Bug 78434 - REGRESSION (r107498): broke two tests on GTK/Qt
Summary: REGRESSION (r107498): broke two tests on GTK/Qt
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKitGTK (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-02-12 02:57 PST by Philippe Normand
Modified: 2012-02-13 02:26 PST (History)
2 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Philippe Normand 2012-02-12 02:57:35 PST
--- /home/slave/webkitgtk/gtk-linux-64-release/build/layout-test-results/fast/dom/Window/window-custom-prototype-expected.txt 
+++ /home/slave/webkitgtk/gtk-linux-64-release/build/layout-test-results/fast/dom/Window/window-custom-prototype-actual.txt 
@@ -3,12 +3,12 @@
 On success, you will see a series of "PASS" messages, followed by "TEST COMPLETE".
 
 
-PASS __proto__ = window; __proto threw exception Error: cyclic __proto__ value.
-PASS __proto__ = chainPointingBackToWindow; __proto__ threw exception Error: cyclic __proto__ value.
-PASS __proto__ = 1; __proto__ is originalWindowPrototype
-PASS __proto__ = 'a string'; __proto__ is originalWindowPrototype
+FAIL __proto__ = window; __proto should throw Error: cyclic __proto__ value. Threw exception ReferenceError: Can't find variable: __proto.
+FAIL __proto__ = chainPointingBackToWindow; __proto__ should throw Error: cyclic __proto__ value. Was [object Object].
+FAIL __proto__ = 1; __proto__ should be [object DOMWindowPrototype]. Was [object Object].
+FAIL __proto__ = 'a string'; __proto__ should be [object DOMWindowPrototype]. Was [object Object].
 PASS __proto__ = anotherObject; __proto__ is anotherObject
-PASS anotherObject.__proto__ = window; __proto__ threw exception Error: cyclic __proto__ value.
+FAIL anotherObject.__proto__ = window; __proto__ should throw Error: cyclic __proto__ value. Was [object Object].
 PASS __proto__ = null; __proto__ is null
 PASS __proto__ = 1; __proto__ is null
 PASS __proto__ = 'a string'; __proto__ is null

--- /home/slave/webkitgtk/gtk-linux-64-release/build/layout-test-results/http/tests/security/cross-frame-access-object-getPrototypeOf-expected.txt 
+++ /home/slave/webkitgtk/gtk-linux-64-release/build/layout-test-results/http/tests/security/cross-frame-access-object-getPrototypeOf-actual.txt 
@@ -1,9 +1,5 @@
-CONSOLE MESSAGE: Unsafe JavaScript attempt to access frame with URL http://localhost:8000/security/resources/cross-frame-iframe-for-object-getPrototypeOf-test.html from frame with URL http://127.0.0.1:8000/security/cross-frame-access-object-getPrototypeOf.html. Domains, protocols and ports must match.
-
-CONSOLE MESSAGE: Unsafe JavaScript attempt to access frame with URL http://localhost:8000/security/resources/cross-frame-iframe-for-object-getPrototypeOf-test.html from frame with URL http://127.0.0.1:8000/security/cross-frame-access-object-getPrototypeOf.html. Domains, protocols and ports must match.
-
 This tests that you can't get the prototype of the window or history objects cross-origin using Object.getPrototypeOf().
 
-PASS: Object.getPrototypeOf(targetWindow) should be 'undefined' and is.
-PASS: Object.getPrototypeOf(targetWindow.history) should be 'undefined' and is.
+*** FAIL: Object.getPrototypeOf(targetWindow) should be 'undefined' but instead is [object DOMWindowPrototype]. ***
+*** FAIL: Object.getPrototypeOf(targetWindow.history) should be 'undefined' but instead is [object HistoryPrototype]. ***
Comment 1 Philippe Normand 2012-02-12 03:02:08 PST
Mac is also affected by this regression.
Comment 2 Csaba Osztrogonác 2012-02-12 10:38:18 PST
I think these are not regression, but progression and these tests needs to be updated. I'll check them.
Comment 3 Gavin Barraclough 2012-02-12 10:51:43 PST
(In reply to comment #2)
> I think these are not regression, but progression and these tests needs to be updated. I'll check them.

Hey Ossy, let me take a look at these.  The first is probably largely a progression, but I might add a couple of extra tests, but the latter is probably a genuine regression I introduced.  Will fix.

cheers,
G.
Comment 4 Gavin Barraclough 2012-02-13 01:29:50 PST
I've temporarily reverted r107498 to easy time pressure, will re-land with these tests passing.
Fixed in r107544.
Comment 5 Csaba Osztrogonác 2012-02-13 02:26:15 PST
(In reply to comment #4)
> I've temporarily reverted r107498 to easy time pressure, will re-land with these tests passing.
> Fixed in r107544.

Thanks, I reverted followup expected updates too - http://trac.webkit.org/changeset/107552