ruby/eval_intern.h

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

325 строки
8.3 KiB
C
Исходник Обычный вид История

#ifndef RUBY_EVAL_INTERN_H
#define RUBY_EVAL_INTERN_H
#include "ruby/ruby.h"
#include "vm_core.h"
static inline void
vm_passed_block_handler_set(rb_execution_context_t *ec, VALUE block_handler)
{
vm_block_handler_verify(block_handler);
ec->passed_block_handler = block_handler;
}
static inline void
pass_passed_block_handler(rb_execution_context_t *ec)
{
VALUE block_handler = rb_vm_frame_block_handler(ec->cfp);
vm_passed_block_handler_set(ec, block_handler);
VM_ENV_FLAGS_SET(ec->cfp->ep, VM_FRAME_FLAG_PASSED);
}
#define PASS_PASSED_BLOCK_HANDLER_EC(ec) pass_passed_block_handler(ec)
#define PASS_PASSED_BLOCK_HANDLER() pass_passed_block_handler(GET_EC())
#ifdef HAVE_STDLIB_H
#include <stdlib.h>
#endif
#ifndef EXIT_SUCCESS
#define EXIT_SUCCESS 0
#endif
#ifndef EXIT_FAILURE
#define EXIT_FAILURE 1
#endif
#include <stdio.h>
#if defined(__wasm__) && !defined(__EMSCRIPTEN__)
# include "wasm/setjmp.h"
#else
# include <setjmp.h>
#endif
#ifdef __APPLE__
# ifdef HAVE_CRT_EXTERNS_H
# include <crt_externs.h>
# else
# include "missing/crt_externs.h"
# endif
#endif
#ifndef HAVE_STRING_H
char *strrchr(const char *, const char);
#endif
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif
#ifdef HAVE_NET_SOCKET_H
#include <net/socket.h>
#endif
#define ruby_setjmp(env) RUBY_SETJMP(env)
#define ruby_longjmp(env,val) RUBY_LONGJMP((env),(val))
#ifdef __CYGWIN__
# ifndef _setjmp
int _setjmp(jmp_buf);
# endif
# ifndef _longjmp
NORETURN(void _longjmp(jmp_buf, int));
# endif
#endif
#include <sys/types.h>
#include <signal.h>
#include <errno.h>
#ifdef HAVE_SYS_SELECT_H
#include <sys/select.h>
#endif
/*
Solaris sys/select.h switches select to select_large_fdset to support larger
file descriptors if FD_SETSIZE is larger than 1024 on 32bit environment.
But Ruby doesn't change FD_SETSIZE because fd_set is allocated dynamically.
So following definition is required to use select_large_fdset.
*/
#ifdef HAVE_SELECT_LARGE_FDSET
#define select(n, r, w, e, t) select_large_fdset((n), (r), (w), (e), (t))
extern int select_large_fdset(int, fd_set *, fd_set *, fd_set *, struct timeval *);
#endif
#ifdef HAVE_SYS_PARAM_H
#include <sys/param.h>
#endif
#include <sys/stat.h>
#define EC_PUSH_TAG(ec) do { \
rb_execution_context_t * const _ec = (ec); \
struct rb_vm_tag _tag; \
_tag.state = TAG_NONE; \
_tag.tag = Qundef; \
_tag.prev = _ec->tag; \
_tag.lock_rec = rb_ec_vm_lock_rec(_ec); \
rb_vm_tag_jmpbuf_init(&_tag.buf); \
#define EC_POP_TAG() \
_ec->tag = _tag.prev; \
rb_vm_tag_jmpbuf_deinit(&_tag.buf); \
} while (0)
#define EC_TMPPOP_TAG() \
_ec->tag = _tag.prev
#define EC_REPUSH_TAG() (void)(_ec->tag = &_tag)
#if defined __GNUC__ && __GNUC__ == 4 && (__GNUC_MINOR__ >= 6 && __GNUC_MINOR__ <= 8) || defined __clang__
/* This macro prevents GCC 4.6--4.8 from emitting maybe-uninitialized warnings.
* This macro also prevents Clang from dumping core in EC_EXEC_TAG().
* (I confirmed Clang 4.0.1 and 5.0.0.)
*/
# define VAR_FROM_MEMORY(var) __extension__(*(__typeof__(var) volatile *)&(var))
# define VAR_INITIALIZED(var) ((var) = VAR_FROM_MEMORY(var))
# define VAR_NOCLOBBERED(var) volatile var
#else
# define VAR_FROM_MEMORY(var) (var)
# define VAR_INITIALIZED(var) ((void)&(var))
# define VAR_NOCLOBBERED(var) var
#endif
static inline void
rb_ec_vm_lock_rec_check(const rb_execution_context_t *ec, unsigned int recorded_lock_rec)
{
unsigned int current_lock_rec = rb_ec_vm_lock_rec(ec);
if (current_lock_rec != recorded_lock_rec) {
rb_ec_vm_lock_rec_release(ec, recorded_lock_rec, current_lock_rec);
}
}
/* clear ec->tag->state, and return the value */
static inline int
rb_ec_tag_state(const rb_execution_context_t *ec)
{
struct rb_vm_tag *tag = ec->tag;
enum ruby_tag_type state = tag->state;
tag->state = TAG_NONE;
rb_ec_vm_lock_rec_check(ec, tag->lock_rec);
2023-08-07 18:28:36 +03:00
RBIMPL_ASSUME(state > TAG_NONE);
RBIMPL_ASSUME(state <= TAG_FATAL);
return state;
}
NORETURN(static inline void rb_ec_tag_jump(const rb_execution_context_t *ec, enum ruby_tag_type st));
static inline void
rb_ec_tag_jump(const rb_execution_context_t *ec, enum ruby_tag_type st)
{
2023-08-07 18:28:36 +03:00
RUBY_ASSERT(st > TAG_NONE && st <= TAG_FATAL, ": Invalid tag jump: %d", (int)st);
ec->tag->state = st;
ruby_longjmp(RB_VM_TAG_JMPBUF_GET(ec->tag->buf), 1);
}
/*
setjmp() in assignment expression rhs is undefined behavior
[ISO/IEC 9899:1999] 7.13.1.1
*/
#define EC_EXEC_TAG() \
(UNLIKELY(ruby_setjmp(RB_VM_TAG_JMPBUF_GET(_tag.buf))) ? rb_ec_tag_state(VAR_FROM_MEMORY(_ec)) : (EC_REPUSH_TAG(), 0))
#define EC_JUMP_TAG(ec, st) rb_ec_tag_jump(ec, st)
#define INTERNAL_EXCEPTION_P(exc) FIXNUM_P(exc)
/* CREF operators */
#define CREF_FL_PUSHED_BY_EVAL IMEMO_FL_USER1
#define CREF_FL_OMOD_SHARED IMEMO_FL_USER2
Lazily create singletons on instance_{exec,eval} (#5146) * Lazily create singletons on instance_{exec,eval} Previously when instance_exec or instance_eval was called on an object, that object would be given a singleton class so that method definitions inside the block would be added to the object rather than its class. This commit aims to improve performance by delaying the creation of the singleton class unless/until one is needed for method definition. Most of the time instance_eval is used without any method definition. This was implemented by adding a flag to the cref indicating that it represents a singleton of the object rather than a class itself. In this case CREF_CLASS returns the object's existing class, but in cases that we are defining a method (either via definemethod or VM_SPECIAL_OBJECT_CBASE which is used for undef and alias). This also happens to fix what I believe is a bug. Previously instance_eval behaved differently with regards to constant access for true/false/nil than for all other objects. I don't think this was intentional. String::Foo = "foo" "".instance_eval("Foo") # => "foo" Integer::Foo = "foo" 123.instance_eval("Foo") # => "foo" TrueClass::Foo = "foo" true.instance_eval("Foo") # NameError: uninitialized constant Foo This also slightly changes the error message when trying to define a method through instance_eval on an object which can't have a singleton class. Before: $ ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': no class/module to add method (TypeError) After: $ ./ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) IMO this error is a small improvement on the original and better matches the (both old and new) message when definging a method using `def self.` $ ruby -e '123.instance_eval{ def self.foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) Co-authored-by: Matthew Draper <matthew@trebex.net> * Remove "under" argument from yield_under * Move CREF_SINGLETON_SET into vm_cref_new * Simplify vm_get_const_base * Fix leaf VM_SPECIAL_OBJECT_CONST_BASE Co-authored-by: Matthew Draper <matthew@trebex.net>
2021-12-03 02:53:39 +03:00
#define CREF_FL_SINGLETON IMEMO_FL_USER3
static inline int CREF_SINGLETON(const rb_cref_t *cref);
static inline VALUE
CREF_CLASS(const rb_cref_t *cref)
{
Lazily create singletons on instance_{exec,eval} (#5146) * Lazily create singletons on instance_{exec,eval} Previously when instance_exec or instance_eval was called on an object, that object would be given a singleton class so that method definitions inside the block would be added to the object rather than its class. This commit aims to improve performance by delaying the creation of the singleton class unless/until one is needed for method definition. Most of the time instance_eval is used without any method definition. This was implemented by adding a flag to the cref indicating that it represents a singleton of the object rather than a class itself. In this case CREF_CLASS returns the object's existing class, but in cases that we are defining a method (either via definemethod or VM_SPECIAL_OBJECT_CBASE which is used for undef and alias). This also happens to fix what I believe is a bug. Previously instance_eval behaved differently with regards to constant access for true/false/nil than for all other objects. I don't think this was intentional. String::Foo = "foo" "".instance_eval("Foo") # => "foo" Integer::Foo = "foo" 123.instance_eval("Foo") # => "foo" TrueClass::Foo = "foo" true.instance_eval("Foo") # NameError: uninitialized constant Foo This also slightly changes the error message when trying to define a method through instance_eval on an object which can't have a singleton class. Before: $ ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': no class/module to add method (TypeError) After: $ ./ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) IMO this error is a small improvement on the original and better matches the (both old and new) message when definging a method using `def self.` $ ruby -e '123.instance_eval{ def self.foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) Co-authored-by: Matthew Draper <matthew@trebex.net> * Remove "under" argument from yield_under * Move CREF_SINGLETON_SET into vm_cref_new * Simplify vm_get_const_base * Fix leaf VM_SPECIAL_OBJECT_CONST_BASE Co-authored-by: Matthew Draper <matthew@trebex.net>
2021-12-03 02:53:39 +03:00
if (CREF_SINGLETON(cref)) {
return CLASS_OF(cref->klass_or_self);
2021-12-24 13:52:44 +03:00
}
else {
Lazily create singletons on instance_{exec,eval} (#5146) * Lazily create singletons on instance_{exec,eval} Previously when instance_exec or instance_eval was called on an object, that object would be given a singleton class so that method definitions inside the block would be added to the object rather than its class. This commit aims to improve performance by delaying the creation of the singleton class unless/until one is needed for method definition. Most of the time instance_eval is used without any method definition. This was implemented by adding a flag to the cref indicating that it represents a singleton of the object rather than a class itself. In this case CREF_CLASS returns the object's existing class, but in cases that we are defining a method (either via definemethod or VM_SPECIAL_OBJECT_CBASE which is used for undef and alias). This also happens to fix what I believe is a bug. Previously instance_eval behaved differently with regards to constant access for true/false/nil than for all other objects. I don't think this was intentional. String::Foo = "foo" "".instance_eval("Foo") # => "foo" Integer::Foo = "foo" 123.instance_eval("Foo") # => "foo" TrueClass::Foo = "foo" true.instance_eval("Foo") # NameError: uninitialized constant Foo This also slightly changes the error message when trying to define a method through instance_eval on an object which can't have a singleton class. Before: $ ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': no class/module to add method (TypeError) After: $ ./ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) IMO this error is a small improvement on the original and better matches the (both old and new) message when definging a method using `def self.` $ ruby -e '123.instance_eval{ def self.foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) Co-authored-by: Matthew Draper <matthew@trebex.net> * Remove "under" argument from yield_under * Move CREF_SINGLETON_SET into vm_cref_new * Simplify vm_get_const_base * Fix leaf VM_SPECIAL_OBJECT_CONST_BASE Co-authored-by: Matthew Draper <matthew@trebex.net>
2021-12-03 02:53:39 +03:00
return cref->klass_or_self;
}
}
static inline VALUE
CREF_CLASS_FOR_DEFINITION(const rb_cref_t *cref)
{
if (CREF_SINGLETON(cref)) {
return rb_singleton_class(cref->klass_or_self);
2021-12-24 13:52:44 +03:00
}
else {
Lazily create singletons on instance_{exec,eval} (#5146) * Lazily create singletons on instance_{exec,eval} Previously when instance_exec or instance_eval was called on an object, that object would be given a singleton class so that method definitions inside the block would be added to the object rather than its class. This commit aims to improve performance by delaying the creation of the singleton class unless/until one is needed for method definition. Most of the time instance_eval is used without any method definition. This was implemented by adding a flag to the cref indicating that it represents a singleton of the object rather than a class itself. In this case CREF_CLASS returns the object's existing class, but in cases that we are defining a method (either via definemethod or VM_SPECIAL_OBJECT_CBASE which is used for undef and alias). This also happens to fix what I believe is a bug. Previously instance_eval behaved differently with regards to constant access for true/false/nil than for all other objects. I don't think this was intentional. String::Foo = "foo" "".instance_eval("Foo") # => "foo" Integer::Foo = "foo" 123.instance_eval("Foo") # => "foo" TrueClass::Foo = "foo" true.instance_eval("Foo") # NameError: uninitialized constant Foo This also slightly changes the error message when trying to define a method through instance_eval on an object which can't have a singleton class. Before: $ ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': no class/module to add method (TypeError) After: $ ./ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) IMO this error is a small improvement on the original and better matches the (both old and new) message when definging a method using `def self.` $ ruby -e '123.instance_eval{ def self.foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) Co-authored-by: Matthew Draper <matthew@trebex.net> * Remove "under" argument from yield_under * Move CREF_SINGLETON_SET into vm_cref_new * Simplify vm_get_const_base * Fix leaf VM_SPECIAL_OBJECT_CONST_BASE Co-authored-by: Matthew Draper <matthew@trebex.net>
2021-12-03 02:53:39 +03:00
return cref->klass_or_self;
}
}
static inline rb_cref_t *
CREF_NEXT(const rb_cref_t *cref)
{
return cref->next;
}
static inline const rb_scope_visibility_t *
CREF_SCOPE_VISI(const rb_cref_t *cref)
{
return &cref->scope_visi;
}
static inline VALUE
CREF_REFINEMENTS(const rb_cref_t *cref)
{
return cref->refinements;
}
static inline void
CREF_REFINEMENTS_SET(rb_cref_t *cref, VALUE refs)
{
RB_OBJ_WRITE(cref, &cref->refinements, refs);
}
static inline int
CREF_PUSHED_BY_EVAL(const rb_cref_t *cref)
{
return cref->flags & CREF_FL_PUSHED_BY_EVAL;
}
static inline void
CREF_PUSHED_BY_EVAL_SET(rb_cref_t *cref)
{
cref->flags |= CREF_FL_PUSHED_BY_EVAL;
}
Lazily create singletons on instance_{exec,eval} (#5146) * Lazily create singletons on instance_{exec,eval} Previously when instance_exec or instance_eval was called on an object, that object would be given a singleton class so that method definitions inside the block would be added to the object rather than its class. This commit aims to improve performance by delaying the creation of the singleton class unless/until one is needed for method definition. Most of the time instance_eval is used without any method definition. This was implemented by adding a flag to the cref indicating that it represents a singleton of the object rather than a class itself. In this case CREF_CLASS returns the object's existing class, but in cases that we are defining a method (either via definemethod or VM_SPECIAL_OBJECT_CBASE which is used for undef and alias). This also happens to fix what I believe is a bug. Previously instance_eval behaved differently with regards to constant access for true/false/nil than for all other objects. I don't think this was intentional. String::Foo = "foo" "".instance_eval("Foo") # => "foo" Integer::Foo = "foo" 123.instance_eval("Foo") # => "foo" TrueClass::Foo = "foo" true.instance_eval("Foo") # NameError: uninitialized constant Foo This also slightly changes the error message when trying to define a method through instance_eval on an object which can't have a singleton class. Before: $ ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': no class/module to add method (TypeError) After: $ ./ruby -e '123.instance_eval { def foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) IMO this error is a small improvement on the original and better matches the (both old and new) message when definging a method using `def self.` $ ruby -e '123.instance_eval{ def self.foo; end }' -e:1:in `block in <main>': can't define singleton (TypeError) Co-authored-by: Matthew Draper <matthew@trebex.net> * Remove "under" argument from yield_under * Move CREF_SINGLETON_SET into vm_cref_new * Simplify vm_get_const_base * Fix leaf VM_SPECIAL_OBJECT_CONST_BASE Co-authored-by: Matthew Draper <matthew@trebex.net>
2021-12-03 02:53:39 +03:00
static inline int
CREF_SINGLETON(const rb_cref_t *cref)
{
return cref->flags & CREF_FL_SINGLETON;
}
static inline void
CREF_SINGLETON_SET(rb_cref_t *cref)
{
cref->flags |= CREF_FL_SINGLETON;
}
static inline int
CREF_OMOD_SHARED(const rb_cref_t *cref)
{
return cref->flags & CREF_FL_OMOD_SHARED;
}
static inline void
CREF_OMOD_SHARED_SET(rb_cref_t *cref)
{
cref->flags |= CREF_FL_OMOD_SHARED;
}
static inline void
CREF_OMOD_SHARED_UNSET(rb_cref_t *cref)
{
cref->flags &= ~CREF_FL_OMOD_SHARED;
}
enum {
RAISED_EXCEPTION = 1,
RAISED_STACKOVERFLOW = 2,
RAISED_NOMEMORY = 4
};
#define rb_ec_raised_set(ec, f) ((ec)->raised_flag |= (f))
#define rb_ec_raised_reset(ec, f) ((ec)->raised_flag &= ~(f))
#define rb_ec_raised_p(ec, f) (((ec)->raised_flag & (f)) != 0)
#define rb_ec_raised_clear(ec) ((ec)->raised_flag = 0)
int rb_ec_set_raised(rb_execution_context_t *ec);
int rb_ec_reset_raised(rb_execution_context_t *ec);
int rb_ec_stack_check(rb_execution_context_t *ec);
VALUE rb_f_eval(int argc, const VALUE *argv, VALUE self);
VALUE rb_make_exception(int argc, const VALUE *argv);
NORETURN(void rb_method_name_error(VALUE, VALUE));
NORETURN(void rb_fiber_start(rb_fiber_t*));
NORETURN(void rb_print_undef(VALUE, ID, rb_method_visibility_t));
NORETURN(void rb_print_undef_str(VALUE, VALUE));
NORETURN(void rb_print_inaccessible(VALUE, ID, rb_method_visibility_t));
NORETURN(void rb_vm_localjump_error(const char *,VALUE, int));
2023-08-07 18:32:45 +03:00
NORETURN(void rb_vm_jump_tag_but_local_jump(enum ruby_tag_type));
2023-08-07 18:32:45 +03:00
VALUE rb_vm_make_jump_tag_but_local_jump(enum ruby_tag_type state, VALUE val);
rb_cref_t *rb_vm_cref(void);
rb_cref_t *rb_vm_cref_replace_with_duplicated_cref(void);
VALUE rb_vm_call_cfunc(VALUE recv, VALUE (*func)(VALUE), VALUE arg, VALUE block_handler, VALUE filename);
void rb_vm_set_progname(VALUE filename);
VALUE rb_vm_cbase(void);
/* vm_backtrace.c */
VALUE rb_ec_backtrace_object(const rb_execution_context_t *ec);
VALUE rb_ec_backtrace_str_ary(const rb_execution_context_t *ec, long lev, long n);
VALUE rb_ec_backtrace_location_ary(const rb_execution_context_t *ec, long lev, long n, bool skip_internal);
#ifndef CharNext /* defined as CharNext[AW] on Windows. */
# ifdef HAVE_MBLEN
# define CharNext(p) rb_char_next(p)
static inline char *
rb_char_next(const char *p)
{
if (p) {
int len = mblen(p, RUBY_MBCHAR_MAXSIZE);
p += len > 0 ? len : 1;
}
return (char *)p;
}
# else
# define CharNext(p) ((p) + 1)
# endif
#endif
#endif /* RUBY_EVAL_INTERN_H */