299642 - Javascript functions don't list their "prototype"-property in a "for ... in" statement.

Status: VERIFIED FIXED

(Core :: JavaScript Engine, defect, P3)

Description

[mozilla]

User-Agent:       Mozilla/5.0 (compatible; Konqueror/3.4; Linux) KHTML/3.4.0 (like Gecko)
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.6) 

According to ECMAscript spec section 13.2 and 15.3.5.2 the 
"prototype"-property of Javascript functions is *not* "DontEnum", and 
should hence be visible when doing a "for .. in" of the function object. 

Reproducible: Always

Comment 1

[mozilla]

Attachment: test-case.

the given test-case should print "prototype".

Comment 2

[Brendan Eich [:brendan]]

Ancient bug.  If f.prototype has been used already, it will be enumerated per
the spec.  Any time a JSClass has a non-stub resolve hook implementation, it
needs a non-stub enumerate hook impl too.

/be

Comment 3

[Brendan Eich [:brendan]]

Attachment: fix

Easy, no need for JSCLASS_NEW_ENUMERATE or replication of code from fun_resolve
-- just trigger fun_resolve via OBJ_LOOKUP_PROPERTY.

/be

Comment 4

[Mike Shaver (:shaver emeritus)]

Comment on attachment 188264 [details] [diff] [review]
fix

And lo, it is righteous. r=shaver

Comment 5

[Brendan Eich [:brendan]]

Fixed, thanks.

/be

Comment 6

[Bob Clary [:bc] (inactive)]

I failed to catch it at the time, but this appears to have caused 

ecma/GlobalObject/15.1.2.1-1.js
ecma/GlobalObject/15.1.2.2-1.js
ecma/GlobalObject/15.1.2.3-1.js
ecma/GlobalObject/15.1.2.4.js
ecma/GlobalObject/15.1.2.5-1.js
ecma/GlobalObject/15.1.2.6.js
ecma/GlobalObject/15.1.2.7.js

to fail since the existing tests expected prototype to not be enumerated.

Checking in 15.1.2.1-1.js;
/cvsroot/mozilla/js/tests/ecma/GlobalObject/15.1.2.1-1.js,v  <--  15.1.2.1-1.js
new revision: 1.5; previous revision: 1.4
done
Checking in 15.1.2.2-1.js;
/cvsroot/mozilla/js/tests/ecma/GlobalObject/15.1.2.2-1.js,v  <--  15.1.2.2-1.js
new revision: 1.5; previous revision: 1.4
done
Checking in 15.1.2.3-1.js;
/cvsroot/mozilla/js/tests/ecma/GlobalObject/15.1.2.3-1.js,v  <--  15.1.2.3-1.js
new revision: 1.5; previous revision: 1.4
done
Checking in 15.1.2.4.js;
/cvsroot/mozilla/js/tests/ecma/GlobalObject/15.1.2.4.js,v  <--  15.1.2.4.js
new revision: 1.5; previous revision: 1.4
done
Checking in 15.1.2.5-1.js;
/cvsroot/mozilla/js/tests/ecma/GlobalObject/15.1.2.5-1.js,v  <--  15.1.2.5-1.js
new revision: 1.5; previous revision: 1.4
done
Checking in 15.1.2.6.js;
/cvsroot/mozilla/js/tests/ecma/GlobalObject/15.1.2.6.js,v  <--  15.1.2.6.js
new revision: 1.5; previous revision: 1.4
done
Checking in 15.1.2.7.js;
/cvsroot/mozilla/js/tests/ecma/GlobalObject/15.1.2.7.js,v  <--  15.1.2.7.js
new revision: 1.5; previous revision: 1.4
done

Comment 7

[Bob Clary [:bc] (inactive)]

Checking in 10.1.6-01.js;
/cvsroot/mozilla/js/tests/js1_5/Function/10.1.6-01.js,v  <--  10.1.6-01.js
new revision: 1.2; previous revision: 1.1

Comment 8

[Bob Clary [:bc] (inactive)]

Test cases for this are in ecma_3/Statements/regress-194364.js and
js1_5/Function/10.1.6-01.js.

Comment 9

[Bob Clary [:bc] (inactive)]

verified fixed 200604-1 1.8.0.2, 1.8, 1.9a1 win/mac/linux