ruby/class.c

2244 строки
60 KiB
C
Исходник Обычный вид История

/**********************************************************************
class.c -
$Author$
created at: Tue Aug 10 15:05:44 JST 1993
* encoding.c: provide basic features for M17N. * parse.y: encoding aware parsing. * parse.y (pragma_encoding): encoding specification pragma. * parse.y (rb_intern3): encoding specified symbols. * string.c (rb_str_length): length based on characters. for older behavior, bytesize method added. * string.c (rb_str_index_m): index based on characters. rindex as well. * string.c (succ_char): encoding aware succeeding string. * string.c (rb_str_reverse): reverse based on characters. * string.c (rb_str_inspect): encoding aware string description. * string.c (rb_str_upcase_bang): encoding aware case conversion. downcase, capitalize, swapcase as well. * string.c (rb_str_tr_bang): tr based on characters. delete, squeeze, tr_s, count as well. * string.c (rb_str_split_m): split based on characters. * string.c (rb_str_each_line): encoding aware each_line. * string.c (rb_str_each_char): added. iteration based on characters. * string.c (rb_str_strip_bang): encoding aware whitespace stripping. lstrip, rstrip as well. * string.c (rb_str_justify): encoding aware justifying (ljust, rjust, center). * string.c (str_encoding): get encoding attribute from a string. * re.c (rb_reg_initialize): encoding aware regular expression * sprintf.c (rb_str_format): formatting (i.e. length count) based on characters. * io.c (rb_io_getc): getc to return one-character string. for older behavior, getbyte method added. * ext/stringio/stringio.c (strio_getc): ditto. * io.c (rb_io_ungetc): allow pushing arbitrary string at the current reading point. * ext/stringio/stringio.c (strio_ungetc): ditto. * ext/strscan/strscan.c: encoding support. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@13261 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2007-08-25 07:29:39 +04:00
Copyright (C) 1993-2007 Yukihiro Matsumoto
**********************************************************************/
/*!
* \addtogroup class
* \{
*/
#include "ruby/internal/config.h"
#include <ctype.h>
#include "constant.h"
Add a cache for class variables Redo of 34a2acdac788602c14bf05fb616215187badd504 and 931138b00696419945dc03e10f033b1f53cd50f3 which were reverted. GitHub PR #4340. This change implements a cache for class variables. Previously there was no cache for cvars. Cvar access is slow due to needing to travel all the way up th ancestor tree before returning the cvar value. The deeper the ancestor tree the slower cvar access will be. The benefits of the cache are more visible with a higher number of included modules due to the way Ruby looks up class variables. The benchmark here includes 26 modules and shows with the cache, this branch is 6.5x faster when accessing class variables. ``` compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105c) [x86_64-darwin19] built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be009) [x86_64-darwin19] | |compare-ruby|built-ruby| |:--------|-----------:|---------:| |vm_cvar | 5.681M| 36.980M| | | -| 6.51x| ``` Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails application. ActiveRecord::Base.logger has 71 ancestors. The more ancestors a tree has, the more clear the speed increase. IE if Base had only one ancestor we'd see no improvement. This benchmark is run on a vanilla Rails application. Benchmark code: ```ruby require "benchmark/ips" require_relative "config/environment" Benchmark.ips do |x| x.report "logger" do ActiveRecord::Base.logger end end ``` Ruby 3.0 master / Rails 6.1: ``` Warming up -------------------------------------- logger 155.251k i/100ms Calculating ------------------------------------- ``` Ruby 3.0 with cvar cache / Rails 6.1: ``` Warming up -------------------------------------- logger 1.546M i/100ms Calculating ------------------------------------- logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s ``` Lastly we ran a benchmark to demonstate the difference between master and our cache when the number of modules increases. This benchmark measures 1 ancestor, 30 ancestors, and 100 ancestors. Ruby 3.0 master: ``` Warming up -------------------------------------- 1 module 1.231M i/100ms 30 modules 432.020k i/100ms 100 modules 145.399k i/100ms Calculating ------------------------------------- 1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s 30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s 100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s Comparison: 1 module: 12209958.3 i/s 30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower 100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower ``` Ruby 3.0 with cvar cache: ``` Warming up -------------------------------------- 1 module 1.641M i/100ms 30 modules 1.655M i/100ms 100 modules 1.620M i/100ms Calculating ------------------------------------- 1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s 30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s 100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s Comparison: 1 module: 16279458.0 i/s 100 modules: 16087484.6 i/s - same-ish: difference falls within error 30 modules: 15891406.2 i/s - same-ish: difference falls within error ``` Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
2021-06-01 20:34:06 +03:00
#include "debug_counter.h"
#include "id_table.h"
#include "internal.h"
#include "internal/class.h"
#include "internal/eval.h"
#include "internal/hash.h"
#include "internal/object.h"
#include "internal/string.h"
#include "internal/variable.h"
#include "ruby/st.h"
#include "vm_core.h"
#define id_attached id__attached__
#define METACLASS_OF(k) RBASIC(k)->klass
#define SET_METACLASS_OF(k, cls) RBASIC_SET_CLASS(k, cls)
Add a cache for class variables Redo of 34a2acdac788602c14bf05fb616215187badd504 and 931138b00696419945dc03e10f033b1f53cd50f3 which were reverted. GitHub PR #4340. This change implements a cache for class variables. Previously there was no cache for cvars. Cvar access is slow due to needing to travel all the way up th ancestor tree before returning the cvar value. The deeper the ancestor tree the slower cvar access will be. The benefits of the cache are more visible with a higher number of included modules due to the way Ruby looks up class variables. The benchmark here includes 26 modules and shows with the cache, this branch is 6.5x faster when accessing class variables. ``` compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105c) [x86_64-darwin19] built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be009) [x86_64-darwin19] | |compare-ruby|built-ruby| |:--------|-----------:|---------:| |vm_cvar | 5.681M| 36.980M| | | -| 6.51x| ``` Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails application. ActiveRecord::Base.logger has 71 ancestors. The more ancestors a tree has, the more clear the speed increase. IE if Base had only one ancestor we'd see no improvement. This benchmark is run on a vanilla Rails application. Benchmark code: ```ruby require "benchmark/ips" require_relative "config/environment" Benchmark.ips do |x| x.report "logger" do ActiveRecord::Base.logger end end ``` Ruby 3.0 master / Rails 6.1: ``` Warming up -------------------------------------- logger 155.251k i/100ms Calculating ------------------------------------- ``` Ruby 3.0 with cvar cache / Rails 6.1: ``` Warming up -------------------------------------- logger 1.546M i/100ms Calculating ------------------------------------- logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s ``` Lastly we ran a benchmark to demonstate the difference between master and our cache when the number of modules increases. This benchmark measures 1 ancestor, 30 ancestors, and 100 ancestors. Ruby 3.0 master: ``` Warming up -------------------------------------- 1 module 1.231M i/100ms 30 modules 432.020k i/100ms 100 modules 145.399k i/100ms Calculating ------------------------------------- 1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s 30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s 100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s Comparison: 1 module: 12209958.3 i/s 30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower 100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower ``` Ruby 3.0 with cvar cache: ``` Warming up -------------------------------------- 1 module 1.641M i/100ms 30 modules 1.655M i/100ms 100 modules 1.620M i/100ms Calculating ------------------------------------- 1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s 30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s 100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s Comparison: 1 module: 16279458.0 i/s 100 modules: 16087484.6 i/s - same-ish: difference falls within error 30 modules: 15891406.2 i/s - same-ish: difference falls within error ``` Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
2021-06-01 20:34:06 +03:00
RUBY_EXTERN rb_serial_t ruby_vm_global_cvar_state;
void
rb_class_subclass_add(VALUE super, VALUE klass)
{
rb_subclass_entry_t *entry, *head;
if (super && super != Qundef) {
entry = ALLOC(rb_subclass_entry_t);
entry->klass = klass;
entry->next = NULL;
2021-01-26 19:49:57 +03:00
head = RCLASS_SUBCLASSES(super);
if (head) {
entry->next = head;
2021-01-26 19:06:31 +03:00
RCLASS_PARENT_SUBCLASSES(head->klass) = &entry->next;
}
2021-01-26 19:49:57 +03:00
RCLASS_SUBCLASSES(super) = entry;
RCLASS_PARENT_SUBCLASSES(klass) = &RCLASS_SUBCLASSES(super);
}
}
static void
rb_module_add_to_subclasses_list(VALUE module, VALUE iclass)
{
rb_subclass_entry_t *entry, *head;
entry = ALLOC(rb_subclass_entry_t);
entry->klass = iclass;
entry->next = NULL;
2021-01-26 19:49:57 +03:00
head = RCLASS_SUBCLASSES(module);
if (head) {
entry->next = head;
2021-01-26 19:12:29 +03:00
RCLASS_MODULE_SUBCLASSES(head->klass) = &entry->next;
}
2021-01-26 19:49:57 +03:00
RCLASS_SUBCLASSES(module) = entry;
RCLASS_MODULE_SUBCLASSES(iclass) = &RCLASS_SUBCLASSES(module);
}
void
rb_class_remove_from_super_subclasses(VALUE klass)
{
2021-08-20 08:03:08 +03:00
rb_subclass_entry_t **prev = RCLASS_PARENT_SUBCLASSES(klass);
2021-08-20 08:03:08 +03:00
if (prev) {
rb_subclass_entry_t *entry = *prev, *next = entry->next;
2021-08-20 08:03:08 +03:00
*prev = next;
if (next) {
RCLASS_PARENT_SUBCLASSES(next->klass) = prev;
}
xfree(entry);
}
2021-01-26 19:06:31 +03:00
RCLASS_PARENT_SUBCLASSES(klass) = NULL;
}
void
rb_class_remove_from_module_subclasses(VALUE klass)
{
2021-08-20 08:03:08 +03:00
rb_subclass_entry_t **prev = RCLASS_MODULE_SUBCLASSES(klass);
2021-08-20 08:03:08 +03:00
if (prev) {
rb_subclass_entry_t *entry = *prev, *next = entry->next;
2021-08-20 08:03:08 +03:00
*prev = next;
if (next) {
RCLASS_MODULE_SUBCLASSES(next->klass) = prev;
}
xfree(entry);
}
2021-01-26 19:12:29 +03:00
RCLASS_MODULE_SUBCLASSES(klass) = NULL;
}
void
rb_class_foreach_subclass(VALUE klass, void (*f)(VALUE, VALUE), VALUE arg)
{
2021-01-26 19:49:57 +03:00
rb_subclass_entry_t *cur = RCLASS_SUBCLASSES(klass);
/* do not be tempted to simplify this loop into a for loop, the order of
operations is important here if `f` modifies the linked list */
while (cur) {
VALUE curklass = cur->klass;
cur = cur->next;
f(curklass, arg);
}
}
static void
class_detach_subclasses(VALUE klass, VALUE arg)
{
rb_class_remove_from_super_subclasses(klass);
}
void
rb_class_detach_subclasses(VALUE klass)
{
rb_class_foreach_subclass(klass, class_detach_subclasses, Qnil);
}
static void
class_detach_module_subclasses(VALUE klass, VALUE arg)
{
rb_class_remove_from_module_subclasses(klass);
}
void
rb_class_detach_module_subclasses(VALUE klass)
{
rb_class_foreach_subclass(klass, class_detach_module_subclasses, Qnil);
}
/**
* Allocates a struct RClass for a new class.
*
* \param flags initial value for basic.flags of the returned class.
* \param klass the class of the returned class.
* \return an uninitialized Class object.
* \pre \p klass must refer \c Class class or an ancestor of Class.
* \pre \code (flags | T_CLASS) != 0 \endcode
* \post the returned class can safely be \c #initialize 'd.
*
* \note this function is not Class#allocate.
*/
static VALUE
class_alloc(VALUE flags, VALUE klass)
{
size_t payload_size = 0;
#if USE_RVARGC
payload_size = sizeof(rb_classext_t);
#endif
RVARGC_NEWOBJ_OF(obj, struct RClass, klass, (flags & T_MASK) | FL_PROMOTED1 /* start from age == 2 */ | (RGENGC_WB_PROTECTED_CLASS ? FL_WB_PROTECTED : 0), payload_size);
#if USE_RVARGC
obj->ptr = (rb_classext_t *)rb_gc_rvargc_object_data((VALUE)obj);
#else
obj->ptr = ZALLOC(rb_classext_t);
#endif
/* ZALLOC
RCLASS_IV_TBL(obj) = 0;
RCLASS_CONST_TBL(obj) = 0;
RCLASS_M_TBL(obj) = 0;
RCLASS_IV_INDEX_TBL(obj) = 0;
RCLASS_SET_SUPER((VALUE)obj, 0);
2021-01-26 19:49:57 +03:00
RCLASS_SUBCLASSES(obj) = NULL;
2021-01-26 19:06:31 +03:00
RCLASS_PARENT_SUBCLASSES(obj) = NULL;
2021-01-26 19:12:29 +03:00
RCLASS_MODULE_SUBCLASSES(obj) = NULL;
*/
RCLASS_SET_ORIGIN((VALUE)obj, (VALUE)obj);
RCLASS_SERIAL(obj) = rb_next_class_serial();
RB_OBJ_WRITE(obj, &RCLASS_REFINED_CLASS(obj), Qnil);
2021-01-26 19:29:09 +03:00
RCLASS_ALLOCATOR(obj) = 0;
return (VALUE)obj;
}
static void
RCLASS_M_TBL_INIT(VALUE c)
{
RCLASS_M_TBL(c) = rb_id_table_create(0);
}
/*!
* A utility function that wraps class_alloc.
*
* allocates a class and initializes safely.
* \param super a class from which the new class derives.
* \return a class object.
* \pre \a super must be a class.
* \post the metaclass of the new class is Class.
*/
VALUE
rb_class_boot(VALUE super)
{
VALUE klass = class_alloc(T_CLASS, rb_cClass);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RCLASS_SET_SUPER(klass, super);
RCLASS_M_TBL_INIT(klass);
return (VALUE)klass;
}
/*!
* Ensures a class can be derived from super.
*
* \param super a reference to an object.
* \exception TypeError if \a super is not a Class or \a super is a singleton class.
*/
void
rb_check_inheritable(VALUE super)
{
if (!RB_TYPE_P(super, T_CLASS)) {
rb_raise(rb_eTypeError, "superclass must be an instance of Class (given an instance of %"PRIsVALUE")",
rb_obj_class(super));
}
if (RBASIC(super)->flags & FL_SINGLETON) {
rb_raise(rb_eTypeError, "can't make subclass of singleton class");
}
if (super == rb_cClass) {
rb_raise(rb_eTypeError, "can't make subclass of Class");
}
}
/*!
* Creates a new class.
* \param super a class from which the new class derives.
* \exception TypeError \a super is not inheritable.
* \exception TypeError \a super is the Class class.
*/
VALUE
rb_class_new(VALUE super)
{
Check_Type(super, T_CLASS);
rb_check_inheritable(super);
return rb_class_boot(super);
}
static void
clone_method(VALUE old_klass, VALUE new_klass, ID mid, const rb_method_entry_t *me)
{
if (me->def->type == VM_METHOD_TYPE_ISEQ) {
rb_cref_t *new_cref;
rb_vm_rewrite_cref(me->def->body.iseq.cref, old_klass, new_klass, &new_cref);
2015-07-22 01:52:59 +03:00
rb_add_method_iseq(new_klass, mid, me->def->body.iseq.iseqptr, new_cref, METHOD_ENTRY_VISI(me));
}
else {
rb_method_entry_set(new_klass, mid, me, METHOD_ENTRY_VISI(me));
}
}
struct clone_method_arg {
VALUE new_klass;
VALUE old_klass;
};
static enum rb_id_table_iterator_result
clone_method_i(ID key, VALUE value, void *data)
{
const struct clone_method_arg *arg = (struct clone_method_arg *)data;
clone_method(arg->old_klass, arg->new_klass, key, (const rb_method_entry_t *)value);
return ID_TABLE_CONTINUE;
}
struct clone_const_arg {
VALUE klass;
struct rb_id_table *tbl;
};
static int
clone_const(ID key, const rb_const_entry_t *ce, struct clone_const_arg *arg)
{
rb_const_entry_t *nce = ALLOC(rb_const_entry_t);
MEMCPY(nce, ce, rb_const_entry_t, 1);
RB_OBJ_WRITTEN(arg->klass, Qundef, ce->value);
RB_OBJ_WRITTEN(arg->klass, Qundef, ce->file);
rb_id_table_insert(arg->tbl, key, (VALUE)nce);
return ID_TABLE_CONTINUE;
}
static enum rb_id_table_iterator_result
clone_const_i(ID key, VALUE value, void *data)
{
return clone_const(key, (const rb_const_entry_t *)value, data);
}
static void
class_init_copy_check(VALUE clone, VALUE orig)
{
if (orig == rb_cBasicObject) {
rb_raise(rb_eTypeError, "can't copy the root class");
}
if (RCLASS_SUPER(clone) != 0 || clone == rb_cBasicObject) {
rb_raise(rb_eTypeError, "already initialized class");
}
if (FL_TEST(orig, FL_SINGLETON)) {
rb_raise(rb_eTypeError, "can't copy singleton class");
}
}
2019-08-09 07:08:01 +03:00
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
static void
copy_tables(VALUE clone, VALUE orig)
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
{
if (RCLASS_IV_TBL(clone)) {
st_free_table(RCLASS_IV_TBL(clone));
RCLASS_IV_TBL(clone) = 0;
}
if (RCLASS_CONST_TBL(clone)) {
rb_free_const_table(RCLASS_CONST_TBL(clone));
RCLASS_CONST_TBL(clone) = 0;
}
* internal.h: remove struct method_table_wrapper. struct method_table_wrapper was introduced to avoid duplicate marking for method tables. For example, `module M1; def foo; end; end` make one method table (mtbl) contains a method `foo`. M1 (T_MODULE) points mtbl. Classes C1 and C2 includes M1, then two T_ICLASS objects are created and they points mtbl too. In this case, three objects (one T_MODULE and two T_ICLASS objects) points same mtbl. On marking phase, these three objects mark same mtbl. To avoid such duplication, struct method_table_wrapper was introduced. However, created two T_ICLASS objects have same or shorter lifetime than M1 (T_MODULE) object. So that we only need to mark mtbl from M1, not from T_ICLASS objects. This patch tries marking only from M1. Note that one `Module#prepend` call creates two T_ICLASS objects. One for refering to a prepending Module object, same as `Module#include`. We don't nedd to care this T_ICLASS. One for moving original mtbl from a prepending class. We need to mark such mtbl from this T_ICLASS object. To mark the mtbl, we need to use `RCLASS_ORIGIN(klass)` on marking from a prepended class `klass`. * class.c: ditto. * eval.c (rb_using_refinement): ditto. * gc.c: ditto. * include/ruby/ruby.h: define m_tbl directly. The definition of struct RClass should be moved to (srcdir)/internal.h. * method.h: remove decl of rb_free_m_tbl_wrapper(). * object.c: use RCLASS_M_TBL() directly. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49862 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-03-06 01:20:14 +03:00
RCLASS_M_TBL(clone) = 0;
if (RCLASS_IV_TBL(orig)) {
st_data_t id;
rb_iv_tbl_copy(clone, orig);
CONST_ID(id, "__tmp_classpath__");
st_delete(RCLASS_IV_TBL(clone), &id, 0);
CONST_ID(id, "__classpath__");
st_delete(RCLASS_IV_TBL(clone), &id, 0);
CONST_ID(id, "__classid__");
st_delete(RCLASS_IV_TBL(clone), &id, 0);
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
if (RCLASS_CONST_TBL(orig)) {
struct clone_const_arg arg;
arg.tbl = RCLASS_CONST_TBL(clone) = rb_id_table_create(0);
arg.klass = clone;
rb_id_table_foreach(RCLASS_CONST_TBL(orig), clone_const_i, &arg);
}
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
}
static bool ensure_origin(VALUE klass);
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
/* :nodoc: */
VALUE
rb_mod_init_copy(VALUE clone, VALUE orig)
{
if (RB_TYPE_P(clone, T_CLASS)) {
class_init_copy_check(clone, orig);
}
if (!OBJ_INIT_COPY(clone, orig)) return clone;
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
/* cloned flag is refer at constant inline cache
* see vm_get_const_key_cref() in vm_insnhelper.c
*/
FL_SET(clone, RCLASS_CLONED);
FL_SET(orig , RCLASS_CLONED);
if (!FL_TEST(CLASS_OF(clone), FL_SINGLETON)) {
RBASIC_SET_CLASS(clone, rb_singleton_class_clone(orig));
rb_singleton_class_attached(RBASIC(clone)->klass, (VALUE)clone);
}
2021-01-26 19:29:09 +03:00
RCLASS_ALLOCATOR(clone) = RCLASS_ALLOCATOR(orig);
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
copy_tables(clone, orig);
if (RCLASS_M_TBL(orig)) {
struct clone_method_arg arg;
arg.old_klass = orig;
arg.new_klass = clone;
RCLASS_M_TBL_INIT(clone);
rb_id_table_foreach(RCLASS_M_TBL(orig), clone_method_i, &arg);
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
if (RCLASS_ORIGIN(orig) == orig) {
RCLASS_SET_SUPER(clone, RCLASS_SUPER(orig));
}
else {
VALUE p = RCLASS_SUPER(orig);
VALUE orig_origin = RCLASS_ORIGIN(orig);
VALUE prev_clone_p = clone;
VALUE origin_stack = rb_ary_tmp_new(2);
VALUE origin[2];
VALUE clone_p = 0;
long origin_len;
int add_subclass;
VALUE clone_origin;
ensure_origin(clone);
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
clone_origin = RCLASS_ORIGIN(clone);
while (p && p != orig_origin) {
if (BUILTIN_TYPE(p) != T_ICLASS) {
rb_bug("non iclass between module/class and origin");
}
clone_p = class_alloc(RBASIC(p)->flags, RBASIC(p)->klass);
RCLASS_SET_SUPER(prev_clone_p, clone_p);
prev_clone_p = clone_p;
RCLASS_M_TBL(clone_p) = RCLASS_M_TBL(p);
RCLASS_CONST_TBL(clone_p) = RCLASS_CONST_TBL(p);
RCLASS_IV_TBL(clone_p) = RCLASS_IV_TBL(p);
2021-01-26 19:29:09 +03:00
RCLASS_ALLOCATOR(clone_p) = RCLASS_ALLOCATOR(p);
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
if (RB_TYPE_P(clone, T_CLASS)) {
RCLASS_SET_INCLUDER(clone_p, clone);
}
add_subclass = TRUE;
if (p != RCLASS_ORIGIN(p)) {
origin[0] = clone_p;
origin[1] = RCLASS_ORIGIN(p);
rb_ary_cat(origin_stack, origin, 2);
}
else if ((origin_len = RARRAY_LEN(origin_stack)) > 1 &&
RARRAY_AREF(origin_stack, origin_len - 1) == p) {
RCLASS_SET_ORIGIN(RARRAY_AREF(origin_stack, (origin_len -= 2)), clone_p);
RICLASS_SET_ORIGIN_SHARED_MTBL(clone_p);
rb_ary_resize(origin_stack, origin_len);
add_subclass = FALSE;
}
if (add_subclass) {
rb_module_add_to_subclasses_list(RBASIC(p)->klass, clone_p);
}
p = RCLASS_SUPER(p);
}
if (p == orig_origin) {
if (clone_p) {
RCLASS_SET_SUPER(clone_p, clone_origin);
RCLASS_SET_SUPER(clone_origin, RCLASS_SUPER(orig_origin));
}
copy_tables(clone_origin, orig_origin);
if (RCLASS_M_TBL(orig_origin)) {
struct clone_method_arg arg;
arg.old_klass = orig;
arg.new_klass = clone;
RCLASS_M_TBL_INIT(clone_origin);
rb_id_table_foreach(RCLASS_M_TBL(orig_origin), clone_method_i, &arg);
}
}
else {
rb_bug("no origin for class that has origin");
}
}
return clone;
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
VALUE
rb_singleton_class_clone(VALUE obj)
{
return rb_singleton_class_clone_and_attach(obj, Qundef);
}
// Clone and return the singleton class of `obj` if it has been created and is attached to `obj`.
VALUE
rb_singleton_class_clone_and_attach(VALUE obj, VALUE attach)
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
{
const VALUE klass = RBASIC(obj)->klass;
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
// Note that `rb_singleton_class()` can create situations where `klass` is
// attached to an object other than `obj`. In which case `obj` does not have
// a material singleton class attached yet and there is no singleton class
// to clone.
if (!(FL_TEST(klass, FL_SINGLETON) && rb_attr_get(klass, id_attached) == obj)) {
// nothing to clone
return klass;
}
else {
/* copy singleton(unnamed) class */
bool klass_of_clone_is_new;
VALUE clone = class_alloc(RBASIC(klass)->flags, 0);
if (BUILTIN_TYPE(obj) == T_CLASS) {
klass_of_clone_is_new = true;
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RBASIC_SET_CLASS(clone, clone);
}
else {
VALUE klass_metaclass_clone = rb_singleton_class_clone(klass);
// When `METACLASS_OF(klass) == klass_metaclass_clone`, it means the
// recursive call did not clone `METACLASS_OF(klass)`.
klass_of_clone_is_new = (METACLASS_OF(klass) != klass_metaclass_clone);
RBASIC_SET_CLASS(clone, klass_metaclass_clone);
}
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RCLASS_SET_SUPER(clone, RCLASS_SUPER(klass));
2021-01-26 19:29:09 +03:00
RCLASS_ALLOCATOR(clone) = RCLASS_ALLOCATOR(klass);
if (RCLASS_IV_TBL(klass)) {
rb_iv_tbl_copy(clone, klass);
}
if (RCLASS_CONST_TBL(klass)) {
struct clone_const_arg arg;
arg.tbl = RCLASS_CONST_TBL(clone) = rb_id_table_create(0);
arg.klass = clone;
rb_id_table_foreach(RCLASS_CONST_TBL(klass), clone_const_i, &arg);
}
if (attach != Qundef) {
rb_singleton_class_attached(clone, attach);
}
RCLASS_M_TBL_INIT(clone);
{
struct clone_method_arg arg;
arg.old_klass = klass;
arg.new_klass = clone;
rb_id_table_foreach(RCLASS_M_TBL(klass), clone_method_i, &arg);
}
if (klass_of_clone_is_new) {
rb_singleton_class_attached(RBASIC(clone)->klass, clone);
}
FL_SET(clone, FL_SINGLETON);
return clone;
}
}
void
rb_singleton_class_attached(VALUE klass, VALUE obj)
{
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
if (FL_TEST(klass, FL_SINGLETON)) {
rb_class_ivar_set(klass, id_attached, obj);
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
}
/*!
* whether k is a meta^(n)-class of Class class
* @retval 1 if \a k is a meta^(n)-class of Class class (n >= 0)
* @retval 0 otherwise
*/
#define META_CLASS_OF_CLASS_CLASS_P(k) (METACLASS_OF(k) == (k))
static int
rb_singleton_class_has_metaclass_p(VALUE sklass)
{
return rb_attr_get(METACLASS_OF(sklass), id_attached) == sklass;
}
int
rb_singleton_class_internal_p(VALUE sklass)
{
return (RB_TYPE_P(rb_attr_get(sklass, id_attached), T_CLASS) &&
!rb_singleton_class_has_metaclass_p(sklass));
}
/*!
* whether k has a metaclass
* @retval 1 if \a k has a metaclass
* @retval 0 otherwise
*/
#define HAVE_METACLASS_P(k) \
(FL_TEST(METACLASS_OF(k), FL_SINGLETON) && \
rb_singleton_class_has_metaclass_p(k))
/*!
* ensures \a klass belongs to its own eigenclass.
* @return the eigenclass of \a klass
* @post \a klass belongs to the returned eigenclass.
* i.e. the attached object of the eigenclass is \a klass.
* @note this macro creates a new eigenclass if necessary.
*/
#define ENSURE_EIGENCLASS(klass) \
(HAVE_METACLASS_P(klass) ? METACLASS_OF(klass) : make_metaclass(klass))
/*!
* Creates a metaclass of \a klass
* \param klass a class
* \return created metaclass for the class
* \pre \a klass is a Class object
* \pre \a klass has no singleton class.
* \post the class of \a klass is the returned class.
* \post the returned class is meta^(n+1)-class when \a klass is a meta^(n)-klass for n >= 0
*/
static inline VALUE
make_metaclass(VALUE klass)
{
VALUE super;
VALUE metaclass = rb_class_boot(Qundef);
FL_SET(metaclass, FL_SINGLETON);
rb_singleton_class_attached(metaclass, klass);
if (META_CLASS_OF_CLASS_CLASS_P(klass)) {
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
SET_METACLASS_OF(klass, metaclass);
SET_METACLASS_OF(metaclass, metaclass);
}
else {
VALUE tmp = METACLASS_OF(klass); /* for a meta^(n)-class klass, tmp is meta^(n)-class of Class class */
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
SET_METACLASS_OF(klass, metaclass);
SET_METACLASS_OF(metaclass, ENSURE_EIGENCLASS(tmp));
}
super = RCLASS_SUPER(klass);
while (RB_TYPE_P(super, T_ICLASS)) super = RCLASS_SUPER(super);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RCLASS_SET_SUPER(metaclass, super ? ENSURE_EIGENCLASS(super) : rb_cClass);
return metaclass;
}
/*!
* Creates a singleton class for \a obj.
* \pre \a obj must not a immediate nor a special const.
* \pre \a obj must not a Class object.
* \pre \a obj has no singleton class.
*/
static inline VALUE
make_singleton_class(VALUE obj)
{
VALUE orig_class = RBASIC(obj)->klass;
VALUE klass = rb_class_boot(orig_class);
FL_SET(klass, FL_SINGLETON);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RBASIC_SET_CLASS(obj, klass);
rb_singleton_class_attached(klass, obj);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
SET_METACLASS_OF(klass, METACLASS_OF(rb_class_real(orig_class)));
return klass;
}
static VALUE
boot_defclass(const char *name, VALUE super)
{
VALUE obj = rb_class_boot(super);
ID id = rb_intern(name);
rb_const_set((rb_cObject ? rb_cObject : obj), id, obj);
rb_vm_add_root_module(obj);
return obj;
}
void
Init_class_hierarchy(void)
{
rb_cBasicObject = boot_defclass("BasicObject", 0);
rb_cObject = boot_defclass("Object", rb_cBasicObject);
rb_gc_register_mark_object(rb_cObject);
/* resolve class name ASAP for order-independence */
rb_set_class_path_string(rb_cObject, rb_cObject, rb_fstring_lit("Object"));
rb_cModule = boot_defclass("Module", rb_cObject);
rb_cClass = boot_defclass("Class", rb_cModule);
rb_const_set(rb_cObject, rb_intern_const("BasicObject"), rb_cBasicObject);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RBASIC_SET_CLASS(rb_cClass, rb_cClass);
RBASIC_SET_CLASS(rb_cModule, rb_cClass);
RBASIC_SET_CLASS(rb_cObject, rb_cClass);
RBASIC_SET_CLASS(rb_cBasicObject, rb_cClass);
}
/*!
* \internal
* Creates a new *singleton class* for an object.
*
* \pre \a obj has no singleton class.
* \note DO NOT USE the function in an extension libraries. Use \ref rb_singleton_class.
* \param obj An object.
* \param unused ignored.
* \return The singleton class of the object.
*/
VALUE
rb_make_metaclass(VALUE obj, VALUE unused)
{
if (BUILTIN_TYPE(obj) == T_CLASS) {
return make_metaclass(obj);
}
else {
return make_singleton_class(obj);
}
}
/*!
* Defines a new class.
* \param id ignored
* \param super A class from which the new class will derive. NULL means \c Object class.
* \return the created class
* \throw TypeError if super is not a \c Class object.
*
* \note the returned class will not be associated with \a id.
* You must explicitly set a class name if necessary.
*/
VALUE
rb_define_class_id(ID id, VALUE super)
{
VALUE klass;
if (!super) super = rb_cObject;
klass = rb_class_new(super);
rb_make_metaclass(klass, RBASIC(super)->klass);
return klass;
}
/*!
* Calls Class#inherited.
* \param super A class which will be called #inherited.
* NULL means Object class.
* \param klass A Class object which derived from \a super
* \return the value \c Class#inherited's returns
* \pre Each of \a super and \a klass must be a \c Class object.
*/
mjit_compile.c: merge initial JIT compiler which has been developed by Takashi Kokubun <takashikkbn@gmail> as YARV-MJIT. Many of its bugs are fixed by wanabe <s.wanabe@gmail.com>. This JIT compiler is designed to be a safe migration path to introduce JIT compiler to MRI. So this commit does not include any bytecode changes or dynamic instruction modifications, which are done in original MJIT. This commit even strips off some aggressive optimizations from YARV-MJIT, and thus it's slower than YARV-MJIT too. But it's still fairly faster than Ruby 2.5 in some benchmarks (attached below). Note that this JIT compiler passes `make test`, `make test-all`, `make test-spec` without JIT, and even with JIT. Not only it's perfectly safe with JIT disabled because it does not replace VM instructions unlike MJIT, but also with JIT enabled it stably runs Ruby applications including Rails applications. I'm expecting this version as just "initial" JIT compiler. I have many optimization ideas which are skipped for initial merging, and you may easily replace this JIT compiler with a faster one by just replacing mjit_compile.c. `mjit_compile` interface is designed for the purpose. common.mk: update dependencies for mjit_compile.c. internal.h: declare `rb_vm_insn_addr2insn` for MJIT. vm.c: exclude some definitions if `-DMJIT_HEADER` is provided to compiler. This avoids to include some functions which take a long time to compile, e.g. vm_exec_core. Some of the purpose is achieved in transform_mjit_header.rb (see `IGNORED_FUNCTIONS`) but others are manually resolved for now. Load mjit_helper.h for MJIT header. mjit_helper.h: New. This is a file used only by JIT-ed code. I'll refactor `mjit_call_cfunc` later. vm_eval.c: add some #ifdef switches to skip compiling some functions like Init_vm_eval. win32/mkexports.rb: export thread/ec functions, which are used by MJIT. include/ruby/defines.h: add MJIT_FUNC_EXPORTED macro alis to clarify that a function is exported only for MJIT. array.c: export a function used by MJIT. bignum.c: ditto. class.c: ditto. compile.c: ditto. error.c: ditto. gc.c: ditto. hash.c: ditto. iseq.c: ditto. numeric.c: ditto. object.c: ditto. proc.c: ditto. re.c: ditto. st.c: ditto. string.c: ditto. thread.c: ditto. variable.c: ditto. vm_backtrace.c: ditto. vm_insnhelper.c: ditto. vm_method.c: ditto. I would like to improve maintainability of function exports, but I believe this way is acceptable as initial merging if we clarify the new exports are for MJIT (so that we can use them as TODO list to fix) and add unit tests to detect unresolved symbols. I'll add unit tests of JIT compilations in succeeding commits. Author: Takashi Kokubun <takashikkbn@gmail.com> Contributor: wanabe <s.wanabe@gmail.com> Part of [Feature #14235] --- * Known issues * Code generated by gcc is faster than clang. The benchmark may be worse in macOS. Following benchmark result is provided by gcc w/ Linux. * Performance is decreased when Google Chrome is running * JIT can work on MinGW, but it doesn't improve performance at least in short running benchmark. * Currently it doesn't perform well with Rails. We'll try to fix this before release. --- * Benchmark reslts Benchmarked with: Intel 4.0GHz i7-4790K with 16GB memory under x86-64 Ubuntu 8 Cores - 2.0.0-p0: Ruby 2.0.0-p0 - r62186: Ruby trunk (early 2.6.0), before MJIT changes - JIT off: On this commit, but without `--jit` option - JIT on: On this commit, and with `--jit` option ** Optcarrot fps Benchmark: https://github.com/mame/optcarrot | |2.0.0-p0 |r62186 |JIT off |JIT on | |:--------|:--------|:--------|:--------|:--------| |fps |37.32 |51.46 |51.31 |58.88 | |vs 2.0.0 |1.00x |1.38x |1.37x |1.58x | ** MJIT benchmarks Benchmark: https://github.com/benchmark-driver/mjit-benchmarks (Original: https://github.com/vnmakarov/ruby/tree/rtl_mjit_branch/MJIT-benchmarks) | |2.0.0-p0 |r62186 |JIT off |JIT on | |:----------|:--------|:--------|:--------|:--------| |aread |1.00 |1.09 |1.07 |2.19 | |aref |1.00 |1.13 |1.11 |2.22 | |aset |1.00 |1.50 |1.45 |2.64 | |awrite |1.00 |1.17 |1.13 |2.20 | |call |1.00 |1.29 |1.26 |2.02 | |const2 |1.00 |1.10 |1.10 |2.19 | |const |1.00 |1.11 |1.10 |2.19 | |fannk |1.00 |1.04 |1.02 |1.00 | |fib |1.00 |1.32 |1.31 |1.84 | |ivread |1.00 |1.13 |1.12 |2.43 | |ivwrite |1.00 |1.23 |1.21 |2.40 | |mandelbrot |1.00 |1.13 |1.16 |1.28 | |meteor |1.00 |2.97 |2.92 |3.17 | |nbody |1.00 |1.17 |1.15 |1.49 | |nest-ntimes|1.00 |1.22 |1.20 |1.39 | |nest-while |1.00 |1.10 |1.10 |1.37 | |norm |1.00 |1.18 |1.16 |1.24 | |nsvb |1.00 |1.16 |1.16 |1.17 | |red-black |1.00 |1.02 |0.99 |1.12 | |sieve |1.00 |1.30 |1.28 |1.62 | |trees |1.00 |1.14 |1.13 |1.19 | |while |1.00 |1.12 |1.11 |2.41 | ** Discourse's script/bench.rb Benchmark: https://github.com/discourse/discourse/blob/v1.8.7/script/bench.rb NOTE: Rails performance was somehow a little degraded with JIT for now. We should fix this. (At least I know opt_aref is performing badly in JIT and I have an idea to fix it. Please wait for the fix.) *** JIT off Your Results: (note for timings- percentile is first, duration is second in millisecs) categories_admin: 50: 17 75: 18 90: 22 99: 29 home_admin: 50: 21 75: 21 90: 27 99: 40 topic_admin: 50: 17 75: 18 90: 22 99: 32 categories: 50: 35 75: 41 90: 43 99: 77 home: 50: 39 75: 46 90: 49 99: 95 topic: 50: 46 75: 52 90: 56 99: 101 *** JIT on Your Results: (note for timings- percentile is first, duration is second in millisecs) categories_admin: 50: 19 75: 21 90: 25 99: 33 home_admin: 50: 24 75: 26 90: 30 99: 35 topic_admin: 50: 19 75: 20 90: 25 99: 30 categories: 50: 40 75: 44 90: 48 99: 76 home: 50: 42 75: 48 90: 51 99: 89 topic: 50: 49 75: 55 90: 58 99: 99 git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62197 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 14:22:28 +03:00
MJIT_FUNC_EXPORTED VALUE
rb_class_inherited(VALUE super, VALUE klass)
{
ID inherited;
if (!super) super = rb_cObject;
CONST_ID(inherited, "inherited");
return rb_funcall(super, inherited, 1, klass);
}
VALUE
rb_define_class(const char *name, VALUE super)
{
VALUE klass;
ID id;
id = rb_intern(name);
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
if (rb_const_defined(rb_cObject, id)) {
klass = rb_const_get(rb_cObject, id);
if (!RB_TYPE_P(klass, T_CLASS)) {
rb_raise(rb_eTypeError, "%s is not a class (%"PRIsVALUE")",
name, rb_obj_class(klass));
}
if (rb_class_real(RCLASS_SUPER(klass)) != super) {
rb_raise(rb_eTypeError, "superclass mismatch for class %s", name);
}
/* Class may have been defined in Ruby and not pin-rooted */
rb_vm_add_root_module(klass);
return klass;
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
if (!super) {
rb_raise(rb_eArgError, "no super class for `%s'", name);
}
klass = rb_define_class_id(id, super);
rb_vm_add_root_module(klass);
rb_const_set(rb_cObject, id, klass);
rb_class_inherited(super, klass);
return klass;
}
VALUE
rb_define_class_under(VALUE outer, const char *name, VALUE super)
{
return rb_define_class_id_under(outer, rb_intern(name), super);
}
/*!
* Defines a class under the namespace of \a outer.
* \param outer a class which contains the new class.
* \param id name of the new class
* \param super a class from which the new class will derive.
* NULL means \c Object class.
* \return the created class
* \throw TypeError if the constant name \a name is already taken but
* the constant is not a \c Class.
* \throw TypeError if the class is already defined but the class can not
* be reopened because its superclass is not \a super.
* \post top-level constant named \a name refers the returned class.
*
* \note if a class named \a name is already defined and its superclass is
* \a super, the function just returns the defined class.
* \note the compaction GC does not move classes returned by this function.
*/
VALUE
rb_define_class_id_under(VALUE outer, ID id, VALUE super)
{
VALUE klass;
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
if (rb_const_defined_at(outer, id)) {
klass = rb_const_get_at(outer, id);
if (!RB_TYPE_P(klass, T_CLASS)) {
rb_raise(rb_eTypeError, "%"PRIsVALUE"::%"PRIsVALUE" is not a class"
" (%"PRIsVALUE")",
outer, rb_id2str(id), rb_obj_class(klass));
}
if (rb_class_real(RCLASS_SUPER(klass)) != super) {
rb_raise(rb_eTypeError, "superclass mismatch for class "
"%"PRIsVALUE"::%"PRIsVALUE""
" (%"PRIsVALUE" is given but was %"PRIsVALUE")",
outer, rb_id2str(id), RCLASS_SUPER(klass), super);
}
/* Class may have been defined in Ruby and not pin-rooted */
rb_vm_add_root_module(klass);
return klass;
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
if (!super) {
rb_raise(rb_eArgError, "no super class for `%"PRIsVALUE"::%"PRIsVALUE"'",
rb_class_path(outer), rb_id2str(id));
}
klass = rb_define_class_id(id, super);
rb_set_class_path_string(klass, outer, rb_id2str(id));
rb_const_set(outer, id, klass);
rb_class_inherited(super, klass);
rb_vm_add_root_module(klass);
return klass;
}
VALUE
rb_module_new(void)
{
VALUE mdl = class_alloc(T_MODULE, rb_cModule);
RCLASS_M_TBL_INIT(mdl);
return (VALUE)mdl;
}
// Kept for compatibility. Use rb_module_new() instead.
VALUE
rb_define_module_id(ID id)
{
2019-10-09 18:34:03 +03:00
return rb_module_new();
}
VALUE
rb_define_module(const char *name)
{
VALUE module;
ID id;
id = rb_intern(name);
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
if (rb_const_defined(rb_cObject, id)) {
module = rb_const_get(rb_cObject, id);
if (!RB_TYPE_P(module, T_MODULE)) {
rb_raise(rb_eTypeError, "%s is not a module (%"PRIsVALUE")",
name, rb_obj_class(module));
}
/* Module may have been defined in Ruby and not pin-rooted */
rb_vm_add_root_module(module);
return module;
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
module = rb_module_new();
rb_vm_add_root_module(module);
rb_const_set(rb_cObject, id, module);
return module;
}
VALUE
rb_define_module_under(VALUE outer, const char *name)
{
return rb_define_module_id_under(outer, rb_intern(name));
}
VALUE
rb_define_module_id_under(VALUE outer, ID id)
{
VALUE module;
if (rb_const_defined_at(outer, id)) {
module = rb_const_get_at(outer, id);
if (!RB_TYPE_P(module, T_MODULE)) {
rb_raise(rb_eTypeError, "%"PRIsVALUE"::%"PRIsVALUE" is not a module"
" (%"PRIsVALUE")",
outer, rb_id2str(id), rb_obj_class(module));
}
/* Module may have been defined in Ruby and not pin-rooted */
rb_gc_register_mark_object(module);
return module;
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
}
module = rb_module_new();
rb_const_set(outer, id, module);
rb_set_class_path_string(module, outer, rb_id2str(id));
rb_gc_register_mark_object(module);
return module;
}
VALUE
rb_include_class_new(VALUE module, VALUE super)
{
VALUE klass = class_alloc(T_ICLASS, rb_cClass);
RCLASS_M_TBL(klass) = RCLASS_M_TBL(module);
RCLASS_SET_ORIGIN(klass, klass);
if (BUILTIN_TYPE(module) == T_ICLASS) {
module = RBASIC(module)->klass;
}
RUBY_ASSERT(!RB_TYPE_P(module, T_ICLASS));
if (!RCLASS_IV_TBL(module)) {
RCLASS_IV_TBL(module) = st_init_numtable();
}
if (!RCLASS_CONST_TBL(module)) {
RCLASS_CONST_TBL(module) = rb_id_table_create(0);
}
RCLASS_IV_TBL(klass) = RCLASS_IV_TBL(module);
Add a cache for class variables Redo of 34a2acdac788602c14bf05fb616215187badd504 and 931138b00696419945dc03e10f033b1f53cd50f3 which were reverted. GitHub PR #4340. This change implements a cache for class variables. Previously there was no cache for cvars. Cvar access is slow due to needing to travel all the way up th ancestor tree before returning the cvar value. The deeper the ancestor tree the slower cvar access will be. The benefits of the cache are more visible with a higher number of included modules due to the way Ruby looks up class variables. The benchmark here includes 26 modules and shows with the cache, this branch is 6.5x faster when accessing class variables. ``` compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105c) [x86_64-darwin19] built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be009) [x86_64-darwin19] | |compare-ruby|built-ruby| |:--------|-----------:|---------:| |vm_cvar | 5.681M| 36.980M| | | -| 6.51x| ``` Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails application. ActiveRecord::Base.logger has 71 ancestors. The more ancestors a tree has, the more clear the speed increase. IE if Base had only one ancestor we'd see no improvement. This benchmark is run on a vanilla Rails application. Benchmark code: ```ruby require "benchmark/ips" require_relative "config/environment" Benchmark.ips do |x| x.report "logger" do ActiveRecord::Base.logger end end ``` Ruby 3.0 master / Rails 6.1: ``` Warming up -------------------------------------- logger 155.251k i/100ms Calculating ------------------------------------- ``` Ruby 3.0 with cvar cache / Rails 6.1: ``` Warming up -------------------------------------- logger 1.546M i/100ms Calculating ------------------------------------- logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s ``` Lastly we ran a benchmark to demonstate the difference between master and our cache when the number of modules increases. This benchmark measures 1 ancestor, 30 ancestors, and 100 ancestors. Ruby 3.0 master: ``` Warming up -------------------------------------- 1 module 1.231M i/100ms 30 modules 432.020k i/100ms 100 modules 145.399k i/100ms Calculating ------------------------------------- 1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s 30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s 100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s Comparison: 1 module: 12209958.3 i/s 30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower 100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower ``` Ruby 3.0 with cvar cache: ``` Warming up -------------------------------------- 1 module 1.641M i/100ms 30 modules 1.655M i/100ms 100 modules 1.620M i/100ms Calculating ------------------------------------- 1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s 30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s 100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s Comparison: 1 module: 16279458.0 i/s 100 modules: 16087484.6 i/s - same-ish: difference falls within error 30 modules: 15891406.2 i/s - same-ish: difference falls within error ``` Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
2021-06-01 20:34:06 +03:00
RCLASS_CVC_TBL(klass) = RCLASS_CVC_TBL(module);
RCLASS_CONST_TBL(klass) = RCLASS_CONST_TBL(module);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RCLASS_SET_SUPER(klass, super);
RBASIC_SET_CLASS(klass, module);
return (VALUE)klass;
}
static int include_modules_at(const VALUE klass, VALUE c, VALUE module, int search_super);
static void
ensure_includable(VALUE klass, VALUE module)
{
rb_class_modify_check(klass);
Check_Type(module, T_MODULE);
if (!NIL_P(rb_refinement_module_get_refined_class(module))) {
rb_raise(rb_eArgError, "refinement module is not allowed");
}
}
void
rb_include_module(VALUE klass, VALUE module)
{
int changed = 0;
ensure_includable(klass, module);
changed = include_modules_at(klass, RCLASS_ORIGIN(klass), module, TRUE);
if (changed < 0)
rb_raise(rb_eArgError, "cyclic include detected");
if (RB_TYPE_P(klass, T_MODULE)) {
2021-01-26 19:49:57 +03:00
rb_subclass_entry_t *iclass = RCLASS_SUBCLASSES(klass);
int do_include = 1;
while (iclass) {
VALUE check_class = iclass->klass;
while (check_class) {
if (RB_TYPE_P(check_class, T_ICLASS) &&
(RBASIC(check_class)->klass == module)) {
do_include = 0;
}
check_class = RCLASS_SUPER(check_class);
}
if (do_include) {
include_modules_at(iclass->klass, RCLASS_ORIGIN(iclass->klass), module, TRUE);
}
iclass = iclass->next;
}
}
}
static enum rb_id_table_iterator_result
add_refined_method_entry_i(ID key, VALUE value, void *data)
* revised r37993 to avoid SEGV/ILL in tests. In r37993, a method entry with VM_METHOD_TYPE_REFINED holds only the original method definition, so ci->me is set to a method entry allocated in the stack, and it causes SEGV/ILL. In this commit, a method entry with VM_METHOD_TYPE_REFINED holds the whole original method entry. Furthermore, rb_thread_mark() is changed to mark cfp->klass to avoid GC for iclasses created by copy_refinement_iclass(). * vm_method.c (rb_method_entry_make): add a method entry with VM_METHOD_TYPE_REFINED to the class refined by the refinement if the target module is a refinement. When a method entry with VM_METHOD_TYPE_UNDEF is invoked by vm_call_method(), a method with the same name is searched in refinements. If such a method is found, the method is invoked. Otherwise, the original method in the refined class (rb_method_definition_t::body.orig_me) is invoked. This change is made to simplify the normal method lookup and to improve the performance of normal method calls. * vm_method.c (EXPR1, search_method, rb_method_entry), vm_eval.c (rb_call0, rb_search_method_entry): do not use refinements for method lookup. * vm_insnhelper.c (vm_call_method): search methods in refinements if ci->me is VM_METHOD_TYPE_REFINED. If the method is called by super (i.e., ci->call == vm_call_super_method), skip the same method entry as the current method to avoid infinite call of the same method. * class.c (include_modules_at): add a refined method entry for each method defined in a module included in a refinement. * class.c (rb_prepend_module): set an empty table to RCLASS_M_TBL(klass) to add refined method entries, because refinements should have priority over prepended modules. * proc.c (mnew): use rb_method_entry_with_refinements() to get a refined method. * vm.c (rb_thread_mark): mark cfp->klass for iclasses created by copy_refinement_iclass(). * vm.c (Init_VM), cont.c (fiber_init): initialize th->cfp->klass. * test/ruby/test_refinement.rb (test_inline_method_cache): do not skip the test because it should pass successfully. * test/ruby/test_refinement.rb (test_redefine_refined_method): new test for the case a refined method is redefined. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38236 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2012-12-06 17:08:41 +04:00
{
rb_add_refined_method_entry((VALUE)data, key);
return ID_TABLE_CONTINUE;
* revised r37993 to avoid SEGV/ILL in tests. In r37993, a method entry with VM_METHOD_TYPE_REFINED holds only the original method definition, so ci->me is set to a method entry allocated in the stack, and it causes SEGV/ILL. In this commit, a method entry with VM_METHOD_TYPE_REFINED holds the whole original method entry. Furthermore, rb_thread_mark() is changed to mark cfp->klass to avoid GC for iclasses created by copy_refinement_iclass(). * vm_method.c (rb_method_entry_make): add a method entry with VM_METHOD_TYPE_REFINED to the class refined by the refinement if the target module is a refinement. When a method entry with VM_METHOD_TYPE_UNDEF is invoked by vm_call_method(), a method with the same name is searched in refinements. If such a method is found, the method is invoked. Otherwise, the original method in the refined class (rb_method_definition_t::body.orig_me) is invoked. This change is made to simplify the normal method lookup and to improve the performance of normal method calls. * vm_method.c (EXPR1, search_method, rb_method_entry), vm_eval.c (rb_call0, rb_search_method_entry): do not use refinements for method lookup. * vm_insnhelper.c (vm_call_method): search methods in refinements if ci->me is VM_METHOD_TYPE_REFINED. If the method is called by super (i.e., ci->call == vm_call_super_method), skip the same method entry as the current method to avoid infinite call of the same method. * class.c (include_modules_at): add a refined method entry for each method defined in a module included in a refinement. * class.c (rb_prepend_module): set an empty table to RCLASS_M_TBL(klass) to add refined method entries, because refinements should have priority over prepended modules. * proc.c (mnew): use rb_method_entry_with_refinements() to get a refined method. * vm.c (rb_thread_mark): mark cfp->klass for iclasses created by copy_refinement_iclass(). * vm.c (Init_VM), cont.c (fiber_init): initialize th->cfp->klass. * test/ruby/test_refinement.rb (test_inline_method_cache): do not skip the test because it should pass successfully. * test/ruby/test_refinement.rb (test_redefine_refined_method): new test for the case a refined method is redefined. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38236 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2012-12-06 17:08:41 +04:00
}
static enum rb_id_table_iterator_result
clear_module_cache_i(ID id, VALUE val, void *data)
{
VALUE klass = (VALUE)data;
rb_clear_method_cache(klass, id);
return ID_TABLE_CONTINUE;
}
static bool
module_in_super_chain(const VALUE klass, VALUE module)
{
struct rb_id_table *const klass_m_tbl = RCLASS_M_TBL(RCLASS_ORIGIN(klass));
if (klass_m_tbl) {
while (module) {
if (klass_m_tbl == RCLASS_M_TBL(module))
return true;
module = RCLASS_SUPER(module);
}
}
return false;
}
static int
do_include_modules_at(const VALUE klass, VALUE c, VALUE module, int search_super, bool check_cyclic)
{
VALUE p, iclass, origin_stack = 0;
int method_changed = 0, constant_changed = 0, add_subclass;
long origin_len;
VALUE klass_origin = RCLASS_ORIGIN(klass);
VALUE original_klass = klass;
if (check_cyclic && module_in_super_chain(klass, module))
return -1;
while (module) {
int c_seen = FALSE;
int superclass_seen = FALSE;
struct rb_id_table *tbl;
if (klass == c) {
c_seen = TRUE;
}
if (klass_origin != c || search_super) {
/* ignore if the module included already in superclasses for include,
* ignore if the module included before origin class for prepend
*/
for (p = RCLASS_SUPER(klass); p; p = RCLASS_SUPER(p)) {
int type = BUILTIN_TYPE(p);
if (klass_origin == p && !search_super)
break;
if (c == p)
c_seen = TRUE;
if (type == T_ICLASS) {
if (RCLASS_M_TBL(p) == RCLASS_M_TBL(module)) {
if (!superclass_seen && c_seen) {
c = p; /* move insertion point */
}
goto skip;
}
}
else if (type == T_CLASS) {
superclass_seen = TRUE;
}
}
}
VALUE super_class = RCLASS_SUPER(c);
// invalidate inline method cache
Add a cache for class variables Redo of 34a2acdac788602c14bf05fb616215187badd504 and 931138b00696419945dc03e10f033b1f53cd50f3 which were reverted. GitHub PR #4340. This change implements a cache for class variables. Previously there was no cache for cvars. Cvar access is slow due to needing to travel all the way up th ancestor tree before returning the cvar value. The deeper the ancestor tree the slower cvar access will be. The benefits of the cache are more visible with a higher number of included modules due to the way Ruby looks up class variables. The benchmark here includes 26 modules and shows with the cache, this branch is 6.5x faster when accessing class variables. ``` compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105c) [x86_64-darwin19] built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be009) [x86_64-darwin19] | |compare-ruby|built-ruby| |:--------|-----------:|---------:| |vm_cvar | 5.681M| 36.980M| | | -| 6.51x| ``` Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails application. ActiveRecord::Base.logger has 71 ancestors. The more ancestors a tree has, the more clear the speed increase. IE if Base had only one ancestor we'd see no improvement. This benchmark is run on a vanilla Rails application. Benchmark code: ```ruby require "benchmark/ips" require_relative "config/environment" Benchmark.ips do |x| x.report "logger" do ActiveRecord::Base.logger end end ``` Ruby 3.0 master / Rails 6.1: ``` Warming up -------------------------------------- logger 155.251k i/100ms Calculating ------------------------------------- ``` Ruby 3.0 with cvar cache / Rails 6.1: ``` Warming up -------------------------------------- logger 1.546M i/100ms Calculating ------------------------------------- logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s ``` Lastly we ran a benchmark to demonstate the difference between master and our cache when the number of modules increases. This benchmark measures 1 ancestor, 30 ancestors, and 100 ancestors. Ruby 3.0 master: ``` Warming up -------------------------------------- 1 module 1.231M i/100ms 30 modules 432.020k i/100ms 100 modules 145.399k i/100ms Calculating ------------------------------------- 1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s 30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s 100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s Comparison: 1 module: 12209958.3 i/s 30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower 100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower ``` Ruby 3.0 with cvar cache: ``` Warming up -------------------------------------- 1 module 1.641M i/100ms 30 modules 1.655M i/100ms 100 modules 1.620M i/100ms Calculating ------------------------------------- 1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s 30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s 100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s Comparison: 1 module: 16279458.0 i/s 100 modules: 16087484.6 i/s - same-ish: difference falls within error 30 modules: 15891406.2 i/s - same-ish: difference falls within error ``` Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
2021-06-01 20:34:06 +03:00
RB_DEBUG_COUNTER_INC(cvar_include_invalidate);
ruby_vm_global_cvar_state++;
tbl = RCLASS_M_TBL(module);
if (tbl && rb_id_table_size(tbl)) {
if (search_super) { // include
if (super_class && !RB_TYPE_P(super_class, T_MODULE)) {
rb_id_table_foreach(tbl, clear_module_cache_i, (void *)super_class);
}
}
else { // prepend
if (!RB_TYPE_P(original_klass, T_MODULE)) {
rb_id_table_foreach(tbl, clear_module_cache_i, (void *)original_klass);
}
}
method_changed = 1;
}
// setup T_ICLASS for the include/prepend module
iclass = rb_include_class_new(module, super_class);
c = RCLASS_SET_SUPER(c, iclass);
RCLASS_SET_INCLUDER(iclass, klass);
add_subclass = TRUE;
if (module != RCLASS_ORIGIN(module)) {
if (!origin_stack) origin_stack = rb_ary_tmp_new(2);
VALUE origin[2] = {iclass, RCLASS_ORIGIN(module)};
rb_ary_cat(origin_stack, origin, 2);
}
else if (origin_stack && (origin_len = RARRAY_LEN(origin_stack)) > 1 &&
RARRAY_AREF(origin_stack, origin_len - 1) == module) {
RCLASS_SET_ORIGIN(RARRAY_AREF(origin_stack, (origin_len -= 2)), iclass);
RICLASS_SET_ORIGIN_SHARED_MTBL(iclass);
rb_ary_resize(origin_stack, origin_len);
add_subclass = FALSE;
}
2021-08-20 08:03:08 +03:00
if (add_subclass) {
VALUE m = module;
if (BUILTIN_TYPE(m) == T_ICLASS) m = RBASIC(m)->klass;
2021-08-20 08:03:08 +03:00
rb_module_add_to_subclasses_list(m, iclass);
}
* revised r37993 to avoid SEGV/ILL in tests. In r37993, a method entry with VM_METHOD_TYPE_REFINED holds only the original method definition, so ci->me is set to a method entry allocated in the stack, and it causes SEGV/ILL. In this commit, a method entry with VM_METHOD_TYPE_REFINED holds the whole original method entry. Furthermore, rb_thread_mark() is changed to mark cfp->klass to avoid GC for iclasses created by copy_refinement_iclass(). * vm_method.c (rb_method_entry_make): add a method entry with VM_METHOD_TYPE_REFINED to the class refined by the refinement if the target module is a refinement. When a method entry with VM_METHOD_TYPE_UNDEF is invoked by vm_call_method(), a method with the same name is searched in refinements. If such a method is found, the method is invoked. Otherwise, the original method in the refined class (rb_method_definition_t::body.orig_me) is invoked. This change is made to simplify the normal method lookup and to improve the performance of normal method calls. * vm_method.c (EXPR1, search_method, rb_method_entry), vm_eval.c (rb_call0, rb_search_method_entry): do not use refinements for method lookup. * vm_insnhelper.c (vm_call_method): search methods in refinements if ci->me is VM_METHOD_TYPE_REFINED. If the method is called by super (i.e., ci->call == vm_call_super_method), skip the same method entry as the current method to avoid infinite call of the same method. * class.c (include_modules_at): add a refined method entry for each method defined in a module included in a refinement. * class.c (rb_prepend_module): set an empty table to RCLASS_M_TBL(klass) to add refined method entries, because refinements should have priority over prepended modules. * proc.c (mnew): use rb_method_entry_with_refinements() to get a refined method. * vm.c (rb_thread_mark): mark cfp->klass for iclasses created by copy_refinement_iclass(). * vm.c (Init_VM), cont.c (fiber_init): initialize th->cfp->klass. * test/ruby/test_refinement.rb (test_inline_method_cache): do not skip the test because it should pass successfully. * test/ruby/test_refinement.rb (test_redefine_refined_method): new test for the case a refined method is redefined. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38236 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2012-12-06 17:08:41 +04:00
if (FL_TEST(klass, RMODULE_IS_REFINEMENT)) {
VALUE refined_class =
rb_refinement_module_get_refined_class(klass);
rb_id_table_foreach(RCLASS_M_TBL(module), add_refined_method_entry_i, (void *)refined_class);
* fix the behavior when a module is included into a refinement. This change is a little tricky, so it might be better to prohibit module inclusion to refinements. * include/ruby/ruby.h (RMODULE_INCLUDED_INTO_REFINEMENT): new flag to represent that a module (iclass) is included into a refinement. * class.c (include_modules_at): set RMODULE_INCLUDED_INTO_REFINEMENT if klass is a refinement. * eval.c (rb_mod_refine): set the superclass of a refinement to the refined class for super. * eval.c (rb_using_refinement): skip the above superclass (the refined class) when creating iclasses for refinements. Otherwise, `using Refinement1; using Refinement2' creates iclasses: <Refinement2> -> <RefinedClass> -> <Refinement1> -> RefinedClass, where <Module> is an iclass for Module, so RefinedClass is searched before Refinement1. The correct iclasses should be <Refinement2> -> <Refinement1> -> RefinedClass. * vm_insnhelper.c (vm_search_normal_superclass): if klass is an iclass for a refinement, use the refinement's superclass instead of the iclass's superclass. Otherwise, multiple refinements are searched by super. For example, if a refinement Refinement2 includes a module M (i.e., Refinement2 -> <M> -> RefinedClass, and if refinements iclasses are <Refinement2> -> <M>' -> <Refinement1> -> RefinedClass, then super in <Refinement2> should use Refinement2's superclass <M> instead of <Refinement2>'s superclass <M>'. * vm_insnhelper.c (vm_search_super_method): do not raise a NotImplementError if current_defind_class is a module included into a refinement. Because of the change of vm_search_normal_superclass(), the receiver might not be an instance of the module('s iclass). * test/ruby/test_refinement.rb: related test. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38298 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2012-12-10 20:05:45 +04:00
FL_SET(c, RMODULE_INCLUDED_INTO_REFINEMENT);
* revised r37993 to avoid SEGV/ILL in tests. In r37993, a method entry with VM_METHOD_TYPE_REFINED holds only the original method definition, so ci->me is set to a method entry allocated in the stack, and it causes SEGV/ILL. In this commit, a method entry with VM_METHOD_TYPE_REFINED holds the whole original method entry. Furthermore, rb_thread_mark() is changed to mark cfp->klass to avoid GC for iclasses created by copy_refinement_iclass(). * vm_method.c (rb_method_entry_make): add a method entry with VM_METHOD_TYPE_REFINED to the class refined by the refinement if the target module is a refinement. When a method entry with VM_METHOD_TYPE_UNDEF is invoked by vm_call_method(), a method with the same name is searched in refinements. If such a method is found, the method is invoked. Otherwise, the original method in the refined class (rb_method_definition_t::body.orig_me) is invoked. This change is made to simplify the normal method lookup and to improve the performance of normal method calls. * vm_method.c (EXPR1, search_method, rb_method_entry), vm_eval.c (rb_call0, rb_search_method_entry): do not use refinements for method lookup. * vm_insnhelper.c (vm_call_method): search methods in refinements if ci->me is VM_METHOD_TYPE_REFINED. If the method is called by super (i.e., ci->call == vm_call_super_method), skip the same method entry as the current method to avoid infinite call of the same method. * class.c (include_modules_at): add a refined method entry for each method defined in a module included in a refinement. * class.c (rb_prepend_module): set an empty table to RCLASS_M_TBL(klass) to add refined method entries, because refinements should have priority over prepended modules. * proc.c (mnew): use rb_method_entry_with_refinements() to get a refined method. * vm.c (rb_thread_mark): mark cfp->klass for iclasses created by copy_refinement_iclass(). * vm.c (Init_VM), cont.c (fiber_init): initialize th->cfp->klass. * test/ruby/test_refinement.rb (test_inline_method_cache): do not skip the test because it should pass successfully. * test/ruby/test_refinement.rb (test_redefine_refined_method): new test for the case a refined method is redefined. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38236 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2012-12-06 17:08:41 +04:00
}
tbl = RCLASS_CONST_TBL(module);
if (tbl && rb_id_table_size(tbl)) constant_changed = 1;
skip:
module = RCLASS_SUPER(module);
}
if (constant_changed) rb_clear_constant_cache();
return method_changed;
}
static int
include_modules_at(const VALUE klass, VALUE c, VALUE module, int search_super)
{
return do_include_modules_at(klass, c, module, search_super, true);
}
static enum rb_id_table_iterator_result
move_refined_method(ID key, VALUE value, void *data)
{
rb_method_entry_t *me = (rb_method_entry_t *)value;
if (me->def->type == VM_METHOD_TYPE_REFINED) {
VALUE klass = (VALUE)data;
struct rb_id_table *tbl = RCLASS_M_TBL(klass);
if (me->def->body.refined.orig_me) {
const rb_method_entry_t *orig_me = me->def->body.refined.orig_me, *new_me;
RB_OBJ_WRITE(me, &me->def->body.refined.orig_me, NULL);
new_me = rb_method_entry_clone(me);
rb_method_table_insert(klass, tbl, key, new_me);
rb_method_entry_copy(me, orig_me);
return ID_TABLE_CONTINUE;
}
else {
rb_method_table_insert(klass, tbl, key, me);
return ID_TABLE_DELETE;
}
}
else {
return ID_TABLE_CONTINUE;
}
}
static enum rb_id_table_iterator_result
cache_clear_refined_method(ID key, VALUE value, void *data)
{
rb_method_entry_t *me = (rb_method_entry_t *) value;
if (me->def->type == VM_METHOD_TYPE_REFINED && me->def->body.refined.orig_me) {
VALUE klass = (VALUE)data;
rb_clear_method_cache(klass, me->called_id);
}
// Refined method entries without an orig_me is going to stay in the method
// table of klass, like before the move, so no need to clear the cache.
return ID_TABLE_CONTINUE;
}
static bool
ensure_origin(VALUE klass)
{
VALUE origin = RCLASS_ORIGIN(klass);
if (origin == klass) {
origin = class_alloc(T_ICLASS, klass);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RCLASS_SET_SUPER(origin, RCLASS_SUPER(klass));
RCLASS_SET_SUPER(klass, origin);
RCLASS_SET_ORIGIN(klass, origin);
* internal.h: remove struct method_table_wrapper. struct method_table_wrapper was introduced to avoid duplicate marking for method tables. For example, `module M1; def foo; end; end` make one method table (mtbl) contains a method `foo`. M1 (T_MODULE) points mtbl. Classes C1 and C2 includes M1, then two T_ICLASS objects are created and they points mtbl too. In this case, three objects (one T_MODULE and two T_ICLASS objects) points same mtbl. On marking phase, these three objects mark same mtbl. To avoid such duplication, struct method_table_wrapper was introduced. However, created two T_ICLASS objects have same or shorter lifetime than M1 (T_MODULE) object. So that we only need to mark mtbl from M1, not from T_ICLASS objects. This patch tries marking only from M1. Note that one `Module#prepend` call creates two T_ICLASS objects. One for refering to a prepending Module object, same as `Module#include`. We don't nedd to care this T_ICLASS. One for moving original mtbl from a prepending class. We need to mark such mtbl from this T_ICLASS object. To mark the mtbl, we need to use `RCLASS_ORIGIN(klass)` on marking from a prepended class `klass`. * class.c: ditto. * eval.c (rb_using_refinement): ditto. * gc.c: ditto. * include/ruby/ruby.h: define m_tbl directly. The definition of struct RClass should be moved to (srcdir)/internal.h. * method.h: remove decl of rb_free_m_tbl_wrapper(). * object.c: use RCLASS_M_TBL() directly. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49862 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-03-06 01:20:14 +03:00
RCLASS_M_TBL(origin) = RCLASS_M_TBL(klass);
RCLASS_M_TBL_INIT(klass);
rb_id_table_foreach(RCLASS_M_TBL(origin), cache_clear_refined_method, (void *)klass);
rb_id_table_foreach(RCLASS_M_TBL(origin), move_refined_method, (void *)klass);
return true;
}
return false;
}
void
rb_prepend_module(VALUE klass, VALUE module)
{
int changed;
bool klass_had_no_origin;
ensure_includable(klass, module);
if (module_in_super_chain(klass, module))
rb_raise(rb_eArgError, "cyclic prepend detected");
klass_had_no_origin = ensure_origin(klass);
changed = do_include_modules_at(klass, klass, module, FALSE, false);
RUBY_ASSERT(changed >= 0); // already checked for cyclic prepend above
if (changed) {
rb_vm_check_redefinition_by_prepend(klass);
}
if (RB_TYPE_P(klass, T_MODULE)) {
2021-01-26 19:49:57 +03:00
rb_subclass_entry_t *iclass = RCLASS_SUBCLASSES(klass);
VALUE klass_origin = RCLASS_ORIGIN(klass);
struct rb_id_table *klass_m_tbl = RCLASS_M_TBL(klass);
struct rb_id_table *klass_origin_m_tbl = RCLASS_M_TBL(klass_origin);
while (iclass) {
if (klass_had_no_origin && klass_origin_m_tbl == RCLASS_M_TBL(iclass->klass)) {
// backfill an origin iclass to handle refinements and future prepends
rb_id_table_foreach(RCLASS_M_TBL(iclass->klass), clear_module_cache_i, (void *)iclass->klass);
RCLASS_M_TBL(iclass->klass) = klass_m_tbl;
VALUE origin = rb_include_class_new(klass_origin, RCLASS_SUPER(iclass->klass));
RCLASS_SET_SUPER(iclass->klass, origin);
RCLASS_SET_INCLUDER(origin, RCLASS_INCLUDER(iclass->klass));
RCLASS_SET_ORIGIN(iclass->klass, origin);
RICLASS_SET_ORIGIN_SHARED_MTBL(origin);
}
include_modules_at(iclass->klass, iclass->klass, module, FALSE);
iclass = iclass->next;
}
}
}
/*
* call-seq:
* mod.included_modules -> array
*
* Returns the list of modules included or prepended in <i>mod</i>
* or one of <i>mod</i>'s ancestors.
*
* module Sub
* end
*
* module Mixin
* prepend Sub
* end
*
* module Outer
* include Mixin
* end
*
* Mixin.included_modules #=> [Sub]
* Outer.included_modules #=> [Sub, Mixin]
*/
VALUE
rb_mod_included_modules(VALUE mod)
{
VALUE ary = rb_ary_new();
VALUE p;
VALUE origin = RCLASS_ORIGIN(mod);
for (p = RCLASS_SUPER(mod); p; p = RCLASS_SUPER(p)) {
if (p != origin && RCLASS_ORIGIN(p) == p && BUILTIN_TYPE(p) == T_ICLASS) {
VALUE m = RBASIC(p)->klass;
if (RB_TYPE_P(m, T_MODULE))
rb_ary_push(ary, m);
}
}
return ary;
}
/*
* call-seq:
* mod.include?(module) -> true or false
*
* Returns <code>true</code> if <i>module</i> is included
* or prepended in <i>mod</i> or one of <i>mod</i>'s ancestors.
*
* module A
* end
* class B
* include A
* end
* class C < B
* end
* B.include?(A) #=> true
* C.include?(A) #=> true
* A.include?(A) #=> false
*/
VALUE
rb_mod_include_p(VALUE mod, VALUE mod2)
{
VALUE p;
Check_Type(mod2, T_MODULE);
for (p = RCLASS_SUPER(mod); p; p = RCLASS_SUPER(p)) {
Ensure origins for all included, prepended, and refined modules This fixes various issues when a module is included in or prepended to a module or class, and then refined, or refined and then included or prepended to a module or class. Implement by renaming ensure_origin to rb_ensure_origin, making it non-static, and calling it when refining a module. Fix Module#initialize_copy to handle origins correctly. Previously, Module#initialize_copy did not handle origins correctly. For example, this code: ```ruby module B; end class A def b; 2 end prepend B end a = A.dup.new class A def b; 1 end end p a.b ``` Printed 1 instead of 2. This is because the super chain for a.singleton_class was: ``` a.singleton_class A.dup B(iclass) B(iclass origin) A(origin) # not A.dup(origin) ``` The B iclasses would not be modified, so the includer entry would be still be set to A and not A.dup. This modifies things so that if the class/module has an origin, all iclasses between the class/module and the origin are duplicated and have the correct includer entry set, and the correct origin is created. This requires other changes to make sure all tests still pass: * rb_undef_methods_from doesn't automatically handle classes with origins, so pass it the origin for Comparable when undefing methods in Complex. This fixed a failure in the Complex tests. * When adding a method, the method cache was not cleared correctly if klass has an origin. Clear the method cache for the klass before switching to the origin of klass. This fixed failures in the autoload tests related to overridding require, without breaking the optimization tests. Also clear the method cache for both the module and origin when removing a method. * Module#include? is fixed to skip origin iclasses. * Refinements are fixed to use the origin class of the module that has an origin. * RCLASS_REFINED_BY_ANY is removed as it was only used in a single place and is no longer needed. * Marshal#dump is fixed to skip iclass origins. * rb_method_entry_make is fixed to handled overridden optimized methods for modules that have origins. Fixes [Bug #16852]
2020-05-24 06:16:27 +03:00
if (BUILTIN_TYPE(p) == T_ICLASS && !FL_TEST(p, RICLASS_IS_ORIGIN)) {
if (RBASIC(p)->klass == mod2) return Qtrue;
}
}
return Qfalse;
}
/*
* call-seq:
* mod.ancestors -> array
*
* Returns a list of modules included/prepended in <i>mod</i>
* (including <i>mod</i> itself).
*
* module Mod
* include Math
* include Comparable
* prepend Enumerable
* end
*
* Mod.ancestors #=> [Enumerable, Mod, Comparable, Math]
* Math.ancestors #=> [Math]
* Enumerable.ancestors #=> [Enumerable]
*/
VALUE
rb_mod_ancestors(VALUE mod)
{
VALUE p, ary = rb_ary_new();
VALUE refined_class = Qnil;
if (FL_TEST(mod, RMODULE_IS_REFINEMENT)) {
refined_class = rb_refinement_module_get_refined_class(mod);
}
for (p = mod; p; p = RCLASS_SUPER(p)) {
if (p == refined_class) break;
if (p != RCLASS_ORIGIN(p)) continue;
if (BUILTIN_TYPE(p) == T_ICLASS) {
rb_ary_push(ary, RBASIC(p)->klass);
}
else {
rb_ary_push(ary, p);
}
}
return ary;
}
static void
ins_methods_push(st_data_t name, st_data_t ary)
{
rb_ary_push((VALUE)ary, ID2SYM((ID)name));
}
static int
ins_methods_i(st_data_t name, st_data_t type, st_data_t ary)
{
switch ((rb_method_visibility_t)type) {
case METHOD_VISI_UNDEF:
case METHOD_VISI_PRIVATE:
break;
default: /* everything but private */
ins_methods_push(name, ary);
break;
}
return ST_CONTINUE;
}
static int
2021-03-21 00:57:30 +03:00
ins_methods_type_i(st_data_t name, st_data_t type, st_data_t ary, rb_method_visibility_t visi)
{
2021-03-21 00:57:30 +03:00
if ((rb_method_visibility_t)type == visi) {
ins_methods_push(name, ary);
}
return ST_CONTINUE;
}
2021-03-21 00:57:30 +03:00
static int
ins_methods_prot_i(st_data_t name, st_data_t type, st_data_t ary)
{
return ins_methods_type_i(name, type, ary, METHOD_VISI_PROTECTED);
}
static int
ins_methods_priv_i(st_data_t name, st_data_t type, st_data_t ary)
{
2021-03-21 00:57:30 +03:00
return ins_methods_type_i(name, type, ary, METHOD_VISI_PRIVATE);
}
static int
ins_methods_pub_i(st_data_t name, st_data_t type, st_data_t ary)
{
2021-03-21 00:57:30 +03:00
return ins_methods_type_i(name, type, ary, METHOD_VISI_PUBLIC);
}
struct method_entry_arg {
st_table *list;
int recur;
};
static enum rb_id_table_iterator_result
method_entry_i(ID key, VALUE value, void *data)
{
const rb_method_entry_t *me = (const rb_method_entry_t *)value;
struct method_entry_arg *arg = (struct method_entry_arg *)data;
rb_method_visibility_t type;
if (me->def->type == VM_METHOD_TYPE_REFINED) {
* method.h: introduce rb_callable_method_entry_t to remove rb_control_frame_t::klass. [Bug #11278], [Bug #11279] rb_method_entry_t data belong to modules/classes. rb_method_entry_t::owner points defined module or class. module M def foo; end end In this case, owner is M. rb_callable_method_entry_t data belong to only classes. For modules, MRI creates corresponding T_ICLASS internally. rb_callable_method_entry_t can also belong to T_ICLASS. rb_callable_method_entry_t::defined_class points T_CLASS or T_ICLASS. rb_method_entry_t data for classes (not for modules) are also rb_callable_method_entry_t data because it is completely same data. In this case, rb_method_entry_t::owner == rb_method_entry_t::defined_class. For example, there are classes C and D, and incldues M, class C; include M; end class D; include M; end then, two T_ICLASS objects for C's super class and D's super class will be created. When C.new.foo is called, then M#foo is searcheed and rb_callable_method_t data is used by VM to invoke M#foo. rb_method_entry_t data is only one for M#foo. However, rb_callable_method_entry_t data are two (and can be more). It is proportional to the number of including (and prepending) classes (the number of T_ICLASS which point to the module). Now, created rb_callable_method_entry_t are collected when the original module M was modified. We can think it is a cache. We need to select what kind of method entry data is needed. To operate definition, then you need to use rb_method_entry_t. You can access them by the following functions. * rb_method_entry(VALUE klass, ID id); * rb_method_entry_with_refinements(VALUE klass, ID id); * rb_method_entry_without_refinements(VALUE klass, ID id); * rb_resolve_refined_method(VALUE refinements, const rb_method_entry_t *me); To invoke methods, then you need to use rb_callable_method_entry_t which you can get by the following APIs corresponding to the above listed functions. * rb_callable_method_entry(VALUE klass, ID id); * rb_callable_method_entry_with_refinements(VALUE klass, ID id); * rb_callable_method_entry_without_refinements(VALUE klass, ID id); * rb_resolve_refined_method_callable(VALUE refinements, const rb_callable_method_entry_t *me); VM pushes rb_callable_method_entry_t, so that rb_vm_frame_method_entry() returns rb_callable_method_entry_t. You can check a super class of current method by rb_callable_method_entry_t::defined_class. * method.h: renamed from rb_method_entry_t::klass to rb_method_entry_t::owner. * internal.h: add rb_classext_struct::callable_m_tbl to cache rb_callable_method_entry_t data. We need to consider abotu this field again because it is only active for T_ICLASS. * class.c (method_entry_i): ditto. * class.c (rb_define_attr): rb_method_entry() does not takes defiend_class_ptr. * gc.c (mark_method_entry): mark RCLASS_CALLABLE_M_TBL() for T_ICLASS. * cont.c (fiber_init): rb_control_frame_t::klass is removed. * proc.c: fix `struct METHOD' data structure because rb_callable_method_t has all information. * vm_core.h: remove several fields. * rb_control_frame_t::klass. * rb_block_t::klass. And catch up changes. * eval.c: catch up changes. * gc.c: ditto. * insns.def: ditto. * vm.c: ditto. * vm_args.c: ditto. * vm_backtrace.c: ditto. * vm_dump.c: ditto. * vm_eval.c: ditto. * vm_insnhelper.c: ditto. * vm_method.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51126 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-03 14:24:50 +03:00
VALUE owner = me->owner;
me = rb_resolve_refined_method(Qnil, me);
if (!me) return ID_TABLE_CONTINUE;
if (!arg->recur && me->owner != owner) return ID_TABLE_CONTINUE;
}
if (!st_is_member(arg->list, key)) {
if (UNDEFINED_METHOD_ENTRY_P(me)) {
type = METHOD_VISI_UNDEF; /* none */
}
else {
type = METHOD_ENTRY_VISI(me);
RUBY_ASSERT(type != METHOD_VISI_UNDEF);
}
st_add_direct(arg->list, key, (st_data_t)type);
}
return ID_TABLE_CONTINUE;
}
static void
add_instance_method_list(VALUE mod, struct method_entry_arg *me_arg)
{
struct rb_id_table *m_tbl = RCLASS_M_TBL(mod);
if (!m_tbl) return;
rb_id_table_foreach(m_tbl, method_entry_i, me_arg);
}
static bool
particular_class_p(VALUE mod)
{
if (!mod) return false;
if (FL_TEST(mod, FL_SINGLETON)) return true;
if (BUILTIN_TYPE(mod) == T_ICLASS) return true;
return false;
}
static VALUE
class_instance_method_list(int argc, const VALUE *argv, VALUE mod, int obj, int (*func) (st_data_t, st_data_t, st_data_t))
{
VALUE ary;
int recur = TRUE, prepended = 0;
struct method_entry_arg me_arg;
if (rb_check_arity(argc, 0, 1)) recur = RTEST(argv[0]);
me_arg.list = st_init_numtable();
me_arg.recur = recur;
if (obj) {
for (; particular_class_p(mod); mod = RCLASS_SUPER(mod)) {
add_instance_method_list(mod, &me_arg);
}
}
if (!recur && RCLASS_ORIGIN(mod) != mod) {
mod = RCLASS_ORIGIN(mod);
prepended = 1;
}
for (; mod; mod = RCLASS_SUPER(mod)) {
add_instance_method_list(mod, &me_arg);
if (BUILTIN_TYPE(mod) == T_ICLASS && !prepended) continue;
if (!recur) break;
}
ary = rb_ary_new2(me_arg.list->num_entries);
st_foreach(me_arg.list, func, ary);
st_free_table(me_arg.list);
return ary;
}
/*
* call-seq:
* mod.instance_methods(include_super=true) -> array
*
* Returns an array containing the names of the public and protected instance
* methods in the receiver. For a module, these are the public and protected methods;
* for a class, they are the instance (not singleton) methods. If the optional
* parameter is <code>false</code>, the methods of any ancestors are not included.
*
* module A
* def method1() end
* end
* class B
* include A
* def method2() end
* end
* class C < B
* def method3() end
* end
*
* A.instance_methods(false) #=> [:method1]
* B.instance_methods(false) #=> [:method2]
* B.instance_methods(true).include?(:method1) #=> true
* C.instance_methods(false) #=> [:method3]
* C.instance_methods.include?(:method2) #=> true
*/
VALUE
rb_class_instance_methods(int argc, const VALUE *argv, VALUE mod)
{
return class_instance_method_list(argc, argv, mod, 0, ins_methods_i);
}
/*
* call-seq:
* mod.protected_instance_methods(include_super=true) -> array
*
* Returns a list of the protected instance methods defined in
* <i>mod</i>. If the optional parameter is <code>false</code>, the
* methods of any ancestors are not included.
*/
VALUE
rb_class_protected_instance_methods(int argc, const VALUE *argv, VALUE mod)
{
return class_instance_method_list(argc, argv, mod, 0, ins_methods_prot_i);
}
/*
* call-seq:
* mod.private_instance_methods(include_super=true) -> array
*
* Returns a list of the private instance methods defined in
* <i>mod</i>. If the optional parameter is <code>false</code>, the
* methods of any ancestors are not included.
*
* module Mod
* def method1() end
* private :method1
* def method2() end
* end
* Mod.instance_methods #=> [:method2]
* Mod.private_instance_methods #=> [:method1]
*/
VALUE
rb_class_private_instance_methods(int argc, const VALUE *argv, VALUE mod)
{
return class_instance_method_list(argc, argv, mod, 0, ins_methods_priv_i);
}
/*
* call-seq:
* mod.public_instance_methods(include_super=true) -> array
*
* Returns a list of the public instance methods defined in <i>mod</i>.
* If the optional parameter is <code>false</code>, the methods of
* any ancestors are not included.
*/
VALUE
rb_class_public_instance_methods(int argc, const VALUE *argv, VALUE mod)
{
return class_instance_method_list(argc, argv, mod, 0, ins_methods_pub_i);
}
/*
* call-seq:
* obj.methods(regular=true) -> array
*
* Returns a list of the names of public and protected methods of
* <i>obj</i>. This will include all the methods accessible in
* <i>obj</i>'s ancestors.
* If the optional parameter is <code>false</code>, it
* returns an array of <i>obj</i>'s public and protected singleton methods,
* the array will not include methods in modules included in <i>obj</i>.
*
* class Klass
* def klass_method()
* end
* end
* k = Klass.new
* k.methods[0..9] #=> [:klass_method, :nil?, :===,
* # :==~, :!, :eql?
* # :hash, :<=>, :class, :singleton_class]
* k.methods.length #=> 56
*
* k.methods(false) #=> []
* def k.singleton_method; end
* k.methods(false) #=> [:singleton_method]
*
* module M123; def m123; end end
* k.extend M123
* k.methods(false) #=> [:singleton_method]
*/
VALUE
rb_obj_methods(int argc, const VALUE *argv, VALUE obj)
{
rb_check_arity(argc, 0, 1);
if (argc > 0 && !RTEST(argv[0])) {
return rb_obj_singleton_methods(argc, argv, obj);
}
return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_i);
}
/*
* call-seq:
* obj.protected_methods(all=true) -> array
*
* Returns the list of protected methods accessible to <i>obj</i>. If
* the <i>all</i> parameter is set to <code>false</code>, only those methods
* in the receiver will be listed.
*/
VALUE
rb_obj_protected_methods(int argc, const VALUE *argv, VALUE obj)
{
return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_prot_i);
}
/*
* call-seq:
* obj.private_methods(all=true) -> array
*
* Returns the list of private methods accessible to <i>obj</i>. If
* the <i>all</i> parameter is set to <code>false</code>, only those methods
* in the receiver will be listed.
*/
VALUE
rb_obj_private_methods(int argc, const VALUE *argv, VALUE obj)
{
return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_priv_i);
}
/*
* call-seq:
* obj.public_methods(all=true) -> array
*
* Returns the list of public methods accessible to <i>obj</i>. If
* the <i>all</i> parameter is set to <code>false</code>, only those methods
* in the receiver will be listed.
*/
VALUE
rb_obj_public_methods(int argc, const VALUE *argv, VALUE obj)
{
return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_pub_i);
}
/*
* call-seq:
* obj.singleton_methods(all=true) -> array
*
* Returns an array of the names of singleton methods for <i>obj</i>.
* If the optional <i>all</i> parameter is true, the list will include
* methods in modules included in <i>obj</i>.
* Only public and protected singleton methods are returned.
*
* module Other
* def three() end
* end
*
* class Single
* def Single.four() end
* end
*
* a = Single.new
*
* def a.one()
* end
*
* class << a
* include Other
* def two()
* end
* end
*
* Single.singleton_methods #=> [:four]
* a.singleton_methods(false) #=> [:two, :one]
* a.singleton_methods #=> [:two, :one, :three]
*/
VALUE
rb_obj_singleton_methods(int argc, const VALUE *argv, VALUE obj)
{
VALUE ary, klass, origin;
struct method_entry_arg me_arg;
struct rb_id_table *mtbl;
int recur = TRUE;
if (rb_check_arity(argc, 0, 1)) recur = RTEST(argv[0]);
if (RB_TYPE_P(obj, T_CLASS) && FL_TEST(obj, FL_SINGLETON)) {
rb_singleton_class(obj);
}
klass = CLASS_OF(obj);
origin = RCLASS_ORIGIN(klass);
me_arg.list = st_init_numtable();
me_arg.recur = recur;
if (klass && FL_TEST(klass, FL_SINGLETON)) {
if ((mtbl = RCLASS_M_TBL(origin)) != 0) rb_id_table_foreach(mtbl, method_entry_i, &me_arg);
klass = RCLASS_SUPER(klass);
}
if (recur) {
while (klass && (FL_TEST(klass, FL_SINGLETON) || RB_TYPE_P(klass, T_ICLASS))) {
if (klass != origin && (mtbl = RCLASS_M_TBL(klass)) != 0) rb_id_table_foreach(mtbl, method_entry_i, &me_arg);
klass = RCLASS_SUPER(klass);
}
}
ary = rb_ary_new2(me_arg.list->num_entries);
st_foreach(me_arg.list, ins_methods_i, ary);
st_free_table(me_arg.list);
return ary;
}
/*!
* \}
*/
/*!
* \addtogroup defmethod
* \{
*/
#ifdef rb_define_method_id
#undef rb_define_method_id
#endif
void
rb_define_method_id(VALUE klass, ID mid, VALUE (*func)(ANYARGS), int argc)
{
rb_add_method_cfunc(klass, mid, func, argc, METHOD_VISI_PUBLIC);
}
drop-in type check for rb_define_method The rb_define_method function takes a pointer to ANYARGS-ed functions, which in fact varies 18 different prototypes. We still need to preserve ANYARGS for storages but why not check the consistencies if possible. Q&As: Q: Where did the magic number "18" came from in the description above? A: Count the case branch of vm_method.c:call_cfunc_invoker_func(). Note also that the 18 branches has lasted for at least 25 years. See also 200e0ee2fd3c1c006c528874a88f684447215524. Q: What is this __weakref__ thing? A: That is a kind of function overloading mechanism that GCC provides. In this case for instance rb_define_method0 is an alias of rb_define_method, with a strong type. Q: What is this __transparent_union__ thing? A: That is another kind of function overloading mechanism that GCC provides. In this case the attributed function pointer is either VALUE(*)(int,VALUE*,VALUE) or VALUE(*)(int,const VALUE*,VALUE). This is better than void* or ANYARGS because we can reject all other possibilities than the two. Q: What does this rb_define_method macro mean? A: It selects appropriate alias of the rb_define_method function, depending on the arity. Q: Why the prototype change of rb_f_notimplement? A: Function pointer to rb_f_notimplement is special cased in vm_method.c:rb_add_method_cfunc(). That should be handled by the __builtin_choose_expr chain inside of rb_define_method macro expansion. In order to do so, comparison like (func == rb_f_notimplement) is inappropriate for __builtin_choose_expr's expression (which must be a compile-time integer constant but the address of rb_f_notimplement is not fixed until the linker). So instead we are using __builtin_types_compatible_p, and in doing so we need to distinguish rb_f_notimplement from others, by type.
2019-08-28 11:11:23 +03:00
#ifdef rb_define_method
#undef rb_define_method
#endif
void
rb_define_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc)
{
rb_add_method_cfunc(klass, rb_intern(name), func, argc, METHOD_VISI_PUBLIC);
}
#ifdef rb_define_protected_method
#undef rb_define_protected_method
#endif
void
rb_define_protected_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc)
{
rb_add_method_cfunc(klass, rb_intern(name), func, argc, METHOD_VISI_PROTECTED);
}
#ifdef rb_define_private_method
#undef rb_define_private_method
#endif
void
rb_define_private_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc)
{
rb_add_method_cfunc(klass, rb_intern(name), func, argc, METHOD_VISI_PRIVATE);
}
void
rb_undef_method(VALUE klass, const char *name)
{
rb_add_method(klass, rb_intern(name), VM_METHOD_TYPE_UNDEF, 0, METHOD_VISI_UNDEF);
}
static enum rb_id_table_iterator_result
undef_method_i(ID name, VALUE value, void *data)
{
VALUE klass = (VALUE)data;
rb_add_method(klass, name, VM_METHOD_TYPE_UNDEF, 0, METHOD_VISI_UNDEF);
return ID_TABLE_CONTINUE;
}
void
rb_undef_methods_from(VALUE klass, VALUE super)
{
struct rb_id_table *mtbl = RCLASS_M_TBL(super);
if (mtbl) {
rb_id_table_foreach(mtbl, undef_method_i, (void *)klass);
}
}
/*!
* \}
*/
/*!
* \addtogroup class
* \{
*/
static inline VALUE
special_singleton_class_of(VALUE obj)
{
switch (obj) {
case Qnil: return rb_cNilClass;
case Qfalse: return rb_cFalseClass;
case Qtrue: return rb_cTrueClass;
default: return Qnil;
}
}
VALUE
rb_special_singleton_class(VALUE obj)
{
return special_singleton_class_of(obj);
}
/*!
* \internal
* Returns the singleton class of \a obj. Creates it if necessary.
*
* \note DO NOT expose the returned singleton class to
* outside of class.c.
* Use \ref rb_singleton_class instead for
* consistency of the metaclass hierarchy.
*/
static VALUE
singleton_class_of(VALUE obj)
{
VALUE klass;
switch (TYPE(obj)) {
case T_FIXNUM:
case T_BIGNUM:
case T_FLOAT:
case T_SYMBOL:
rb_raise(rb_eTypeError, "can't define singleton");
case T_FALSE:
case T_TRUE:
case T_NIL:
klass = special_singleton_class_of(obj);
if (NIL_P(klass))
rb_bug("unknown immediate %p", (void *)obj);
return klass;
case T_STRING:
if (FL_TEST_RAW(obj, RSTRING_FSTR)) {
rb_raise(rb_eTypeError, "can't define singleton");
}
}
klass = RBASIC(obj)->klass;
if (!(FL_TEST(klass, FL_SINGLETON) &&
rb_attr_get(klass, id_attached) == obj)) {
rb_serial_t serial = RCLASS_SERIAL(klass);
klass = rb_make_metaclass(obj, klass);
RCLASS_SERIAL(klass) = serial;
}
RB_FL_SET_RAW(klass, RB_OBJ_FROZEN_RAW(obj));
return klass;
}
void
rb_freeze_singleton_class(VALUE x)
{
/* should not propagate to meta-meta-class, and so on */
if (!(RBASIC(x)->flags & FL_SINGLETON)) {
VALUE klass = RBASIC_CLASS(x);
if (klass && (klass = RCLASS_ORIGIN(klass)) != 0 &&
FL_TEST(klass, (FL_SINGLETON|FL_FREEZE)) == FL_SINGLETON) {
OBJ_FREEZE_RAW(klass);
}
}
}
/*!
* Returns the singleton class of \a obj, or nil if obj is not a
* singleton object.
*
* \param obj an arbitrary object.
* \return the singleton class or nil.
*/
VALUE
rb_singleton_class_get(VALUE obj)
{
VALUE klass;
if (SPECIAL_CONST_P(obj)) {
return rb_special_singleton_class(obj);
}
klass = RBASIC(obj)->klass;
if (!FL_TEST(klass, FL_SINGLETON)) return Qnil;
if (rb_attr_get(klass, id_attached) != obj) return Qnil;
return klass;
}
/*!
* Returns the singleton class of \a obj. Creates it if necessary.
*
* \param obj an arbitrary object.
* \throw TypeError if \a obj is an Integer or a Symbol.
* \return the singleton class.
*
* \post \a obj has its own singleton class.
* \post if \a obj is a class,
* the returned singleton class also has its own
* singleton class in order to keep consistency of the
* inheritance structure of metaclasses.
* \note a new singleton class will be created
* if \a obj does not have it.
* \note the singleton classes for nil, true and false are:
* NilClass, TrueClass and FalseClass.
*/
VALUE
rb_singleton_class(VALUE obj)
{
VALUE klass = singleton_class_of(obj);
/* ensures an exposed class belongs to its own eigenclass */
if (RB_TYPE_P(obj, T_CLASS)) (void)ENSURE_EIGENCLASS(klass);
return klass;
}
/*!
* \}
*/
/*!
* \addtogroup defmethod
* \{
*/
#ifdef rb_define_singleton_method
#undef rb_define_singleton_method
#endif
/*!
* Defines a singleton method for \a obj.
* \param obj an arbitrary object
* \param name name of the singleton method
* \param func the method body
* \param argc the number of parameters, or -1 or -2. see \ref defmethod.
*/
void
rb_define_singleton_method(VALUE obj, const char *name, VALUE (*func)(ANYARGS), int argc)
{
rb_define_method(singleton_class_of(obj), name, func, argc);
}
#ifdef rb_define_module_function
#undef rb_define_module_function
#endif
void
rb_define_module_function(VALUE module, const char *name, VALUE (*func)(ANYARGS), int argc)
{
rb_define_private_method(module, name, func, argc);
rb_define_singleton_method(module, name, func, argc);
}
#ifdef rb_define_global_function
#undef rb_define_global_function
#endif
void
rb_define_global_function(const char *name, VALUE (*func)(ANYARGS), int argc)
{
rb_define_module_function(rb_mKernel, name, func, argc);
}
void
rb_define_alias(VALUE klass, const char *name1, const char *name2)
{
rb_alias(klass, rb_intern(name1), rb_intern(name2));
}
void
rb_define_attr(VALUE klass, const char *name, int read, int write)
{
rb_attr(klass, rb_intern(name), read, write, FALSE);
}
mjit_compile.c: merge initial JIT compiler which has been developed by Takashi Kokubun <takashikkbn@gmail> as YARV-MJIT. Many of its bugs are fixed by wanabe <s.wanabe@gmail.com>. This JIT compiler is designed to be a safe migration path to introduce JIT compiler to MRI. So this commit does not include any bytecode changes or dynamic instruction modifications, which are done in original MJIT. This commit even strips off some aggressive optimizations from YARV-MJIT, and thus it's slower than YARV-MJIT too. But it's still fairly faster than Ruby 2.5 in some benchmarks (attached below). Note that this JIT compiler passes `make test`, `make test-all`, `make test-spec` without JIT, and even with JIT. Not only it's perfectly safe with JIT disabled because it does not replace VM instructions unlike MJIT, but also with JIT enabled it stably runs Ruby applications including Rails applications. I'm expecting this version as just "initial" JIT compiler. I have many optimization ideas which are skipped for initial merging, and you may easily replace this JIT compiler with a faster one by just replacing mjit_compile.c. `mjit_compile` interface is designed for the purpose. common.mk: update dependencies for mjit_compile.c. internal.h: declare `rb_vm_insn_addr2insn` for MJIT. vm.c: exclude some definitions if `-DMJIT_HEADER` is provided to compiler. This avoids to include some functions which take a long time to compile, e.g. vm_exec_core. Some of the purpose is achieved in transform_mjit_header.rb (see `IGNORED_FUNCTIONS`) but others are manually resolved for now. Load mjit_helper.h for MJIT header. mjit_helper.h: New. This is a file used only by JIT-ed code. I'll refactor `mjit_call_cfunc` later. vm_eval.c: add some #ifdef switches to skip compiling some functions like Init_vm_eval. win32/mkexports.rb: export thread/ec functions, which are used by MJIT. include/ruby/defines.h: add MJIT_FUNC_EXPORTED macro alis to clarify that a function is exported only for MJIT. array.c: export a function used by MJIT. bignum.c: ditto. class.c: ditto. compile.c: ditto. error.c: ditto. gc.c: ditto. hash.c: ditto. iseq.c: ditto. numeric.c: ditto. object.c: ditto. proc.c: ditto. re.c: ditto. st.c: ditto. string.c: ditto. thread.c: ditto. variable.c: ditto. vm_backtrace.c: ditto. vm_insnhelper.c: ditto. vm_method.c: ditto. I would like to improve maintainability of function exports, but I believe this way is acceptable as initial merging if we clarify the new exports are for MJIT (so that we can use them as TODO list to fix) and add unit tests to detect unresolved symbols. I'll add unit tests of JIT compilations in succeeding commits. Author: Takashi Kokubun <takashikkbn@gmail.com> Contributor: wanabe <s.wanabe@gmail.com> Part of [Feature #14235] --- * Known issues * Code generated by gcc is faster than clang. The benchmark may be worse in macOS. Following benchmark result is provided by gcc w/ Linux. * Performance is decreased when Google Chrome is running * JIT can work on MinGW, but it doesn't improve performance at least in short running benchmark. * Currently it doesn't perform well with Rails. We'll try to fix this before release. --- * Benchmark reslts Benchmarked with: Intel 4.0GHz i7-4790K with 16GB memory under x86-64 Ubuntu 8 Cores - 2.0.0-p0: Ruby 2.0.0-p0 - r62186: Ruby trunk (early 2.6.0), before MJIT changes - JIT off: On this commit, but without `--jit` option - JIT on: On this commit, and with `--jit` option ** Optcarrot fps Benchmark: https://github.com/mame/optcarrot | |2.0.0-p0 |r62186 |JIT off |JIT on | |:--------|:--------|:--------|:--------|:--------| |fps |37.32 |51.46 |51.31 |58.88 | |vs 2.0.0 |1.00x |1.38x |1.37x |1.58x | ** MJIT benchmarks Benchmark: https://github.com/benchmark-driver/mjit-benchmarks (Original: https://github.com/vnmakarov/ruby/tree/rtl_mjit_branch/MJIT-benchmarks) | |2.0.0-p0 |r62186 |JIT off |JIT on | |:----------|:--------|:--------|:--------|:--------| |aread |1.00 |1.09 |1.07 |2.19 | |aref |1.00 |1.13 |1.11 |2.22 | |aset |1.00 |1.50 |1.45 |2.64 | |awrite |1.00 |1.17 |1.13 |2.20 | |call |1.00 |1.29 |1.26 |2.02 | |const2 |1.00 |1.10 |1.10 |2.19 | |const |1.00 |1.11 |1.10 |2.19 | |fannk |1.00 |1.04 |1.02 |1.00 | |fib |1.00 |1.32 |1.31 |1.84 | |ivread |1.00 |1.13 |1.12 |2.43 | |ivwrite |1.00 |1.23 |1.21 |2.40 | |mandelbrot |1.00 |1.13 |1.16 |1.28 | |meteor |1.00 |2.97 |2.92 |3.17 | |nbody |1.00 |1.17 |1.15 |1.49 | |nest-ntimes|1.00 |1.22 |1.20 |1.39 | |nest-while |1.00 |1.10 |1.10 |1.37 | |norm |1.00 |1.18 |1.16 |1.24 | |nsvb |1.00 |1.16 |1.16 |1.17 | |red-black |1.00 |1.02 |0.99 |1.12 | |sieve |1.00 |1.30 |1.28 |1.62 | |trees |1.00 |1.14 |1.13 |1.19 | |while |1.00 |1.12 |1.11 |2.41 | ** Discourse's script/bench.rb Benchmark: https://github.com/discourse/discourse/blob/v1.8.7/script/bench.rb NOTE: Rails performance was somehow a little degraded with JIT for now. We should fix this. (At least I know opt_aref is performing badly in JIT and I have an idea to fix it. Please wait for the fix.) *** JIT off Your Results: (note for timings- percentile is first, duration is second in millisecs) categories_admin: 50: 17 75: 18 90: 22 99: 29 home_admin: 50: 21 75: 21 90: 27 99: 40 topic_admin: 50: 17 75: 18 90: 22 99: 32 categories: 50: 35 75: 41 90: 43 99: 77 home: 50: 39 75: 46 90: 49 99: 95 topic: 50: 46 75: 52 90: 56 99: 101 *** JIT on Your Results: (note for timings- percentile is first, duration is second in millisecs) categories_admin: 50: 19 75: 21 90: 25 99: 33 home_admin: 50: 24 75: 26 90: 30 99: 35 topic_admin: 50: 19 75: 20 90: 25 99: 30 categories: 50: 40 75: 44 90: 48 99: 76 home: 50: 42 75: 48 90: 51 99: 89 topic: 50: 49 75: 55 90: 58 99: 99 git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62197 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 14:22:28 +03:00
MJIT_FUNC_EXPORTED VALUE
rb_keyword_error_new(const char *error, VALUE keys)
{
long i = 0, len = RARRAY_LEN(keys);
VALUE error_message = rb_sprintf("%s keyword%.*s", error, len > 1, "s");
if (len > 0) {
rb_str_cat_cstr(error_message, ": ");
while (1) {
const VALUE k = RARRAY_AREF(keys, i);
rb_str_append(error_message, rb_inspect(k));
if (++i >= len) break;
rb_str_cat_cstr(error_message, ", ");
}
}
return rb_exc_new_str(rb_eArgError, error_message);
}
NORETURN(static void rb_keyword_error(const char *error, VALUE keys));
static void
rb_keyword_error(const char *error, VALUE keys)
{
rb_exc_raise(rb_keyword_error_new(error, keys));
}
NORETURN(static void unknown_keyword_error(VALUE hash, const ID *table, int keywords));
static void
unknown_keyword_error(VALUE hash, const ID *table, int keywords)
{
int i;
for (i = 0; i < keywords; i++) {
st_data_t key = ID2SYM(table[i]);
rb_hash_stlike_delete(hash, &key, NULL);
}
rb_keyword_error("unknown", rb_hash_keys(hash));
}
static int
separate_symbol(st_data_t key, st_data_t value, st_data_t arg)
{
VALUE *kwdhash = (VALUE *)arg;
if (!SYMBOL_P(key)) kwdhash++;
if (!*kwdhash) *kwdhash = rb_hash_new();
rb_hash_aset(*kwdhash, (VALUE)key, (VALUE)value);
return ST_CONTINUE;
}
VALUE
rb_extract_keywords(VALUE *orighash)
{
VALUE parthash[2] = {0, 0};
VALUE hash = *orighash;
if (RHASH_EMPTY_P(hash)) {
*orighash = 0;
return hash;
}
rb_hash_foreach(hash, separate_symbol, (st_data_t)&parthash);
*orighash = parthash[1];
if (parthash[1] && RBASIC_CLASS(hash) != rb_cHash) {
RBASIC_SET_CLASS(parthash[1], RBASIC_CLASS(hash));
}
return parthash[0];
}
int
rb_get_kwargs(VALUE keyword_hash, const ID *table, int required, int optional, VALUE *values)
{
int i = 0, j;
int rest = 0;
VALUE missing = Qnil;
st_data_t key;
#define extract_kwarg(keyword, val) \
(key = (st_data_t)(keyword), values ? \
2019-07-16 02:48:57 +03:00
(rb_hash_stlike_delete(keyword_hash, &key, &(val)) || ((val) = Qundef, 0)) : \
rb_hash_stlike_lookup(keyword_hash, key, NULL))
if (NIL_P(keyword_hash)) keyword_hash = 0;
if (optional < 0) {
rest = 1;
optional = -1-optional;
}
if (required) {
for (; i < required; i++) {
VALUE keyword = ID2SYM(table[i]);
if (keyword_hash) {
2019-07-16 03:22:11 +03:00
if (extract_kwarg(keyword, values[i])) {
continue;
}
}
if (NIL_P(missing)) missing = rb_ary_tmp_new(1);
rb_ary_push(missing, keyword);
}
if (!NIL_P(missing)) {
rb_keyword_error("missing", missing);
}
}
j = i;
if (optional && keyword_hash) {
for (i = 0; i < optional; i++) {
2019-07-16 03:22:11 +03:00
if (extract_kwarg(ID2SYM(table[required+i]), values[required+i])) {
j++;
}
}
}
if (!rest && keyword_hash) {
if (RHASH_SIZE(keyword_hash) > (unsigned int)(values ? 0 : j)) {
unknown_keyword_error(keyword_hash, table, required+optional);
}
}
2019-07-16 02:48:57 +03:00
if (values && !keyword_hash) {
2019-07-16 03:22:11 +03:00
for (i = 0; i < required + optional; i++) {
values[i] = Qundef;
}
2019-07-16 02:48:57 +03:00
}
return j;
#undef extract_kwarg
}
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
struct rb_scan_args_t {
int kw_flag;
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
int n_lead;
int n_opt;
int n_trail;
bool f_var;
bool f_hash;
bool f_block;
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
};
static void
rb_scan_args_parse(int kw_flag, const char *fmt, struct rb_scan_args_t *arg)
{
const char *p = fmt;
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
memset(arg, 0, sizeof(*arg));
arg->kw_flag = kw_flag;
if (ISDIGIT(*p)) {
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
arg->n_lead = *p - '0';
p++;
if (ISDIGIT(*p)) {
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
arg->n_opt = *p - '0';
p++;
}
}
if (*p == '*') {
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
arg->f_var = 1;
p++;
}
if (ISDIGIT(*p)) {
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
arg->n_trail = *p - '0';
p++;
}
if (*p == ':') {
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
arg->f_hash = 1;
p++;
}
if (*p == '&') {
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
arg->f_block = 1;
p++;
}
if (*p != '\0') {
rb_fatal("bad scan arg format: %s", fmt);
}
}
static int
rb_scan_args_assign(const struct rb_scan_args_t *arg, int argc, const VALUE *const argv, va_list vargs)
{
int i, argi = 0;
VALUE *var, hash = Qnil;
#define rb_scan_args_next_param() va_arg(vargs, VALUE *)
const int kw_flag = arg->kw_flag;
const int n_lead = arg->n_lead;
const int n_opt = arg->n_opt;
const int n_trail = arg->n_trail;
const int n_mand = n_lead + n_trail;
const bool f_var = arg->f_var;
const bool f_hash = arg->f_hash;
const bool f_block = arg->f_block;
/* capture an option hash - phase 1: pop from the argv */
if (f_hash && argc > 0) {
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
VALUE last = argv[argc - 1];
if (rb_scan_args_keyword_p(kw_flag, last)) {
hash = rb_hash_dup(last);
argc--;
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
}
}
if (argc < n_mand) {
goto argc_error;
}
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
/* capture leading mandatory arguments */
for (i = 0; i < n_lead; i++) {
var = rb_scan_args_next_param();
if (var) *var = argv[argi];
argi++;
}
/* capture optional arguments */
for (i = 0; i < n_opt; i++) {
var = rb_scan_args_next_param();
if (argi < argc - n_trail) {
if (var) *var = argv[argi];
argi++;
}
else {
if (var) *var = Qnil;
}
}
/* capture variable length arguments */
if (f_var) {
int n_var = argc - argi - n_trail;
var = rb_scan_args_next_param();
if (0 < n_var) {
if (var) *var = rb_ary_new_from_values(n_var, &argv[argi]);
argi += n_var;
}
else {
if (var) *var = rb_ary_new();
}
}
/* capture trailing mandatory arguments */
for (i = 0; i < n_trail; i++) {
var = rb_scan_args_next_param();
if (var) *var = argv[argi];
argi++;
}
/* capture an option hash - phase 2: assignment */
if (f_hash) {
var = rb_scan_args_next_param();
if (var) *var = hash;
}
/* capture iterator block */
if (f_block) {
var = rb_scan_args_next_param();
if (rb_block_given_p()) {
*var = rb_block_proc();
}
else {
*var = Qnil;
}
}
if (argi == argc) {
return argc;
}
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
argc_error:
return -(argc + 1);
#undef rb_scan_args_next_param
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
}
2020-02-14 09:38:42 +03:00
static int
rb_scan_args_result(const struct rb_scan_args_t *const arg, int argc)
{
const int n_lead = arg->n_lead;
const int n_opt = arg->n_opt;
const int n_trail = arg->n_trail;
const int n_mand = n_lead + n_trail;
const bool f_var = arg->f_var;
if (argc >= 0) {
return argc;
2020-02-14 09:38:42 +03:00
}
argc = -argc - 1;
rb_error_arity(argc, n_mand, f_var ? UNLIMITED_ARGUMENTS : n_mand + n_opt);
UNREACHABLE_RETURN(-1);
2020-02-14 09:38:42 +03:00
}
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
#undef rb_scan_args
int
rb_scan_args(int argc, const VALUE *argv, const char *fmt, ...)
{
va_list vargs;
struct rb_scan_args_t arg;
rb_scan_args_parse(RB_SCAN_ARGS_PASS_CALLED_KEYWORDS, fmt, &arg);
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
va_start(vargs,fmt);
argc = rb_scan_args_assign(&arg, argc, argv, vargs);
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
va_end(vargs);
2020-02-14 09:38:42 +03:00
return rb_scan_args_result(&arg, argc);
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
}
2020-02-03 07:07:34 +03:00
#undef rb_scan_args_kw
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
int
rb_scan_args_kw(int kw_flag, int argc, const VALUE *argv, const char *fmt, ...)
{
va_list vargs;
struct rb_scan_args_t arg;
rb_scan_args_parse(kw_flag, fmt, &arg);
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
va_start(vargs,fmt);
argc = rb_scan_args_assign(&arg, argc, argv, vargs);
Make rb_scan_args handle keywords more similar to Ruby methods (#2460) Cfuncs that use rb_scan_args with the : entry suffer similar keyword argument separation issues that Ruby methods suffer if the cfuncs accept optional or variable arguments. This makes the following changes to : handling. * Treats as **kw, prompting keyword argument separation warnings if called with a positional hash. * Do not look for an option hash if empty keywords are provided. For backwards compatibility, treat an empty keyword splat as a empty mandatory positional hash argument, but emit a a warning, as this behavior will be removed in Ruby 3. The argument number check needs to be moved lower so it can correctly handle an empty positional argument being added. * If the last argument is nil and it is necessary to treat it as an option hash in order to make sure all arguments are processed, continue to treat the last argument as the option hash. Emit a warning in this case, as this behavior will be removed in Ruby 3. * If splitting the keyword hash into two hashes, issue a warning, as we will not be splitting hashes in Ruby 3. * If the keyword argument is required to fill a mandatory positional argument, continue to do so, but emit a warning as this behavior will be going away in Ruby 3. * If keyword arguments are provided and the last argument is not a hash, that indicates something wrong. This can happen if a cfunc is calling rb_scan_args multiple times, and providing arguments that were not passed to it from Ruby. Callers need to switch to the new rb_scan_args_kw function, which allows passing of whether keywords were provided. This commit fixes all warnings caused by the changes above. It switches some function calls to *_kw versions with appropriate kw_splat flags. If delegating arguments, RB_PASS_CALLED_KEYWORDS is used. If creating new arguments, RB_PASS_KEYWORDS is used if the last argument is a hash to be treated as keywords. In open_key_args in io.c, use rb_scan_args_kw. In this case, the arguments provided come from another C function, not Ruby. The last argument may or may not be a hash, so we can't set keyword argument mode. However, if it is a hash, we don't want to warn when treating it as keywords. In Ruby files, make sure to appropriately use keyword splats or literal keywords when calling Cfuncs that now issue keyword argument separation warnings through rb_scan_args. Also, make sure not to pass nil in place of an option hash. Work around Kernel#warn warnings due to problems in the Rubygems override of the method. There is an open pull request to fix these issues in Rubygems, but part of the Rubygems tests for their override fail on ruby-head due to rb_scan_args not recognizing empty keyword splats, which this commit fixes. Implementation wise, adding rb_scan_args_kw is kind of a pain, because rb_scan_args takes a variable number of arguments. In order to not duplicate all the code, the function internals need to be split into two functions taking a va_list, and to avoid passing in a ton of arguments, a single struct argument is used to handle the variables previously local to the function.
2019-09-25 21:18:49 +03:00
va_end(vargs);
2020-02-14 09:38:42 +03:00
return rb_scan_args_result(&arg, argc);
}
/*!
* \}
*/