Primitive this values of accessors should not be coerced in strict mode #3854
Conversation
dbatyai
added
bug
jerry-core/vm/vm.c
Outdated
|
|
||
| if (JERRY_UNLIKELY (!ecma_is_value_object (object))) | ||
| if (JERRY_UNLIKELY (!ecma_is_value_object (base) |
There was a problem hiding this comment.
I think this part is always true if the second part is true.
There was a problem hiding this comment.
Since this case is extremely unlikely, I added the first condition to reduce the number of checks when base is an object, which is more likely. I can remove it if you want.
There was a problem hiding this comment.
This looks like a very hoth path, and this is checked twice. What about moving the ecma_is_value_prop_name part before this, and combine this part with the rest?
There was a problem hiding this comment.
Unfortunatley the errors must be thrown in this order, the base object needs to be checked first, followed by the property name. The checks could be reduced by adding a completely separate code path for the non-object case and duplicating/factoring out the property name coercion. What do you think?
There was a problem hiding this comment.
This is a key function of vm, so in this case I would not mind the extra binary size.
| JERRY_ASSERT (!ECMA_IS_VALUE_ERROR (object)); | ||
|
|
||
| object_p = ecma_get_object_from_value (object); | ||
| ecma_op_ordinary_object_prevent_extensions (object_p); |
There was a problem hiding this comment.
This has been added as part of #2917, it seems to me that it's correct.
JerryScript-DCO-1.0-Signed-off-by: Dániel Bátyai [email protected]
JerryScript-DCO-1.0-Signed-off-by: Dániel Bátyai [email protected]