset_hash_type_arg *p = (struct reset_hash_type_arg *)arg; - HASH_ASSERT(RHASH_ST_TABLE_P(p->hash)); - RHASH_ST_TABLE(p->hash)->type = p->orighash; + struct assoc_arg *p = (struct assoc_arg*)arg; + st_data_t data; + if (st_lookup(p->tbl, p->key, &data)) return (VALUE)data; return Qundef; } @@ -4119,8 +4135,8 @@ assoc_i(VALUE key, VALUE val, VALUE arg) VALUE *args = (VALUE *)arg; if (RTEST(rb_equal(args[0], key))) { - args[1] = rb_assoc_new(key, val); - return ST_STOP; + args[1] = rb_assoc_new(key, val); + return ST_STOP; } return ST_CONTINUE; } @@ -4129,7 +4145,7 @@ assoc_i(VALUE key, VALUE val, VALUE arg) * call-seq: * hash.assoc(key) -> new_array or nil * - * If the given +key+ is found, returns a 2-element \Array containing that key and its value: + * If the given +key+ is found, returns a 2-element Array containing that key and its value: * h = {foo: 0, bar: 1, baz: 2} * h.assoc(:bar) # => [:bar, 1] * @@ -4139,31 +4155,31 @@ assoc_i(VALUE key, VALUE val, VALUE arg) static VALUE rb_hash_assoc(VALUE hash, VALUE key) { - st_table *table; - const struct st_hash_type *orighash; VALUE args[2]; if (RHASH_EMPTY_P(hash)) return Qnil; - ar_force_convert_table(hash, __FILE__, __LINE__); - HASH_ASSERT(RHASH_ST_TABLE_P(hash)); - table = RHASH_ST_TABLE(hash); - orighash = table->type; - - if (orighash != &identhash) { - VALUE value; - struct reset_hash_type_arg ensure_arg; - struct st_hash_type assochash; - - assochash.compare = assoc_cmp; - assochash.hash = orighash->hash; - table->type = &assochash; - args[0] = hash; - args[1] = key; - ensure_arg.hash = hash; - ensure_arg.orighash = orighash; - value = rb_ensure(lookup2_call, (VALUE)&args, reset_hash_type, (VALUE)&ensure_arg); - if (value != Qundef) return rb_assoc_new(key, value); + if (RHASH_ST_TABLE_P(hash) && !RHASH_IDENTHASH_P(hash)) { + VALUE value = Qundef; + st_table assoctable = *RHASH_ST_TABLE(hash); + assoctable.type = &(struct st_hash_type){ + .compare = assoc_cmp, + .hash = assoctable.type->hash, + }; + VALUE arg = (VALUE)&(struct assoc_arg){ + .tbl = &assoctable, + .key = (st_data_t)key, + }; + + if (RB_OBJ_FROZEN(hash)) { + value = assoc_lookup(arg); + } + else { + hash_iter_lev_inc(hash); + value = rb_ensure(assoc_lookup, arg, hash_foreach_ensure, hash); + } + hash_verify(hash); + if (!UNDEF_P(value)) return rb_assoc_new(key, value); } args[0] = key; @@ -4178,8 +4194,8 @@ rassoc_i(VALUE key, VALUE val, VALUE arg) VALUE *args = (VALUE *)arg; if (RTEST(rb_equal(args[0], val))) { - args[1] = rb_assoc_new(key, val); - return ST_STOP; + args[1] = rb_assoc_new(key, val); + return ST_STOP; } return ST_CONTINUE; } @@ -4188,9 +4204,9 @@ rassoc_i(VALUE key, VALUE val, VALUE arg) * call-seq: * hash.rassoc(value) -> new_array or nil * - * Returns a new 2-element \Array consisting of the key and value + * Returns a new 2-element Array consisting of the key and value * of the first-found entry whose value is <tt>==</tt> to value - * (see {Entry Order}[#class-Hash-label-Entry+Order]): + * (see {Entry Order}[rdoc-ref:Hash@Entry+Order]): * h = {foo: 0, bar: 1, baz: 1} * h.rassoc(1) # => [:bar, 1] * @@ -4225,7 +4241,7 @@ flatten_i(VALUE key, VALUE val, VALUE ary) * hash.flatten -> new_array * hash.flatten(level) -> new_array * - * Returns a new \Array object that is a 1-dimensional flattening of +self+. + * Returns a new Array object that is a 1-dimensional flattening of +self+. * * --- * @@ -4233,7 +4249,7 @@ flatten_i(VALUE key, VALUE val, VALUE ary) * h = {foo: 0, bar: [:bat, 3], baz: 2} * h.flatten # => [:foo, 0, :bar, [:bat, 3], :baz, 2] * - * Takes the depth of recursive flattening from \Integer argument +level+: + * Takes the depth of recursive flattening from Integer argument +level+: * h = {foo: 0, bar: [:bat, [:baz, [:bat, ]]]} * h.flatten(1) # => [:foo, 0, :bar, [:bat, [:baz, [:bat]]]] * h.flatten(2) # => [:foo, 0, :bar, :bat, [:baz, [:bat]]] @@ -4259,26 +4275,26 @@ rb_hash_flatten(int argc, VALUE *argv, VALUE hash) rb_check_arity(argc, 0, 1); if (argc) { - int level = NUM2INT(argv[0]); + int level = NUM2INT(argv[0]); - if (level == 0) return rb_hash_to_a(hash); + if (level == 0) return rb_hash_to_a(hash); - ary = rb_ary_new_capa(RHASH_SIZE(hash) * 2); - rb_hash_foreach(hash, flatten_i, ary); - level--; + ary = rb_ary_new_capa(RHASH_SIZE(hash) * 2); + rb_hash_foreach(hash, flatten_i, ary); + level--; - if (level > 0) { - VALUE ary_flatten_level = INT2FIX(level); - rb_funcallv(ary, id_flatten_bang, 1, &ary_flatten_level); - } - else if (level < 0) { - /* flatten recursively */ - rb_funcallv(ary, id_flatten_bang, 0, 0); - } + if (level > 0) { + VALUE ary_flatten_level = INT2FIX(level); + rb_funcallv(ary, id_flatten_bang, 1, &ary_flatten_level); + } + else if (level < 0) { + /* flatten recursively */ + rb_funcallv(ary, id_flatten_bang, 0, 0); + } } else { - ary = rb_ary_new_capa(RHASH_SIZE(hash) * 2); - rb_hash_foreach(hash, flatten_i, ary); + ary = rb_ary_new_capa(RHASH_SIZE(hash) * 2); + rb_hash_foreach(hash, flatten_i, ary); } return ary; @@ -4288,16 +4304,7 @@ static int delete_if_nil(VALUE key, VALUE value, VALUE hash) { if (NIL_P(value)) { - return ST_DELETE; - } - return ST_CONTINUE; -} - -static int -set_if_not_nil(VALUE key, VALUE value, VALUE hash) -{ - if (!NIL_P(value)) { - rb_hash_aset(hash, key, value); + return ST_DELETE; } return ST_CONTINUE; } @@ -4315,9 +4322,13 @@ set_if_not_nil(VALUE key, VALUE value, VALUE hash) static VALUE rb_hash_compact(VALUE hash) { - VALUE result = rb_hash_new(); + VALUE result = rb_hash_dup(hash); if (!RHASH_EMPTY_P(hash)) { - rb_hash_foreach(hash, set_if_not_nil, result); + rb_hash_foreach(result, delete_if_nil, result); + compact_after_delete(result); + } + else if (rb_hash_compare_by_id_p(hash)) { + result = rb_hash_compare_by_id(result); } return result; } @@ -4340,15 +4351,13 @@ rb_hash_compact_bang(VALUE hash) rb_hash_modify_check(hash); n = RHASH_SIZE(hash); if (n) { - rb_hash_foreach(hash, delete_if_nil, hash); + rb_hash_foreach(hash, delete_if_nil, hash); if (n != RHASH_SIZE(hash)) - return hash; + return hash; } return Qnil; } -static st_table *rb_init_identtable_with_size(st_index_t size); - /* * call-seq: * hash.compare_by_identity -> self @@ -4386,16 +4395,33 @@ rb_hash_compare_by_id(VALUE hash) if (rb_hash_compare_by_id_p(hash)) return hash; rb_hash_modify_check(hash); - ar_force_convert_table(hash, __FILE__, __LINE__); - HASH_ASSERT(RHASH_ST_TABLE_P(hash)); + if (hash_iterating_p(hash)) { + rb_raise(rb_eRuntimeError, "compare_by_identity during iteration"); + } + + if (RHASH_TABLE_EMPTY_P(hash)) { + // Fast path: There's nothing to rehash, so we don't need a `tmp` table. + // We're most likely an AR table, so this will need an allocation. + ar_force_convert_table(hash, __FILE__, __LINE__); + HASH_ASSERT(RHASH_ST_TABLE_P(hash)); + + RHASH_ST_TABLE(hash)->type = &identhash; + } + else { + // Slow path: Need to rehash the members of `self` into a new + // `tmp` table using the new `identhash` compare/hash functions. + tmp = hash_alloc(0); + hash_st_table_init(tmp, &identhash, RHASH_SIZE(hash)); + identtable = RHASH_ST_TABLE(tmp); - tmp = hash_alloc(0); - identtable = rb_init_identtable_with_size(RHASH_SIZE(hash)); - RHASH_ST_TABLE_SET(tmp, identtable); - rb_hash_foreach(hash, rb_hash_rehash_i, (VALUE)tmp); - st_free_table(RHASH_ST_TABLE(hash)); - RHASH_ST_TABLE_SET(hash, identtable); - RHASH_ST_CLEAR(tmp); + rb_hash_foreach(hash, rb_hash_rehash_i, (VALUE)tmp); + rb_hash_free(hash); + + // We know for sure `identtable` is an st table, + // so we can skip `ar_force_convert_table` here. + RHASH_ST_TABLE_SET(hash, identtable); + RHASH_ST_CLEAR(tmp); + } return hash; } @@ -4407,17 +4433,17 @@ rb_hash_compare_by_id(VALUE hash) * Returns +true+ if #compare_by_identity has been called, +false+ otherwise. */ -MJIT_FUNC_EXPORTED VALUE +VALUE rb_hash_compare_by_id_p(VALUE hash) { - return RBOOL(RHASH_ST_TABLE_P(hash) && RHASH_ST_TABLE(hash)->type == &identhash); + return RBOOL(RHASH_IDENTHASH_P(hash)); } VALUE rb_ident_hash_new(void) { VALUE hash = rb_hash_new(); - RHASH_ST_TABLE_SET(hash, st_init_table(&identhash)); + hash_st_table_init(hash, &identhash, 0); return hash; } @@ -4425,7 +4451,7 @@ VALUE rb_ident_hash_new_with_size(st_index_t size) { VALUE hash = rb_hash_new(); - RHASH_ST_TABLE_SET(hash, st_init_table_with_size(&identhash, size)); + hash_st_table_init(hash, &identhash, size); return hash; } @@ -4435,19 +4461,13 @@ rb_init_identtable(void) return st_init_table(&identhash); } -static st_table * -rb_init_identtable_with_size(st_index_t size) -{ - return st_init_table_with_size(&identhash, size); -} - static int any_p_i(VALUE key, VALUE value, VALUE arg) { VALUE ret = rb_yield(rb_assoc_new(key, value)); if (RTEST(ret)) { - *(VALUE *)arg = Qtrue; - return ST_STOP; + *(VALUE *)arg = Qtrue; + return ST_STOP; } return ST_CONTINUE; } @@ -4457,8 +4477,8 @@ any_p_i_fast(VALUE key, VALUE value, VALUE arg) { VALUE ret = rb_yield_values(2, key, value); if (RTEST(ret)) { - *(VALUE *)arg = Qtrue; - return ST_STOP; + *(VALUE *)arg = Qtrue; + return ST_STOP; } return ST_CONTINUE; } @@ -4468,8 +4488,8 @@ any_p_i_pattern(VALUE key, VALUE value, VALUE arg) { VALUE ret = rb_funcall(((VALUE *)arg)[1], idEqq, 1, rb_assoc_new(key, value)); if (RTEST(ret)) { - *(VALUE *)arg = Qtrue; - return ST_STOP; + *(VALUE *)arg = Qtrue; + return ST_STOP; } return ST_CONTINUE; } @@ -4483,6 +4503,9 @@ any_p_i_pattern(VALUE key, VALUE value, VALUE arg) * Returns +true+ if any element satisfies a given criterion; * +false+ otherwise. * + * If +self+ has no element, returns +false+ and argument or block + * are not used. + * * With no argument and no block, * returns +true+ if +self+ is non-empty; +false+ if empty. * @@ -4501,6 +4524,8 @@ any_p_i_pattern(VALUE key, VALUE value, VALUE arg) * h = {foo: 0, bar: 1, baz: 2} * h.any? {|key, value| value < 3 } # => true * h.any? {|key, value| value > 3 } # => false + * + * Related: Enumerable#any? */ static VALUE @@ -4515,19 +4540,19 @@ rb_hash_any_p(int argc, VALUE *argv, VALUE hash) if (rb_block_given_p()) { rb_warn("given block not used"); } - args[1] = argv[0]; + args[1] = argv[0]; - rb_hash_foreach(hash, any_p_i_pattern, (VALUE)args); + rb_hash_foreach(hash, any_p_i_pattern, (VALUE)args); } else { - if (!rb_block_given_p()) { - /* yields pairs, never false */ - return Qtrue; - } + if (!rb_block_given_p()) { + /* yields pairs, never false */ + return Qtrue; + } if (rb_block_pair_yield_optimizable()) - rb_hash_foreach(hash, any_p_i_fast, (VALUE)args); - else - rb_hash_foreach(hash, any_p_i, (VALUE)args); + rb_hash_foreach(hash, any_p_i_fast, (VALUE)args); + else + rb_hash_foreach(hash, any_p_i, (VALUE)args); } return args[0]; } @@ -4552,7 +4577,7 @@ rb_hash_any_p(int argc, VALUE *argv, VALUE hash) * h = {foo: {bar: [:a, :b, :c]}} * h.dig(:foo, :bar, 2) # => :c * - * This method will use the {default values}[#class-Hash-label-Default+Values] + * This method will use the {default values}[rdoc-ref:Hash@Default+Values] * for keys that are not present: * h = {foo: {bar: [:a, :b, :c]}} * h.dig(:hello) # => nil @@ -4576,7 +4601,7 @@ hash_le_i(VALUE key, VALUE value, VALUE arg) { VALUE *args = (VALUE *)arg; VALUE v = rb_hash_lookup2(args[0], key, Qundef); - if (v != Qundef && rb_equal(value, v)) return ST_CONTINUE; + if (!UNDEF_P(v) && rb_equal(value, v)) return ST_CONTINUE; args[1] = Qfalse; return ST_STOP; } @@ -4678,7 +4703,7 @@ hash_proc_call(RB_BLOCK_CALL_FUNC_ARGLIST(key, hash)) * call-seq: * hash.to_proc -> proc * - * Returns a \Proc object that maps a key to its value: + * Returns a Proc object that maps a key to its value: * h = {foo: 0, bar: 1, baz: 2} * proc = h.to_proc * proc.class # => Proc @@ -4692,6 +4717,7 @@ rb_hash_to_proc(VALUE hash) return rb_func_lambda_new(hash_proc_call, hash, 1, 1); } +/* :nodoc: */ static VALUE rb_hash_deconstruct_keys(VALUE hash, VALUE keys) { @@ -4722,14 +4748,13 @@ rb_hash_add_new_element(VALUE hash, VALUE key, VALUE val) args[1] = val; if (RHASH_AR_TABLE_P(hash)) { - hash_ar_table(hash); - ret = ar_update(hash, (st_data_t)key, add_new_i, (st_data_t)args); if (ret != -1) { return ret; } - ar_try_convert_table(hash); + ar_force_convert_table(hash, __FILE__, __LINE__); } + tbl = RHASH_TBL_RAW(hash); return st_update(tbl, (st_data_t)key, add_new_i, (st_data_t)args); @@ -4762,15 +4787,6 @@ rb_hash_bulk_insert(long argc, const VALUE *argv, VALUE hash) if (argc > 0) { st_index_t size = argc / 2; - if (RHASH_TABLE_NULL_P(hash)) { - if (size <= RHASH_AR_TABLE_MAX_SIZE) { - hash_ar_table(hash); - } - else { - RHASH_TBL_RAW(hash); - } - } - if (RHASH_AR_TABLE_P(hash) && (RHASH_AR_TABLE_SIZE(hash) + size <= RHASH_AR_TABLE_MAX_SIZE)) { ar_bulk_insert(hash, argc, argv); @@ -4808,6 +4824,9 @@ extern char **environ; #define ENVNMATCH(s1, s2, n) (memcmp((s1), (s2), (n)) == 0) #endif +#define ENV_LOCK() RB_VM_LOCK_ENTER() +#define ENV_UNLOCK() RB_VM_LOCK_LEAVE() + static inline rb_encoding * env_encoding(void) { @@ -4828,12 +4847,6 @@ env_enc_str_new(const char *ptr, long len, rb_encoding *enc) } static VALUE -env_enc_str_new_cstr(const char *ptr, rb_encoding *enc) -{ - return env_enc_str_new(ptr, strlen(ptr), enc); -} - -static VALUE env_str_new(const char *ptr, long len) { return env_enc_str_new(ptr, len, env_encoding()); @@ -4846,28 +4859,47 @@ env_str_new2(const char *ptr) return env_str_new(ptr, strlen(ptr)); } -static const char TZ_ENV[] = "TZ"; - static VALUE -env_name_new(const char *name, const char *ptr) +getenv_with_lock(const char *name) +{ + VALUE ret; + ENV_LOCK(); + { + const char *val = getenv(name); + ret = env_str_new2(val); + } + ENV_UNLOCK(); + return ret; +} + +static bool +has_env_with_lock(const char *name) { - return env_enc_str_new_cstr(ptr, env_encoding()); + const char *val; + + ENV_LOCK(); + { + val = getenv(name); + } + ENV_UNLOCK(); + + return val ? true : false; } +static const char TZ_ENV[] = "TZ"; + static void * -get_env_cstr( - VALUE str, - const char *name) +get_env_cstr(VALUE str, const char *name) { char *var; rb_encoding *enc = rb_enc_get(str); if (!rb_enc_asciicompat(enc)) { - rb_raise(rb_eArgError, "bad environment variable %s: ASCII incompatible encoding: %s", - name, rb_enc_name(enc)); + rb_raise(rb_eArgError, "bad environment variable %s: ASCII incompatible encoding: %s", + name, rb_enc_name(enc)); } var = RSTRING_PTR(str); if (memchr(var, '\0', RSTRING_LEN(str))) { - rb_raise(rb_eArgError, "bad environment variable %s: contains null byte", name); + rb_raise(rb_eArgError, "bad environment variable %s: contains null byte", name); } return rb_str_fill_terminator(str, 1); /* ASCII compatible */ } @@ -4879,7 +4911,7 @@ static inline const char * env_name(volatile VALUE *s) { const char *name; - SafeStringValue(*s); + StringValue(*s); get_env_ptr(name, *s); return name; } @@ -4906,17 +4938,13 @@ static VALUE env_delete(VALUE name) { const char *nam = env_name(name); - const char *val = getenv(nam); - reset_by_modified_env(nam); + VALUE val = getenv_with_lock(nam); - if (val) { - VALUE value = env_str_new2(val); - - ruby_setenv(nam, 0); - return value; + if (!NIL_P(val)) { + ruby_setenv(nam, 0); } - return Qnil; + return val; } /* @@ -4942,7 +4970,7 @@ env_delete(VALUE name) * ENV.delete('foo') { |name| raise 'ignored' } # => "0" * * Raises an exception if +name+ is invalid. - * See {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]. + * See {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]. */ static VALUE env_delete_m(VALUE obj, VALUE name) @@ -4964,19 +4992,14 @@ env_delete_m(VALUE obj, VALUE name) * Returns +nil+ if the named variable does not exist. * * Raises an exception if +name+ is invalid. - * See {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]. + * See {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]. */ static VALUE rb_f_getenv(VALUE obj, VALUE name) { - const char *nam, *env; - - nam = env_name(name); - env = getenv(nam); - if (env) { - return env_name_new(nam, env); - } - return Qnil; + const char *nam = env_name(name); + VALUE env = getenv_with_lock(nam); + return env; } /* @@ -5002,38 +5025,33 @@ rb_f_getenv(VALUE obj, VALUE name) * and neither default value nor block is given: * ENV.fetch('foo') # Raises KeyError (key not found: "foo") * Raises an exception if +name+ is invalid. - * See {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]. + * See {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]. */ static VALUE env_fetch(int argc, VALUE *argv, VALUE _) { VALUE key; long block_given; - const char *nam, *env; + const char *nam; + VALUE env; rb_check_arity(argc, 1, 2); key = argv[0]; block_given = rb_block_given_p(); if (block_given && argc == 2) { - rb_warn("block supersedes default value argument"); + rb_warn("block supersedes default value argument"); } nam = env_name(key); - env = getenv(nam); - if (!env) { - if (block_given) return rb_yield(key); - if (argc == 1) { - rb_key_err_raise(rb_sprintf("key not found: \"%"PRIsVALUE"\"", key), envtbl, key); - } - return argv[1]; - } - return env_name_new(nam, env); -} + env = getenv_with_lock(nam); -int -rb_env_path_tainted(void) -{ - rb_warn_deprecated_to_remove_at(3.2, "rb_env_path_tainted", NULL); - return 0; + if (NIL_P(env)) { + if (block_given) return rb_yield(key); + if (argc == 1) { + rb_key_err_raise(rb_sprintf("key not found: \"%"PRIsVALUE"\"", key), envtbl, key); + } + return argv[1]; + } + return env; } #if defined(_WIN32) || (defined(HAVE_SETENV) && defined(HAVE_UNSETENV)) @@ -5043,7 +5061,7 @@ in_origenv(const char *str) { char **env; for (env = origenviron; *env; ++env) { - if (*env == str) return 1; + if (*env == str) return 1; } return 0; } @@ -5051,13 +5069,15 @@ in_origenv(const char *str) static int envix(const char *nam) { + // should be locked + register int i, len = strlen(nam); char **env; env = GET_ENVIRON(environ); for (i = 0; env[i]; i++) { - if (ENVNMATCH(env[i],nam,len) && env[i][len] == '=') - break; /* memcmp must come first to avoid */ + if (ENVNMATCH(env[i],nam,len) && env[i][len] == '=') + break; /* memcmp must come first to avoid */ } /* potential SEGV's */ FREE_ENVIRON(environ); return i; @@ -5087,16 +5107,16 @@ static int check_envsize(size_t n) { if (_WIN32_WINNT < 0x0600 && rb_w32_osver() < 6) { - /* https://msdn.microsoft.com/en-us/library/windows/desktop/ms682653(v=vs.85).aspx */ - /* Windows Server 2003 and Windows XP: The maximum size of the - * environment block for the process is 32,767 characters. */ - WCHAR* p = GetEnvironmentStringsW(); - if (!p) return -1; /* never happen */ - n += getenvsize(p); - FreeEnvironmentStringsW(p); - if (n >= getenvblocksize()) { - return -1; - } + /* https://msdn.microsoft.com/en-us/library/windows/desktop/ms682653(v=vs.85).aspx */ + /* Windows Server 2003 and Windows XP: The maximum size of the + * environment block for the process is 32,767 characters. */ + WCHAR* p = GetEnvironmentStringsW(); + if (!p) return -1; /* never happen */ + n += getenvsize(p); + FreeEnvironmentStringsW(p); + if (n >= getenvblocksize()) { + return -1; + } } return 0; } @@ -5117,7 +5137,7 @@ static const char * check_envname(const char *name) { if (strchr(name, '=')) { - invalid_envname(name); + invalid_envname(name); } return name; } @@ -5138,56 +5158,78 @@ ruby_setenv(const char *name, const char *value) check_envname(name); len = MultiByteToWideChar(CP_UTF8, 0, name, -1, NULL, 0); if (value) { - int len2; - len2 = MultiByteToWideChar(CP_UTF8, 0, value, -1, NULL, 0); - if (check_envsize((size_t)len + len2)) { /* len and len2 include '\0' */ - goto fail; /* 2 for '=' & '\0' */ - } - wname = ALLOCV_N(WCHAR, buf, len + len2); - wvalue = wname + len; - MultiByteToWideChar(CP_UTF8, 0, name, -1, wname, len); - MultiByteToWideChar(CP_UTF8, 0, value, -1, wvalue, len2); + int len2; + len2 = MultiByteToWideChar(CP_UTF8, 0, value, -1, NULL, 0); + if (check_envsize((size_t)len + len2)) { /* len and len2 include '\0' */ + goto fail; /* 2 for '=' & '\0' */ + } + wname = ALLOCV_N(WCHAR, buf, len + len2); + wvalue = wname + len; + MultiByteToWideChar(CP_UTF8, 0, name, -1, wname, len); + MultiByteToWideChar(CP_UTF8, 0, value, -1, wvalue, len2); #ifndef HAVE__WPUTENV_S - wname[len-1] = L'='; + wname[len-1] = L'='; #endif } else { - wname = ALLOCV_N(WCHAR, buf, len + 1); - MultiByteToWideChar(CP_UTF8, 0, name, -1, wname, len); - wvalue = wname + len; - *wvalue = L'\0'; + wname = ALLOCV_N(WCHAR, buf, len + 1); + MultiByteToWideChar(CP_UTF8, 0, name, -1, wname, len); + wvalue = wname + len; + *wvalue = L'\0'; #ifndef HAVE__WPUTENV_S - wname[len-1] = L'='; + wname[len-1] = L'='; #endif } + + ENV_LOCK(); + { #ifndef HAVE__WPUTENV_S - failed = _wputenv(wname); + failed = _wputenv(wname); #else - failed = _wputenv_s(wname, wvalue); + failed = _wputenv_s(wname, wvalue); #endif + } + ENV_UNLOCK(); + ALLOCV_END(buf); /* even if putenv() failed, clean up and try to delete the * variable from the system area. */ if (!value || !*value) { - /* putenv() doesn't handle empty value */ - if (!SetEnvironmentVariable(name, value) && - GetLastError() != ERROR_ENVVAR_NOT_FOUND) goto fail; + /* putenv() doesn't handle empty value */ + if (!SetEnvironmentVariable(name, value) && + GetLastError() != ERROR_ENVVAR_NOT_FOUND) goto fail; } if (failed) { fail: - invalid_envname(name); + invalid_envname(name); } #elif defined(HAVE_SETENV) && defined(HAVE_UNSETENV) if (value) { - if (setenv(name, value, 1)) - rb_sys_fail_str(rb_sprintf("setenv(%s)", name)); + int ret; + ENV_LOCK(); + { + ret = setenv(name, value, 1); + } + ENV_UNLOCK(); + + if (ret) rb_sys_fail_sprintf("setenv(%s)", name); } else { #ifdef VOID_UNSETENV - unsetenv(name); + ENV_LOCK(); + { + unsetenv(name); + } + ENV_UNLOCK(); #else - if (unsetenv(name)) - rb_sys_fail_str(rb_sprintf("unsetenv(%s)", name)); + int ret; + ENV_LOCK(); + { + ret = unsetenv(name); + } + ENV_UNLOCK(); + + if (ret) rb_sys_fail_sprintf("unsetenv(%s)", name); #endif } #elif defined __sun @@ -5201,64 +5243,85 @@ ruby_setenv(const char *name, const char *value) check_envname(name); len = strlen(name); if (value) { - mem_size = len + strlen(value) + 2; - mem_ptr = malloc(mem_size); - if (mem_ptr == NULL) - rb_sys_fail_str(rb_sprintf("malloc("PRIuSIZE")", mem_size)); - snprintf(mem_ptr, mem_size, "%s=%s", name, value); - } - for (env_ptr = GET_ENVIRON(environ); (str = *env_ptr) != 0; ++env_ptr) { - if (!strncmp(str, name, len) && str[len] == '=') { - if (!in_origenv(str)) free(str); - while ((env_ptr[0] = env_ptr[1]) != 0) env_ptr++; - break; - } + mem_size = len + strlen(value) + 2; + mem_ptr = malloc(mem_size); + if (mem_ptr == NULL) + rb_sys_fail_sprintf("malloc(%"PRIuSIZE")", mem_size); + snprintf(mem_ptr, mem_size, "%s=%s", name, value); + } + + ENV_LOCK(); + { + for (env_ptr = GET_ENVIRON(environ); (str = *env_ptr) != 0; ++env_ptr) { + if (!strncmp(str, name, len) && str[len] == '=') { + if (!in_origenv(str)) free(str); + while ((env_ptr[0] = env_ptr[1]) != 0) env_ptr++; + break; + } + } } + ENV_UNLOCK(); + if (value) { - if (putenv(mem_ptr)) { - free(mem_ptr); - rb_sys_fail_str(rb_sprintf("putenv(%s)", name)); - } + int ret; + ENV_LOCK(); + { + ret = putenv(mem_ptr); + } + ENV_UNLOCK(); + + if (ret) { + free(mem_ptr); + rb_sys_fail_sprintf("putenv(%s)", name); + } } #else /* WIN32 */ size_t len; int i; - i=envix(name); /* where does it go? */ - - if (environ == origenviron) { /* need we copy environment? */ - int j; - int max; - char **tmpenv; - - for (max = i; environ[max]; max++) ; - tmpenv = ALLOC_N(char*, max+2); - for (j=0; j<max; j++) /* copy environment */ - tmpenv[j] = ruby_strdup(environ[j]); - tmpenv[max] = 0; - environ = tmpenv; /* tell exec where it is now */ - } - if (environ[i]) { - char **envp = origenviron; - while (*envp && *envp != environ[i]) envp++; - if (!*envp) - xfree(environ[i]); - if (!value) { - while (environ[i]) { - environ[i] = environ[i+1]; - i++; - } - return; - } - } - else { /* does not exist yet */ - if (!value) return; - REALLOC_N(environ, char*, i+2); /* just expand it a bit */ - environ[i+1] = 0; /* make sure it's null terminated */ - } - len = strlen(name) + strlen(value) + 2; - environ[i] = ALLOC_N(char, len); - snprintf(environ[i],len,"%s=%s",name,value); /* all that work just for this */ + ENV_LOCK(); + { + i = envix(name); /* where does it go? */ + + if (environ == origenviron) { /* need we copy environment? */ + int j; + int max; + char **tmpenv; + + for (max = i; environ[max]; max++) ; + tmpenv = ALLOC_N(char*, max+2); + for (j=0; j<max; j++) /* copy environment */ + tmpenv[j] = ruby_strdup(environ[j]); + tmpenv[max] = 0; + environ = tmpenv; /* tell exec where it is now */ + } + + if (environ[i]) { + char **envp = origenviron; + while (*envp && *envp != environ[i]) envp++; + if (!*envp) + xfree(environ[i]); + if (!value) { + while (environ[i]) { + environ[i] = environ[i+1]; + i++; + } + goto finish; + } + } + else { /* does not exist yet */ + if (!value) goto finish; + REALLOC_N(environ, char*, i+2); /* just expand it a bit */ + environ[i+1] = 0; /* make sure it's null terminated */ + } + + len = strlen(name) + strlen(value) + 2; + environ[i] = ALLOC_N(char, len); + snprintf(environ[i],len,"%s=%s",name,value); /* all that work just for this */ + + finish:; + } + ENV_UNLOCK(); #endif /* WIN32 */ } @@ -5273,11 +5336,9 @@ ruby_unsetenv(const char *name) * ENV[name] = value -> value * ENV.store(name, value) -> value * - * ENV.store is an alias for ENV.[]=. - * * Creates, updates, or deletes the named environment variable, returning the value. * Both +name+ and +value+ may be instances of String. - * See {Valid Names and Values}[#class-ENV-label-Valid+Names+and+Values]. + * See {Valid Names and Values}[rdoc-ref:ENV@Valid+Names+and+Values]. * * - If the named environment variable does not exist: * - If +value+ is +nil+, does nothing. @@ -5310,7 +5371,7 @@ ruby_unsetenv(const char *name) * ENV.include?('bar') # => false * * Raises an exception if +name+ or +value+ is invalid. - * See {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]. + * See {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]. */ static VALUE env_aset_m(VALUE obj, VALUE nm, VALUE val) @@ -5325,10 +5386,10 @@ env_aset(VALUE nm, VALUE val) if (NIL_P(val)) { env_delete(nm); - return Qnil; + return Qnil; } - SafeStringValue(nm); - SafeStringValue(val); + StringValue(nm); + StringValue(val); /* nm can be modified in `val.to_str`, don't get `name` before * check for `val` */ get_env_ptr(name, nm); @@ -5342,23 +5403,26 @@ env_aset(VALUE nm, VALUE val) static VALUE env_keys(int raw) { - char **env; - VALUE ary; rb_encoding *enc = raw ? 0 : rb_locale_encoding(); - - ary = rb_ary_new(); - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s) { - const char *p = *env; - size_t l = s - p; - VALUE e = raw ? rb_utf8_str_new(p, l) : env_enc_str_new(p, l, enc); - rb_ary_push(ary, e); - } - env++; + VALUE ary = rb_ary_new(); + + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + while (*env) { + char *s = strchr(*env, '='); + if (s) { + const char *p = *env; + size_t l = s - p; + VALUE e = raw ? rb_utf8_str_new(p, l) : env_enc_str_new(p, l, enc); + rb_ary_push(ary, e); + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); + return ary; } @@ -5370,7 +5434,7 @@ env_keys(int raw) * ENV.replace('foo' => '0', 'bar' => '1') * ENV.keys # => ['bar', 'foo'] * The order of the names is OS-dependent. - * See {About Ordering}[#class-ENV-label-About+Ordering]. + * See {About Ordering}[rdoc-ref:ENV@About+Ordering]. * * Returns the empty Array if ENV is empty. */ @@ -5387,13 +5451,18 @@ rb_env_size(VALUE ehash, VALUE args, VALUE eobj) char **env; long cnt = 0; - env = GET_ENVIRON(environ); - for (; *env ; ++env) { - if (strchr(*env, '=')) { - cnt++; - } + ENV_LOCK(); + { + env = GET_ENVIRON(environ); + for (; *env ; ++env) { + if (strchr(*env, '=')) { + cnt++; + } + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); + return LONG2FIX(cnt); } @@ -5423,7 +5492,7 @@ env_each_key(VALUE ehash) RETURN_SIZED_ENUMERATOR(ehash, 0, 0, rb_env_size); keys = env_keys(FALSE); for (i=0; i<RARRAY_LEN(keys); i++) { - rb_yield(RARRAY_AREF(keys, i)); + rb_yield(RARRAY_AREF(keys, i)); } return ehash; } @@ -5431,19 +5500,23 @@ env_each_key(VALUE ehash) static VALUE env_values(void) { - VALUE ary; - char **env; + VALUE ary = rb_ary_new(); - ary = rb_ary_new(); - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s) { - rb_ary_push(ary, env_str_new2(s+1)); - } - env++; + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + + while (*env) { + char *s = strchr(*env, '='); + if (s) { + rb_ary_push(ary, env_str_new2(s+1)); + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); + return ary; } @@ -5455,7 +5528,7 @@ env_values(void) * ENV.replace('foo' => '0', 'bar' => '1') * ENV.values # => ['1', '0'] * The order of the values is OS-dependent. - * See {About Ordering}[#class-ENV-label-About+Ordering]. + * See {About Ordering}[rdoc-ref:ENV@About+Ordering]. * * Returns the empty Array if ENV is empty. */ @@ -5491,7 +5564,7 @@ env_each_value(VALUE ehash) RETURN_SIZED_ENUMERATOR(ehash, 0, 0, rb_env_size); values = env_values(); for (i=0; i<RARRAY_LEN(values); i++) { - rb_yield(RARRAY_AREF(values, i)); + rb_yield(RARRAY_AREF(values, i)); } return ehash; } @@ -5503,7 +5576,7 @@ env_each_value(VALUE ehash) * ENV.each_pair { |name, value| block } -> ENV * ENV.each_pair -> an_enumerator * - * Yields each environment variable name and its value as a 2-element \Array: + * Yields each environment variable name and its value as a 2-element Array: * h = {} * ENV.each_pair { |name, value| h[name] = value } # => ENV * h # => {"bar"=>"1", "foo"=>"0"} @@ -5517,34 +5590,39 @@ env_each_value(VALUE ehash) static VALUE env_each_pair(VALUE ehash) { - char **env; - VALUE ary; long i; RETURN_SIZED_ENUMERATOR(ehash, 0, 0, rb_env_size); - ary = rb_ary_new(); - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s) { - rb_ary_push(ary, env_str_new(*env, s-*env)); - rb_ary_push(ary, env_str_new2(s+1)); - } - env++; + VALUE ary = rb_ary_new(); + + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + + while (*env) { + char *s = strchr(*env, '='); + if (s) { + rb_ary_push(ary, env_str_new(*env, s-*env)); + rb_ary_push(ary, env_str_new2(s+1)); + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); if (rb_block_pair_yield_optimizable()) { - for (i=0; i<RARRAY_LEN(ary); i+=2) { - rb_yield_values(2, RARRAY_AREF(ary, i), RARRAY_AREF(ary, i+1)); - } + for (i=0; i<RARRAY_LEN(ary); i+=2) { + rb_yield_values(2, RARRAY_AREF(ary, i), RARRAY_AREF(ary, i+1)); + } } else { - for (i=0; i<RARRAY_LEN(ary); i+=2) { - rb_yield(rb_assoc_new(RARRAY_AREF(ary, i), RARRAY_AREF(ary, i+1))); - } + for (i=0; i<RARRAY_LEN(ary); i+=2) { + rb_yield(rb_assoc_new(RARRAY_AREF(ary, i), RARRAY_AREF(ary, i+1))); + } } + return ehash; } @@ -5581,13 +5659,13 @@ env_reject_bang(VALUE ehash) keys = env_keys(FALSE); RBASIC_CLEAR_CLASS(keys); for (i=0; i<RARRAY_LEN(keys); i++) { - VALUE val = rb_f_getenv(Qnil, RARRAY_AREF(keys, i)); - if (!NIL_P(val)) { - if (RTEST(rb_yield_values(2, RARRAY_AREF(keys, i), val))) { + VALUE val = rb_f_getenv(Qnil, RARRAY_AREF(keys, i)); + if (!NIL_P(val)) { + if (RTEST(rb_yield_values(2, RARRAY_AREF(keys, i), val))) { env_delete(RARRAY_AREF(keys, i)); - del++; - } - } + del++; + } + } } RB_GC_GUARD(keys); if (del == 0) return Qnil; @@ -5634,10 +5712,10 @@ env_delete_if(VALUE ehash) * Returns +nil+ in the Array for each name that is not an ENV name: * ENV.values_at('foo', 'bat', 'bar', 'bam') # => ["0", nil, "1", nil] * - * Returns an empty \Array if no names given. + * Returns an empty Array if no names given. * * Raises an exception if any name is invalid. - * See {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]. + * See {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]. */ static VALUE env_values_at(int argc, VALUE *argv, VALUE _) @@ -5647,7 +5725,7 @@ env_values_at(int argc, VALUE *argv, VALUE _) result = rb_ary_new(); for (i=0; i<argc; i++) { - rb_ary_push(result, rb_f_getenv(Qnil, argv[i])); + rb_ary_push(result, rb_f_getenv(Qnil, argv[i])); } return result; } @@ -5659,8 +5737,6 @@ env_values_at(int argc, VALUE *argv, VALUE _) * ENV.filter { |name, value| block } -> hash of name/value pairs * ENV.filter -> an_enumerator * - * ENV.filter is an alias for ENV.select. - * * Yields each environment variable name and its value as a 2-element Array, * returning a Hash of the names and values for which the block returns a truthy value: * ENV.replace('foo' => '0', 'bar' => '1', 'baz' => '2') @@ -5684,13 +5760,13 @@ env_select(VALUE ehash) result = rb_hash_new(); keys = env_keys(FALSE); for (i = 0; i < RARRAY_LEN(keys); ++i) { - VALUE key = RARRAY_AREF(keys, i); - VALUE val = rb_f_getenv(Qnil, key); - if (!NIL_P(val)) { - if (RTEST(rb_yield_values(2, key, val))) { - rb_hash_aset(result, key, val); - } - } + VALUE key = RARRAY_AREF(keys, i); + VALUE val = rb_f_getenv(Qnil, key); + if (!NIL_P(val)) { + if (RTEST(rb_yield_values(2, key, val))) { + rb_hash_aset(result, key, val); + } + } } RB_GC_GUARD(keys); @@ -5704,8 +5780,6 @@ env_select(VALUE ehash) * ENV.filter! { |name, value| block } -> ENV or nil * ENV.filter! -> an_enumerator * - * ENV.filter! is an alias for ENV.select!. - * * Yields each environment variable name and its value as a 2-element Array, * deleting each entry for which the block returns +false+ or +nil+, * and returning ENV if any deletions made, or +nil+ otherwise: @@ -5745,13 +5819,13 @@ env_select_bang(VALUE ehash) keys = env_keys(FALSE); RBASIC_CLEAR_CLASS(keys); for (i=0; i<RARRAY_LEN(keys); i++) { - VALUE val = rb_f_getenv(Qnil, RARRAY_AREF(keys, i)); - if (!NIL_P(val)) { - if (!RTEST(rb_yield_values(2, RARRAY_AREF(keys, i), val))) { + VALUE val = rb_f_getenv(Qnil, RARRAY_AREF(keys, i)); + if (!NIL_P(val)) { + if (!RTEST(rb_yield_values(2, RARRAY_AREF(keys, i), val))) { env_delete(RARRAY_AREF(keys, i)); - del++; - } - } + del++; + } + } } RB_GC_GUARD(keys); if (del == 0) return Qnil; @@ -5793,7 +5867,7 @@ env_keep_if(VALUE ehash) * ENV.slice('foo', 'baz') # => {"foo"=>"0", "baz"=>"2"} * ENV.slice('baz', 'foo') # => {"baz"=>"2", "foo"=>"0"} * Raises an exception if any of the +names+ is invalid - * (see {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]): + * (see {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]): * ENV.slice('foo', 'bar', :bat) # Raises TypeError (no implicit conversion of Symbol into String) */ static VALUE @@ -5873,27 +5947,31 @@ env_to_s(VALUE _) static VALUE env_inspect(VALUE _) { - char **env; - VALUE str, i; + VALUE i; + VALUE str = rb_str_buf_new2("{"); - str = rb_str_buf_new2("{"); - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - - if (env != environ) { - rb_str_buf_cat2(str, ", "); - } - if (s) { - rb_str_buf_cat2(str, "\""); - rb_str_buf_cat(str, *env, s-*env); - rb_str_buf_cat2(str, "\"=>"); - i = rb_inspect(rb_str_new2(s+1)); - rb_str_buf_append(str, i); - } - env++; + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + while (*env) { + char *s = strchr(*env, '='); + + if (env != environ) { + rb_str_buf_cat2(str, ", "); + } + if (s) { + rb_str_buf_cat2(str, "\""); + rb_str_buf_cat(str, *env, s-*env); + rb_str_buf_cat2(str, "\"=>"); + i = rb_inspect(rb_str_new2(s+1)); + rb_str_buf_append(str, i); + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); + rb_str_buf_cat2(str, "}"); return str; @@ -5911,20 +5989,23 @@ env_inspect(VALUE _) static VALUE env_to_a(VALUE _) { - char **env; - VALUE ary; + VALUE ary = rb_ary_new(); - ary = rb_ary_new(); - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s) { - rb_ary_push(ary, rb_assoc_new(env_str_new(*env, s-*env), - env_str_new2(s+1))); - } - env++; + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + while (*env) { + char *s = strchr(*env, '='); + if (s) { + rb_ary_push(ary, rb_assoc_new(env_str_new(*env, s-*env), + env_str_new2(s+1))); + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); + return ary; } @@ -5942,6 +6023,22 @@ env_none(VALUE _) return Qnil; } +static int +env_size_with_lock(void) +{ + int i = 0; + + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + while (env[i]) i++; + FREE_ENVIRON(environ); + } + ENV_UNLOCK(); + + return i; +} + /* * call-seq: * ENV.length -> an_integer @@ -5955,14 +6052,7 @@ env_none(VALUE _) static VALUE env_size(VALUE _) { - int i; - char **env; - - env = GET_ENVIRON(environ); - for (i=0; env[i]; i++) - ; - FREE_ENVIRON(environ); - return INT2FIX(i); + return INT2FIX(env_size_with_lock()); } /* @@ -5978,15 +6068,19 @@ env_size(VALUE _) static VALUE env_empty_p(VALUE _) { - char **env; + bool empty = true; - env = GET_ENVIRON(environ); - if (env[0] == 0) { - FREE_ENVIRON(environ); - return Qtrue; + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + if (env[0] != 0) { + empty = false; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); - return Qfalse; + ENV_UNLOCK(); + + return RBOOL(empty); } /* @@ -5996,8 +6090,6 @@ env_empty_p(VALUE _) * ENV.member?(name) -> true or false * ENV.key?(name) -> true or false * - * ENV.has_key?, ENV.member?, and ENV.key? are aliases for ENV.include?. - * * Returns +true+ if there is an environment variable with the given +name+: * ENV.replace('foo' => '0', 'bar' => '1') * ENV.include?('foo') # => true @@ -6017,10 +6109,8 @@ env_empty_p(VALUE _) static VALUE env_has_key(VALUE env, VALUE key) { - const char *s; - - s = env_name(key); - return RBOOL(getenv(s)); + const char *s = env_name(key); + return RBOOL(has_env_with_lock(s)); } /* @@ -6046,12 +6136,15 @@ env_has_key(VALUE env, VALUE key) static VALUE env_assoc(VALUE env, VALUE key) { - const char *s, *e; + const char *s = env_name(key); + VALUE e = getenv_with_lock(s); - s = env_name(key); - e = getenv(s); - if (e) return rb_assoc_new(key, env_str_new2(e)); - return Qnil; + if (!NIL_P(e)) { + return rb_assoc_new(key, e); + } + else { + return Qnil; + } } /* @@ -6069,24 +6162,30 @@ env_assoc(VALUE env, VALUE key) static VALUE env_has_value(VALUE dmy, VALUE obj) { - char **env; - obj = rb_check_string_type(obj); if (NIL_P(obj)) return Qnil; - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s++) { - long len = strlen(s); - if (RSTRING_LEN(obj) == len && strncmp(s, RSTRING_PTR(obj), len) == 0) { - FREE_ENVIRON(environ); - return Qtrue; - } - } - env++; + + VALUE ret = Qfalse; + + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + while (*env) { + char *s = strchr(*env, '='); + if (s++) { + long len = strlen(s); + if (RSTRING_LEN(obj) == len && strncmp(s, RSTRING_PTR(obj), len) == 0) { + ret = Qtrue; + break; + } + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); - return Qfalse; + ENV_UNLOCK(); + + return ret; } /* @@ -6099,32 +6198,39 @@ env_has_value(VALUE dmy, VALUE obj) * ENV.replace('foo' => '0', 'bar' => '0') * ENV.rassoc('0') # => ["bar", "0"] * The order in which environment variables are examined is OS-dependent. - * See {About Ordering}[#class-ENV-label-About+Ordering]. + * See {About Ordering}[rdoc-ref:ENV@About+Ordering]. * * Returns +nil+ if there is no such environment variable. */ static VALUE env_rassoc(VALUE dmy, VALUE obj) { - char **env; - obj = rb_check_string_type(obj); if (NIL_P(obj)) return Qnil; - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s++) { - long len = strlen(s); - if (RSTRING_LEN(obj) == len && strncmp(s, RSTRING_PTR(obj), len) == 0) { - VALUE result = rb_assoc_new(rb_str_new(*env, s-*env-1), obj); - FREE_ENVIRON(environ); - return result; - } - } - env++; + + VALUE result = Qnil; + + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + + while (*env) { + const char *p = *env; + char *s = strchr(p, '='); + if (s++) { + long len = strlen(s); + if (RSTRING_LEN(obj) == len && strncmp(s, RSTRING_PTR(obj), len) == 0) { + result = rb_assoc_new(rb_str_new(p, s-p-1), obj); + break; + } + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); - return Qnil; + ENV_UNLOCK(); + + return result; } /* @@ -6135,55 +6241,61 @@ env_rassoc(VALUE dmy, VALUE obj) * ENV.replace('foo' => '0', 'bar' => '0') * ENV.key('0') # => "foo" * The order in which environment variables are examined is OS-dependent. - * See {About Ordering}[#class-ENV-label-About+Ordering]. + * See {About Ordering}[rdoc-ref:ENV@About+Ordering]. * * Returns +nil+ if there is no such value. * * Raises an exception if +value+ is invalid: * ENV.key(Object.new) # raises TypeError (no implicit conversion of Object into String) - * See {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]. + * See {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]. */ static VALUE env_key(VALUE dmy, VALUE value) { - char **env; - VALUE str; - - SafeStringValue(value); - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s++) { - long len = strlen(s); - if (RSTRING_LEN(value) == len && strncmp(s, RSTRING_PTR(value), len) == 0) { - str = env_str_new(*env, s-*env-1); - FREE_ENVIRON(environ); - return str; - } - } - env++; + StringValue(value); + VALUE str = Qnil; + + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + while (*env) { + char *s = strchr(*env, '='); + if (s++) { + long len = strlen(s); + if (RSTRING_LEN(value) == len && strncmp(s, RSTRING_PTR(value), len) == 0) { + str = env_str_new(*env, s-*env-1); + break; + } + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); - return Qnil; + ENV_UNLOCK(); + + return str; } static VALUE env_to_hash(void) { - char **env; - VALUE hash; + VALUE hash = rb_hash_new(); - hash = rb_hash_new(); - env = GET_ENVIRON(environ); - while (*env) { - char *s = strchr(*env, '='); - if (s) { - rb_hash_aset(hash, env_str_new(*env, s-*env), - env_str_new2(s+1)); - } - env++; + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + while (*env) { + char *s = strchr(*env, '='); + if (s) { + rb_hash_aset(hash, env_str_new(*env, s-*env), + env_str_new2(s+1)); + } + env++; + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); + return hash; } @@ -6312,27 +6424,36 @@ env_freeze(VALUE self) * ENV.shift # => ['bar', '1'] * ENV.to_hash # => {'foo' => '0'} * Exactly which environment variable is "first" is OS-dependent. - * See {About Ordering}[#class-ENV-label-About+Ordering]. + * See {About Ordering}[rdoc-ref:ENV@About+Ordering]. * * Returns +nil+ if the environment is empty. */ static VALUE env_shift(VALUE _) { - char **env; VALUE result = Qnil; + VALUE key = Qnil; - env = GET_ENVIRON(environ); - if (*env) { - char *s = strchr(*env, '='); - if (s) { - VALUE key = env_str_new(*env, s-*env); - VALUE val = env_str_new2(getenv(RSTRING_PTR(key))); - env_delete(key); - result = rb_assoc_new(key, val); - } + ENV_LOCK(); + { + char **env = GET_ENVIRON(environ); + if (*env) { + const char *p = *env; + char *s = strchr(p, '='); + if (s) { + key = env_str_new(p, s-p); + VALUE val = env_str_new2(getenv(RSTRING_PTR(key))); + result = rb_assoc_new(key, val); + } + } + FREE_ENVIRON(environ); } - FREE_ENVIRON(environ); + ENV_UNLOCK(); + + if (!NIL_P(key)) { + env_delete(key); + } + return result; } @@ -6349,7 +6470,7 @@ env_shift(VALUE _) * ENV.invert # => {"0"=>"foo"} * Note that the order of the ENV processing is OS-dependent, * which means that the order of overwriting is also OS-dependent. - * See {About Ordering}[#class-ENV-label-About+Ordering]. + * See {About Ordering}[rdoc-ref:ENV@About+Ordering]. */ static VALUE env_invert(VALUE _) @@ -6399,7 +6520,7 @@ env_replace_i(VALUE key, VALUE val, VALUE keys) * ENV.to_hash # => {"bar"=>"1", "foo"=>"0"} * * Raises an exception if a name or value is invalid - * (see {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]): + * (see {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]): * ENV.replace('foo' => '0', :bar => '1') # Raises TypeError (no implicit conversion of Symbol into String) * ENV.replace('foo' => '0', 'bar' => 1) # Raises TypeError (no implicit conversion of Integer into String) * ENV.to_hash # => {"bar"=>"1", "foo"=>"0"} @@ -6434,7 +6555,7 @@ env_update_block_i(VALUE key, VALUE val, VALUE _) { VALUE oldval = rb_f_getenv(Qnil, key); if (!NIL_P(oldval)) { - val = rb_yield_values(3, key, oldval, val); + val = rb_yield_values(3, key, oldval, val); } env_aset(key, val); return ST_CONTINUE; @@ -6442,12 +6563,12 @@ env_update_block_i(VALUE key, VALUE val, VALUE _) /* * call-seq: - * ENV.update(hash) -> ENV - * ENV.update(hash) { |name, env_val, hash_val| block } -> ENV - * ENV.merge!(hash) -> ENV - * ENV.merge!(hash) { |name, env_val, hash_val| block } -> ENV - * - * ENV.update is an alias for ENV.merge!. + * ENV.update -> ENV + * ENV.update(*hashes) -> ENV + * ENV.update(*hashes) { |name, env_val, hash_val| block } -> ENV + * ENV.merge! -> ENV + * ENV.merge!(*hashes) -> ENV + * ENV.merge!(*hashes) { |name, env_val, hash_val| block } -> ENV * * Adds to ENV each key/value pair in the given +hash+; returns ENV: * ENV.replace('foo' => '0', 'bar' => '1') @@ -6461,14 +6582,14 @@ env_update_block_i(VALUE key, VALUE val, VALUE _) * the block's return value becomes the new name: * ENV.merge!('foo' => '5') { |name, env_val, hash_val | env_val + hash_val } # => {"bar"=>"1", "foo"=>"45"} * Raises an exception if a name or value is invalid - * (see {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]); + * (see {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]); * ENV.replace('foo' => '0', 'bar' => '1') * ENV.merge!('foo' => '6', :bar => '7', 'baz' => '9') # Raises TypeError (no implicit conversion of Symbol into String) * ENV # => {"bar"=>"1", "foo"=>"6"} * ENV.merge!('foo' => '7', 'bar' => 8, 'baz' => '9') # Raises TypeError (no implicit conversion of Integer into String) * ENV # => {"bar"=>"1", "foo"=>"7"} * Raises an exception if the block returns an invalid name: - * (see {Invalid Names and Values}[#class-ENV-label-Invalid+Names+and+Values]): + * (see {Invalid Names and Values}[rdoc-ref:ENV@Invalid+Names+and+Values]): * ENV.merge!('bat' => '8', 'foo' => '9') { |name, env_val, hash_val | 10 } # Raises TypeError (no implicit conversion of Integer into String) * ENV # => {"bar"=>"1", "bat"=>"8", "foo"=>"7"} * @@ -6477,41 +6598,39 @@ env_update_block_i(VALUE key, VALUE val, VALUE _) * those following are ignored. */ static VALUE -env_update(VALUE env, VALUE hash) +env_update(int argc, VALUE *argv, VALUE env) { - if (env == hash) return env; - hash = to_hash(hash); rb_foreach_func *func = rb_block_given_p() ? env_update_block_i : env_update_i; - rb_hash_foreach(hash, func, 0); + for (int i = 0; i < argc; ++i) { + VALUE hash = argv[i]; + if (env == hash) continue; + hash = to_hash(hash); + rb_hash_foreach(hash, func, 0); + } return env; } +NORETURN(static VALUE env_clone(int, VALUE *, VALUE)); /* * call-seq: - * ENV.clone(freeze: nil) -> ENV + * ENV.clone(freeze: nil) # raises TypeError * - * Returns ENV itself, and warns because ENV is a wrapper for the - * process-wide environment variables and a clone is useless. - * If +freeze+ keyword is given and not +nil+ or +false+, raises ArgumentError. - * If +freeze+ keyword is given and +true+, raises TypeError, as ENV storage - * cannot be frozen. + * Raises TypeError, because ENV is a wrapper for the process-wide + * environment variables and a clone is useless. + * Use #to_h to get a copy of ENV data as a hash. */ static VALUE env_clone(int argc, VALUE *argv, VALUE obj) { if (argc) { - VALUE opt, kwfreeze; + VALUE opt; if (rb_scan_args(argc, argv, "0:", &opt) < argc) { - kwfreeze = rb_get_freeze_opt(1, &opt); - if (RTEST(kwfreeze)) { - rb_raise(rb_eTypeError, "cannot freeze ENV"); - } + rb_get_freeze_opt(1, &opt); } } - rb_warn_deprecated("ENV.clone", "ENV.to_h"); - return envtbl; + rb_raise(rb_eTypeError, "Cannot clone ENV, use ENV.to_h to get a copy of ENV as a hash"); } NORETURN(static VALUE env_dup(VALUE)); @@ -6528,28 +6647,39 @@ env_dup(VALUE obj) rb_raise(rb_eTypeError, "Cannot dup ENV, use ENV.to_h to get a copy of ENV as a hash"); } +static const rb_data_type_t env_data_type = { + "ENV", + { + NULL, + NULL, + NULL, + NULL, + }, + 0, 0, RUBY_TYPED_FREE_IMMEDIATELY | RUBY_TYPED_WB_PROTECTED, +}; + /* - * A \Hash maps each of its unique keys to a specific value. + * A +Hash+ maps each of its unique keys to a specific value. * - * A \Hash has certain similarities to an \Array, but: - * - An \Array index is always an \Integer. - * - A \Hash key can be (almost) any object. + * A +Hash+ has certain similarities to an Array, but: + * - An Array index is always an Integer. + * - A +Hash+ key can be (almost) any object. * - * === \Hash \Data Syntax + * === +Hash+ \Data Syntax * - * The older syntax for \Hash data uses the "hash rocket," <tt>=></tt>: + * The older syntax for +Hash+ data uses the "hash rocket," <tt>=></tt>: * * h = {:foo => 0, :bar => 1, :baz => 2} * h # => {:foo=>0, :bar=>1, :baz=>2} * - * Alternatively, but only for a \Hash key that's a \Symbol, + * Alternatively, but only for a +Hash+ key that's a Symbol, * you can use a newer JSON-style syntax, - * where each bareword becomes a \Symbol: + * where each bareword becomes a Symbol: * * h = {foo: 0, bar: 1, baz: 2} * h # => {:foo=>0, :bar=>1, :baz=>2} * - * You can also use a \String in place of a bareword: + * You can also use a String in place of a bareword: * * h = {'foo': 0, 'bar': 1, 'baz': 2} * h # => {:foo=>0, :bar=>1, :baz=>2} @@ -6565,26 +6695,34 @@ env_dup(VALUE obj) * # Raises SyntaxError (syntax error, unexpected ':', expecting =>): * h = {0: 'zero'} * + * +Hash+ value can be omitted, meaning that value will be fetched from the context + * by the name of the key: + * + * x = 0 + * y = 100 + * h = {x:, y:} + * h # => {:x=>0, :y=>100} + * * === Common Uses * - * You can use a \Hash to give names to objects: + * You can use a +Hash+ to give names to objects: * * person = {name: 'Matz', language: 'Ruby'} * person # => {:name=>"Matz", :language=>"Ruby"} * - * You can use a \Hash to give names to method arguments: + * You can use a +Hash+ to give names to method arguments: * * def some_method(hash) * p hash * end * some_method({foo: 0, bar: 1, baz: 2}) # => {:foo=>0, :bar=>1, :baz=>2} * - * Note: when the last argument in a method call is a \Hash, + * Note: when the last argument in a method call is a +Hash+, * the curly braces may be omitted: * * some_method(foo: 0, bar: 1, baz: 2) # => {:foo=>0, :bar=>1, :baz=>2} * - * You can use a \Hash to initialize an object: + * You can use a +Hash+ to initialize an object: * * class Dev * attr_accessor :name, :language @@ -6596,61 +6734,57 @@ env_dup(VALUE obj) * matz = Dev.new(name: 'Matz', language: 'Ruby') * matz # => #<Dev: @name="Matz", @language="Ruby"> * - * === Creating a \Hash + * === Creating a +Hash+ * - * Here are three ways to create a \Hash: + * You can create a +Hash+ object explicitly with: * - * - \Method <tt>Hash.new</tt> - * - \Method <tt>Hash[]</tt> - * - Literal form: <tt>{}</tt>. + * - A {hash literal}[rdoc-ref:syntax/literals.rdoc@Hash+Literals]. * - * --- + * You can convert certain objects to Hashes with: * - * You can create a \Hash by calling method Hash.new. + * - \Method #Hash. * - * Create an empty Hash: + * You can create a +Hash+ by calling method Hash.new. + * + * Create an empty +Hash+: * * h = Hash.new * h # => {} * h.class # => Hash * - * --- + * You can create a +Hash+ by calling method Hash.[]. * - * You can create a \Hash by calling method Hash.[]. - * - * Create an empty Hash: + * Create an empty +Hash+: * * h = Hash[] * h # => {} * - * Create a \Hash with initial entries: + * Create a +Hash+ with initial entries: * * h = Hash[foo: 0, bar: 1, baz: 2] * h # => {:foo=>0, :bar=>1, :baz=>2} * - * --- - * - * You can create a \Hash by using its literal form (curly braces). + * You can create a +Hash+ by using its literal form (curly braces). * - * Create an empty \Hash: + * Create an empty +Hash+: * * h = {} * h # => {} * - * Create a \Hash with initial entries: + * Create a +Hash+ with initial entries: * * h = {foo: 0, bar: 1, baz: 2} * h # => {:foo=>0, :bar=>1, :baz=>2} * * - * === \Hash Value Basics + * === +Hash+ Value Basics * - * The simplest way to retrieve a \Hash value (instance method #[]): + * The simplest way to retrieve a +Hash+ value (instance method #[]): * * h = {foo: 0, bar: 1, baz: 2} * h[:foo] # => 0 * - * The simplest way to create or update a \Hash value (instance method #[]=): + * The simplest way to create or update a +Hash+ value (instance method #[]=): * * h = {foo: 0, bar: 1, baz: 2} * h[:bat] = 3 # => 3 @@ -6658,7 +6792,7 @@ env_dup(VALUE obj) * h[:foo] = 4 # => 4 * h # => {:foo=>4, :bar=>1, :baz=>2, :bat=>3} * - * The simplest way to delete a \Hash entry (instance method #delete): + * The simplest way to delete a +Hash+ entry (instance method #delete): * * h = {foo: 0, bar: 1, baz: 2} * h.delete(:bar) # => 1 @@ -6666,13 +6800,13 @@ env_dup(VALUE obj) * * === Entry Order * - * A \Hash object presents its entries in the order of their creation. This is seen in: + * A +Hash+ object presents its entries in the order of their creation. This is seen in: * * - Iterative methods such as <tt>each</tt>, <tt>each_key</tt>, <tt>each_pair</tt>, <tt>each_value</tt>. * - Other order-sensitive methods such as <tt>shift</tt>, <tt>keys</tt>, <tt>values</tt>. - * - The \String returned by method <tt>inspect</tt>. + * - The String returned by method <tt>inspect</tt>. * - * A new \Hash has its initial ordering per the given entries: + * A new +Hash+ has its initial ordering per the given entries: * * h = Hash[foo: 0, bar: 1] * h # => {:foo=>0, :bar=>1} @@ -6693,18 +6827,18 @@ env_dup(VALUE obj) * h[:foo] = 5 * h # => {:bar=>1, :baz=>3, :foo=>5} * - * === \Hash Keys + * === +Hash+ Keys * - * ==== \Hash Key Equivalence + * ==== +Hash+ Key Equivalence * * Two objects are treated as the same \hash key when their <code>hash</code> value * is identical and the two objects are <code>eql?</code> to each other. * - * ==== Modifying an Active \Hash Key + * ==== Modifying an Active +Hash+ Key * - * Modifying a \Hash key while it is in use damages the hash's index. + * Modifying a +Hash+ key while it is in use damages the hash's index. * - * This \Hash has keys that are Arrays: + * This +Hash+ has keys that are Arrays: * * a0 = [ :foo, :bar ] * a1 = [ :baz, :bat ] @@ -6718,7 +6852,7 @@ env_dup(VALUE obj) * a0[0] = :bam * a0.hash # => 1069447059 * - * And damages the \Hash index: + * And damages the +Hash+ index: * * h.include?(a0) # => false * h[a0] # => nil @@ -6729,9 +6863,9 @@ env_dup(VALUE obj) * h.include?(a0) # => true * h[a0] # => 0 * - * A \String key is always safe. - * That's because an unfrozen \String - * passed as a key will be replaced by a duplicated and frozen \String: + * A String key is always safe. + * That's because an unfrozen String + * passed as a key will be replaced by a duplicated and frozen String: * * s = 'foo' * s.frozen? # => false @@ -6739,15 +6873,15 @@ env_dup(VALUE obj) * first_key = h.keys.first * first_key.frozen? # => true * - * ==== User-Defined \Hash Keys + * ==== User-Defined +Hash+ Keys * - * To be useable as a \Hash key, objects must implement the methods <code>hash</code> and <code>eql?</code>. - * Note: this requirement does not apply if the \Hash uses #compare_by_identity since comparison will then + * To be useable as a +Hash+ key, objects must implement the methods <code>hash</code> and <code>eql?</code>. + * Note: this requirement does not apply if the +Hash+ uses #compare_by_identity since comparison will then * rely on the keys' object id instead of <code>hash</code> and <code>eql?</code>. * - * \Object defines basic implementation for <code>hash</code> and <code>eq?</code> that makes each object + * Object defines basic implementation for <code>hash</code> and <code>eq?</code> that makes each object * a distinct key. Typically, user-defined classes will want to override these methods to provide meaningful - * behavior, or for example inherit \Struct that has useful definitions for these. + * behavior, or for example inherit Struct that has useful definitions for these. * * A typical implementation of <code>hash</code> is based on the * object's data while <code>eql?</code> is usually aliased to the overridden @@ -6770,7 +6904,7 @@ env_dup(VALUE obj) * alias eql? == * * def hash - * @author.hash ^ @title.hash # XOR + * [self.class, @author, @title].hash * end * end * @@ -6828,9 +6962,9 @@ env_dup(VALUE obj) * * To use a mutable object as default, it is recommended to use a default proc * - * ==== Default \Proc + * ==== Default Proc * - * When the default proc for a \Hash is set (i.e., not +nil+), + * When the default proc for a +Hash+ is set (i.e., not +nil+), * the default value returned by method #[] is determined by the default proc alone. * * You can retrieve the default proc with method #default_proc: @@ -6848,7 +6982,7 @@ env_dup(VALUE obj) * * When the default proc is set (i.e., not +nil+) * and method #[] is called with with a non-existent key, - * #[] calls the default proc with both the \Hash object itself and the missing key, + * #[] calls the default proc with both the +Hash+ object itself and the missing key, * then returns the proc's return value: * * h = Hash.new { |hash, key| "Default value for #{key}" } @@ -6868,133 +7002,137 @@ env_dup(VALUE obj) * * Note that setting the default proc will clear the default value and vice versa. * + * Be aware that a default proc that modifies the hash is not thread-safe in the + * sense that multiple threads can call into the default proc concurrently for the + * same key. + * * === What's Here * - * First, what's elsewhere. \Class \Hash: + * First, what's elsewhere. \Class +Hash+: * - * - Inherits from {class Object}[Object.html#class-Object-label-What-27s+Here]. - * - Includes {module Enumerable}[Enumerable.html#module-Enumerable-label-What-27s+Here], + * - Inherits from {class Object}[rdoc-ref:Object@What-27s+Here]. + * - Includes {module Enumerable}[rdoc-ref:Enumerable@What-27s+Here], * which provides dozens of additional methods. * - * Here, class \Hash provides methods that are useful for: + * Here, class +Hash+ provides methods that are useful for: * - * - {Creating a Hash}[#class-Hash-label-Methods+for+Creating+a+Hash] - * - {Setting Hash State}[#class-Hash-label-Methods+for+Setting+Hash+State] - * - {Querying}[#class-Hash-label-Methods+for+Querying] - * - {Comparing}[#class-Hash-label-Methods+for+Comparing] - * - {Fetching}[#class-Hash-label-Methods+for+Fetching] - * - {Assigning}[#class-Hash-label-Methods+for+Assigning] - * - {Deleting}[#class-Hash-label-Methods+for+Deleting] - * - {Iterating}[#class-Hash-label-Methods+for+Iterating] - * - {Converting}[#class-Hash-label-Methods+for+Converting] - * - {Transforming Keys and Values}[#class-Hash-label-Methods+for+Transforming+Keys+and+Values] - * - {And more....}[#class-Hash-label-Other+Methods] + * - {Creating a Hash}[rdoc-ref:Hash@Methods+for+Creating+a+Hash] + * - {Setting Hash State}[rdoc-ref:Hash@Methods+for+Setting+Hash+State] + * - {Querying}[rdoc-ref:Hash@Methods+for+Querying] + * - {Comparing}[rdoc-ref:Hash@Methods+for+Comparing] + * - {Fetching}[rdoc-ref:Hash@Methods+for+Fetching] + * - {Assigning}[rdoc-ref:Hash@Methods+for+Assigning] + * - {Deleting}[rdoc-ref:Hash@Methods+for+Deleting] + * - {Iterating}[rdoc-ref:Hash@Methods+for+Iterating] + * - {Converting}[rdoc-ref:Hash@Methods+for+Converting] + * - {Transforming Keys and Values}[rdoc-ref:Hash@Methods+for+Transforming+Keys+and+Values] + * - {And more....}[rdoc-ref:Hash@Other+Methods] * - * \Class \Hash also includes methods from module Enumerable. + * \Class +Hash+ also includes methods from module Enumerable. * - * ==== Methods for Creating a \Hash + * ==== Methods for Creating a +Hash+ * - * ::[]:: Returns a new hash populated with given objects. - * ::new:: Returns a new empty hash. - * ::try_convert:: Returns a new hash created from a given object. + * - ::[]: Returns a new hash populated with given objects. + * - ::new: Returns a new empty hash. + * - ::try_convert: Returns a new hash created from a given object. * - * ==== Methods for Setting \Hash State + * ==== Methods for Setting +Hash+ State * - * #compare_by_identity:: Sets +self+ to consider only identity in comparing keys. - * #default=:: Sets the default to a given value. - * #default_proc=:: Sets the default proc to a given proc. - * #rehash:: Rebuilds the hash table by recomputing the hash index for each key. + * - #compare_by_identity: Sets +self+ to consider only identity in comparing keys. + * - #default=: Sets the default to a given value. + * - #default_proc=: Sets the default proc to a given proc. + * - #rehash: Rebuilds the hash table by recomputing the hash index for each key. * * ==== Methods for Querying * - * #any?:: Returns whether any element satisfies a given criterion. - * #compare_by_identity?:: Returns whether the hash considers only identity when comparing keys. - * #default:: Returns the default value, or the default value for a given key. - * #default_proc:: Returns the default proc. - * #empty?:: Returns whether there are no entries. - * #eql?:: Returns whether a given object is equal to +self+. - * #hash:: Returns the integer hash code. - * #has_value?:: Returns whether a given object is a value in +self+. - * #include?, #has_key?, #member?, #key?:: Returns whether a given object is a key in +self+. - * #length, #size:: Returns the count of entries. - * #value?:: Returns whether a given object is a value in +self+. + * - #any?: Returns whether any element satisfies a given criterion. + * - #compare_by_identity?: Returns whether the hash considers only identity when comparing keys. + * - #default: Returns the default value, or the default value for a given key. + * - #default_proc: Returns the default proc. + * - #empty?: Returns whether there are no entries. + * - #eql?: Returns whether a given object is equal to +self+. + * - #hash: Returns the integer hash code. + * - #has_value?: Returns whether a given object is a value in +self+. + * - #include?, #has_key?, #member?, #key?: Returns whether a given object is a key in +self+. + * - #length, #size: Returns the count of entries. + * - #value?: Returns whether a given object is a value in +self+. * * ==== Methods for Comparing * - * {#<}[#method-i-3C]:: Returns whether +self+ is a proper subset of a given object. - * {#<=}[#method-i-3C-3D]:: Returns whether +self+ is a subset of a given object. - * {#==}[#method-i-3D-3D]:: Returns whether a given object is equal to +self+. - * {#>}[#method-i-3E]:: Returns whether +self+ is a proper superset of a given object - * {#>=}[#method-i-3E-3D]:: Returns whether +self+ is a proper superset of a given object. + * - #<: Returns whether +self+ is a proper subset of a given object. + * - #<=: Returns whether +self+ is a subset of a given object. + * - #==: Returns whether a given object is equal to +self+. + * - #>: Returns whether +self+ is a proper superset of a given object + * - #>=: Returns whether +self+ is a superset of a given object. * * ==== Methods for Fetching * - * #[]:: Returns the value associated with a given key. - * #assoc:: Returns a 2-element array containing a given key and its value. - * #dig:: Returns the object in nested objects that is specified - * by a given key and additional arguments. - * #fetch:: Returns the value for a given key. - * #fetch_values:: Returns array containing the values associated with given keys. - * #key:: Returns the key for the first-found entry with a given value. - * #keys:: Returns an array containing all keys in +self+. - * #rassoc:: Returns a 2-element array consisting of the key and value - of the first-found entry having a given value. - * #values:: Returns an array containing all values in +self+/ - * #values_at:: Returns an array containing values for given keys. + * - #[]: Returns the value associated with a given key. + * - #assoc: Returns a 2-element array containing a given key and its value. + * - #dig: Returns the object in nested objects that is specified + * by a given key and additional arguments. + * - #fetch: Returns the value for a given key. + * - #fetch_values: Returns array containing the values associated with given keys. + * - #key: Returns the key for the first-found entry with a given value. + * - #keys: Returns an array containing all keys in +self+. + * - #rassoc: Returns a 2-element array consisting of the key and value + * of the first-found entry having a given value. + * - #values: Returns an array containing all values in +self+/ + * - #values_at: Returns an array containing values for given keys. * * ==== Methods for Assigning * - * #[]=, #store:: Associates a given key with a given value. - * #merge:: Returns the hash formed by merging each given hash into a copy of +self+. - * #merge!, #update:: Merges each given hash into +self+. - * #replace:: Replaces the entire contents of +self+ with the contents of a givan hash. + * - #[]=, #store: Associates a given key with a given value. + * - #merge: Returns the hash formed by merging each given hash into a copy of +self+. + * - #merge!, #update: Merges each given hash into +self+. + * - #replace: Replaces the entire contents of +self+ with the contents of a given hash. * * ==== Methods for Deleting * * These methods remove entries from +self+: * - * #clear:: Removes all entries from +self+. - * #compact!:: Removes all +nil+-valued entries from +self+. - * #delete:: Removes the entry for a given key. - * #delete_if:: Removes entries selected by a given block. - * #filter!, #select!:: Keep only those entries selected by a given block. - * #keep_if:: Keep only those entries selected by a given block. - * #reject!:: Removes entries selected by a given block. - * #shift:: Removes and returns the first entry. + * - #clear: Removes all entries from +self+. + * - #compact!: Removes all +nil+-valued entries from +self+. + * - #delete: Removes the entry for a given key. + * - #delete_if: Removes entries selected by a given block. + * - #filter!, #select!: Keep only those entries selected by a given block. + * - #keep_if: Keep only those entries selected by a given block. + * - #reject!: Removes entries selected by a given block. + * - #shift: Removes and returns the first entry. * * These methods return a copy of +self+ with some entries removed: * - * #compact:: Returns a copy of +self+ with all +nil+-valued entries removed. - * #except:: Returns a copy of +self+ with entries removed for specified keys. - * #filter, #select:: Returns a copy of +self+ with only those entries selected by a given block. - * #reject:: Returns a copy of +self+ with entries removed as specified by a given block. - * #slice:: Returns a hash containing the entries for given keys. + * - #compact: Returns a copy of +self+ with all +nil+-valued entries removed. + * - #except: Returns a copy of +self+ with entries removed for specified keys. + * - #filter, #select: Returns a copy of +self+ with only those entries selected by a given block. + * - #reject: Returns a copy of +self+ with entries removed as specified by a given block. + * - #slice: Returns a hash containing the entries for given keys. * * ==== Methods for Iterating - * #each, #each_pair:: Calls a given block with each key-value pair. - * #each_key:: Calls a given block with each key. - * #each_value:: Calls a given block with each value. + * - #each, #each_pair: Calls a given block with each key-value pair. + * - #each_key: Calls a given block with each key. + * - #each_value: Calls a given block with each value. * * ==== Methods for Converting * - * #inspect, #to_s:: Returns a new String containing the hash entries. - * #to_a:: Returns a new array of 2-element arrays; - * each nested array contains a key-value pair from +self+. - * #to_h:: Returns +self+ if a \Hash; - * if a subclass of \Hash, returns a \Hash containing the entries from +self+. - * #to_hash:: Returns +self+. - * #to_proc:: Returns a proc that maps a given key to its value. + * - #inspect, #to_s: Returns a new String containing the hash entries. + * - #to_a: Returns a new array of 2-element arrays; + * each nested array contains a key-value pair from +self+. + * - #to_h: Returns +self+ if a +Hash+; + * if a subclass of +Hash+, returns a +Hash+ containing the entries from +self+. + * - #to_hash: Returns +self+. + * - #to_proc: Returns a proc that maps a given key to its value. * * ==== Methods for Transforming Keys and Values * - * #transform_keys:: Returns a copy of +self+ with modified keys. - * #transform_keys!:: Modifies keys in +self+ - * #transform_values:: Returns a copy of +self+ with modified values. - * #transform_values!:: Modifies values in +self+. + * - #transform_keys: Returns a copy of +self+ with modified keys. + * - #transform_keys!: Modifies keys in +self+ + * - #transform_values: Returns a copy of +self+ with modified values. + * - #transform_values!: Modifies values in +self+. * * ==== Other Methods - * #flatten:: Returns an array that is a 1-dimensional flattening of +self+. - * #invert:: Returns a hash with the each key-value pair inverted. + * - #flatten: Returns an array that is a 1-dimensional flattening of +self+. + * - #invert: Returns a hash with the each key-value pair inverted. * */ @@ -7002,7 +7140,6 @@ void Init_Hash(void) { id_hash = rb_intern_const("hash"); - id_default = rb_intern_const("default"); id_flatten_bang = rb_intern_const("flatten!"); id_hash_iter_lev = rb_make_internal_id(); @@ -7104,15 +7241,15 @@ Init_Hash(void) /* Document-class: ENV * - * ENV is a hash-like accessor for environment variables. + * +ENV+ is a hash-like accessor for environment variables. * * === Interaction with the Operating System * - * The ENV object interacts with the operating system's environment variables: + * The +ENV+ object interacts with the operating system's environment variables: * - * - When you get the value for a name in ENV, the value is retrieved from among the current environment variables. - * - When you create or set a name-value pair in ENV, the name and value are immediately set in the environment variables. - * - When you delete a name-value pair in ENV, it is immediately deleted from the environment variables. + * - When you get the value for a name in +ENV+, the value is retrieved from among the current environment variables. + * - When you create or set a name-value pair in +ENV+, the name and value are immediately set in the environment variables. + * - When you delete a name-value pair in +ENV+, it is immediately deleted from the environment variables. * * === Names and Values * @@ -7162,24 +7299,104 @@ Init_Hash(void) * * === About Ordering * - * ENV enumerates its name/value pairs in the order found + * +ENV+ enumerates its name/value pairs in the order found * in the operating system's environment variables. - * Therefore the ordering of ENV content is OS-dependent, and may be indeterminate. + * Therefore the ordering of +ENV+ content is OS-dependent, and may be indeterminate. * * This will be seen in: - * - A Hash returned by an ENV method. - * - An Enumerator returned by an ENV method. + * - A Hash returned by an +ENV+ method. + * - An Enumerator returned by an +ENV+ method. * - An Array returned by ENV.keys, ENV.values, or ENV.to_a. * - The String returned by ENV.inspect. * - The Array returned by ENV.shift. * - The name returned by ENV.key. * * === About the Examples - * Some methods in ENV return ENV itself. Typically, there are many environment variables. - * It's not useful to display a large ENV in the examples here, - * so most example snippets begin by resetting the contents of ENV: - * - ENV.replace replaces ENV with a new collection of entries. - * - ENV.clear empties ENV. + * Some methods in +ENV+ return +ENV+ itself. Typically, there are many environment variables. + * It's not useful to display a large +ENV+ in the examples here, + * so most example snippets begin by resetting the contents of +ENV+: + * - ENV.replace replaces +ENV+ with a new collection of entries. + * - ENV.clear empties +ENV+. + * + * === What's Here + * + * First, what's elsewhere. \Class +ENV+: + * + * - Inherits from {class Object}[rdoc-ref:Object@What-27s+Here]. + * - Extends {module Enumerable}[rdoc-ref:Enumerable@What-27s+Here], + * + * Here, class +ENV+ provides methods that are useful for: + * + * - {Querying}[rdoc-ref:ENV@Methods+for+Querying] + * - {Assigning}[rdoc-ref:ENV@Methods+for+Assigning] + * - {Deleting}[rdoc-ref:ENV@Methods+for+Deleting] + * - {Iterating}[rdoc-ref:ENV@Methods+for+Iterating] + * - {Converting}[rdoc-ref:ENV@Methods+for+Converting] + * - {And more ....}[rdoc-ref:ENV@More+Methods] + * + * ==== Methods for Querying + * + * - ::[]: Returns the value for the given environment variable name if it exists: + * - ::empty?: Returns whether +ENV+ is empty. + * - ::has_value?, ::value?: Returns whether the given value is in +ENV+. + * - ::include?, ::has_key?, ::key?, ::member?: Returns whether the given name + is in +ENV+. + * - ::key: Returns the name of the first entry with the given value. + * - ::size, ::length: Returns the number of entries. + * - ::value?: Returns whether any entry has the given value. + * + * ==== Methods for Assigning + * + * - ::[]=, ::store: Creates, updates, or deletes the named environment variable. + * - ::clear: Removes every environment variable; returns +ENV+: + * - ::update, ::merge!: Adds to +ENV+ each key/value pair in the given hash. + * - ::replace: Replaces the entire content of the +ENV+ + * with the name/value pairs in the given hash. + * + * ==== Methods for Deleting + * + * - ::delete: Deletes the named environment variable name if it exists. + * - ::delete_if: Deletes entries selected by the block. + * - ::keep_if: Deletes entries not selected by the block. + * - ::reject!: Similar to #delete_if, but returns +nil+ if no change was made. + * - ::select!, ::filter!: Deletes entries selected by the block. + * - ::shift: Removes and returns the first entry. + * + * ==== Methods for Iterating + * + * - ::each, ::each_pair: Calls the block with each name/value pair. + * - ::each_key: Calls the block with each name. + * - ::each_value: Calls the block with each value. + * + * ==== Methods for Converting + * + * - ::assoc: Returns a 2-element array containing the name and value + * of the named environment variable if it exists: + * - ::clone: Returns +ENV+ (and issues a warning). + * - ::except: Returns a hash of all name/value pairs except those given. + * - ::fetch: Returns the value for the given name. + * - ::inspect: Returns the contents of +ENV+ as a string. + * - ::invert: Returns a hash whose keys are the +ENV+ values, + and whose values are the corresponding +ENV+ names. + * - ::keys: Returns an array of all names. + * - ::rassoc: Returns the name and value of the first found entry + * that has the given value. + * - ::reject: Returns a hash of those entries not rejected by the block. + * - ::select, ::filter: Returns a hash of name/value pairs selected by the block. + * - ::slice: Returns a hash of the given names and their corresponding values. + * - ::to_a: Returns the entries as an array of 2-element Arrays. + * - ::to_h: Returns a hash of entries selected by the block. + * - ::to_hash: Returns a hash of all entries. + * - ::to_s: Returns the string <tt>'ENV'</tt>. + * - ::values: Returns all values as an array. + * - ::values_at: Returns an array of the values for the given name. + * + * ==== More Methods + * + * - ::dup: Raises an exception. + * - ::freeze: Raises an exception. + * - ::rehash: Returns +nil+, without modifying +ENV+. + * */ /* @@ -7187,8 +7404,10 @@ Init_Hash(void) * envtbl = rb_define_class("ENV", rb_cObject); */ origenviron = environ; - envtbl = rb_obj_alloc(rb_cObject); + envtbl = TypedData_Wrap_Struct(rb_cObject, &env_data_type, NULL); rb_extend_object(envtbl, rb_mEnumerable); + FL_SET_RAW(envtbl, RUBY_FL_SHAREABLE); + rb_define_singleton_method(envtbl, "[]", rb_f_getenv, 1); rb_define_singleton_method(envtbl, "fetch", env_fetch, -1); @@ -7214,8 +7433,8 @@ Init_Hash(void) rb_define_singleton_method(envtbl, "freeze", env_freeze, 0); rb_define_singleton_method(envtbl, "invert", env_invert, 0); rb_define_singleton_method(envtbl, "replace", env_replace, 1); - rb_define_singleton_method(envtbl, "update", env_update, 1); - rb_define_singleton_method(envtbl, "merge!", env_update, 1); + rb_define_singleton_method(envtbl, "update", env_update, -1); + rb_define_singleton_method(envtbl, "merge!", env_update, -1); rb_define_singleton_method(envtbl, "inspect", env_inspect, 0); rb_define_singleton_method(envtbl, "rehash", env_none, 0); rb_define_singleton_method(envtbl, "to_a", env_to_a, 0); @@ -7247,7 +7466,7 @@ Init_Hash(void) rb_undef_method(envtbl_class, "initialize_dup"); /* - * ENV is a Hash-like accessor for environment variables. + * +ENV+ is a Hash-like accessor for environment variables. * * See ENV (the class) for more details. */ |