diff options
| author | NARUSE, Yui <nurse@users.noreply.github.com> | 2024-03-21 10:45:01 +0900 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-03-21 01:45:01 +0000 |
| commit | a63e979853783601a60228b45741f8b3776e5507 (patch) | |
| tree | 8396278af85906d600486360bfb7e466aaef6161 /internal | |
| parent | a406c516685f1950269c4e43d13cc748f0bfbc06 (diff) | |
merge revision(s) d19d683a354530a27b4cbb049223f8dc70c75849,de1a586ecc2ee7f465f0c0a69291054136a3a819: [Backport #20250] (#10308)
rb_obj_setup: do not copy RUBY_FL_SEEN_OBJ_ID
[Bug #20250]
We're seting up a new instance, so it never had an associated
object_id.
proc.c: get rid of `CLONESETUP`
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
[Bug #20253]
All the way down to Ruby 1.9, `Proc`, `Method`, `UnboundMethod`
and `Binding` always had their own specific clone and dup routine.
This caused various discrepancies with how other objects behave
on `dup` and `clone. [Bug #20250], [Bug #20253].
This commit get rid of `CLONESETUP` and use the the same codepath
as all other types, so ensure consistency.
NB: It's still not accepting the `freeze` keyword argument on `clone`.
Co-Authored-By: Étienne Barrié <etienne.barrie@gmail.com>
Diffstat (limited to 'internal')
| -rw-r--r-- | internal/object.h | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/internal/object.h b/internal/object.h index 06595bdd91..903e2d29a5 100644 --- a/internal/object.h +++ b/internal/object.h @@ -16,6 +16,8 @@ VALUE rb_class_search_ancestor(VALUE klass, VALUE super); NORETURN(void rb_undefined_alloc(VALUE klass)); double rb_num_to_dbl(VALUE val); VALUE rb_obj_dig(int argc, VALUE *argv, VALUE self, VALUE notfound); +VALUE rb_obj_clone_setup(VALUE obj, VALUE clone, VALUE kwfreeze); +VALUE rb_obj_dup_setup(VALUE obj, VALUE dup); VALUE rb_immutable_obj_clone(int, VALUE *, VALUE); VALUE rb_check_convert_type_with_id(VALUE,int,const char*,ID); int rb_bool_expected(VALUE, const char *, int raise); |