Bug 65286 - V8 should correctly serialize Boolean, Number and String objects
: V8 should correctly serialize Boolean, Number and String objects
Status: RESOLVED FIXED
: WebKit
New Bugs
: 528+ (Nightly build)
: Unspecified Unspecified
: P2 Normal
Assigned To:
:
:
:
: 65292 100404
  Show dependency treegraph
 
Reported: 2011-07-27 14:43 PST by
Modified: 2012-10-25 12:11 PST (History)


Attachments
Changes to SerializedScriptValue + window-postmessage-clone. (20.89 KB, patch)
2011-07-27 14:50 PST, Luke Zarko
levin: review-
Review Patch | Details | Formatted Diff | Diff
Address comments. (20.89 KB, patch)
2011-07-27 15:38 PST, Luke Zarko
levin: review+
webkit.review.bot: commit‑queue-
Review Patch | Details | Formatted Diff | Diff
Fix a broken expectation (22.59 KB, patch)
2011-07-28 11:25 PST, Luke Zarko
no flags Review Patch | Details | Formatted Diff | Diff


Note

You need to log in before you can comment on or make changes to this bug.


Description From 2011-07-27 14:43:10 PST
New APIs were exposed to allow V8 to correctly serialize Boolean, Number and String objects, as well as detect certain native errors as required by the HTML5 Structured Clone algorithm (http://www.whatwg.org/specs/web-apps/current-work/multipage/urls.html#safe-passing-of-structured-data). SerializedScriptValue should make use of these APIs.
------- Comment #1 From 2011-07-27 14:50:26 PST -------
Created an attachment (id=102188) [details]
Changes to SerializedScriptValue + window-postmessage-clone.
------- Comment #2 From 2011-07-27 15:29:53 PST -------
(From update of attachment 102188 [details])
View in context: https://bugs.webkit.org/attachment.cgi?id=102188&action=review

Just a few minor things to clean up.

Thanks!

> LayoutTests/ChangeLog:14
> +        ():

Either fix this by adding a function name or remove it.

> LayoutTests/fast/dom/Window/script-tests/postmessage-clone.js:224
> +        'return [t1,t1,t2,f1,f1,f2]'

put spaces after commas

> LayoutTests/fast/dom/Window/window-postmessage-clone-expected.txt:6
> +FAIL: 'postMessage(var x = 0; try { eval("badref"); } catch(e) { x = e; } x)' should throw but didn't

We should file a bug for these failures for SquirellFish.
------- Comment #3 From 2011-07-27 15:38:09 PST -------
Created an attachment (id=102198) [details]
Address comments.
------- Comment #4 From 2011-07-27 15:49:43 PST -------
(In reply to comment #3)
> Created an attachment (id=102198) [details] [details]
> Address comments.

Filed a bug for the JSC bindings at https://bugs.webkit.org/show_bug.cgi?id=65292 .
------- Comment #5 From 2011-07-27 21:10:14 PST -------
(From update of attachment 102198 [details])
Attachment 102198 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/9265064

New failing tests:
fast/loader/stateobjects/pushstate-object-types.html
------- Comment #6 From 2011-07-28 11:25:08 PST -------
Created an attachment (id=102271) [details]
Fix a broken expectation

The test failure above came from an expectation that Error objects could be serialized. In the new revision of the standard this is no longer the case.
------- Comment #7 From 2011-07-28 16:56:41 PST -------
(From update of attachment 102271 [details])
Clearing flags on attachment: 102271

Committed r91959: <http://trac.webkit.org/changeset/91959>
------- Comment #8 From 2011-07-28 16:56:45 PST -------
All reviewed patches have been landed.  Closing bug.