Bug 264991

Summary:	'NotEnumerable' causing Selection API 'idlharness' stringification test failure
Product:	WebKit	Reporter:	Ahmad Saleem <ahmad.saleem792>
Component:	HTML Editing	Assignee:	Nobody <webkit-unassigned>
Status:	RESOLVED FIXED
Severity:	Normal	CC:	mike, rniwa, webkit-bug-importer, wenson_hsieh
Priority:	P2	Keywords:	BrowserCompat, InRadar, WPTImpact
Version:	WebKit Nightly Build
Hardware:	Unspecified
OS:	Unspecified

Description Ahmad Saleem 2023-11-16 16:01:34 PST

Hi Team,

While trying to investigate 'browser specific failure', I noticed that Safari / WebKit fails one 'idlharness' test from Selection API:

WPT Test Case: https://wpt.fyi/results/selection/idlharness.window.html?label=master&label=experimental&aligned=&q=safari%3Afail+idlharness

^ Failing: Selection interface: stringifier

While loading 'Assert', I noticed that it was failing here:

Fail	
assert_true(false, "property should be enumerable")
 /resources/idlharness.js:2642:20

and while looking into 'DOMSelection.idl', I noticed that we have following:

https://github.com/WebKit/WebKit/blob/54a799c39b9970cface5982aefe8e5f6cc508bf7/Source/WebCore/page/DOMSelection.idl#L70

[NotEnumerable] DOMString toString();

and change above to:

DOMString toString();

___

It leads us to pass all test cases in 'idlharness'.

I just wanted to raise so we can track this failure and fix the issue.

Thanks!

Comment 1 Ahmad Saleem 2023-11-16 16:46:46 PST

Web-Spec: https://w3c.github.io/selection-api/#dom-selection-stringifier

Comment 2 Ahmad Saleem 2023-11-16 16:52:33 PST

Just to update, we have following:

undefined setBaseAndExtent(Node? baseNode, unsigned long baseOffset, Node? extentNode, unsigned long extentOffset);

but Web-Spec has following:

undefined setBaseAndExtent(Node anchorNode, unsigned long anchorOffset, Node focusNode, unsigned long focusOffset);

___

Web-Spec: https://w3c.github.io/selection-api/#selection-interface

Comment 3 Ryosuke Niwa 2023-11-17 15:46:03 PST

It seems like a good change assuming that other browser engines pass those tests.

Comment 4 Ahmad Saleem 2023-11-17 16:21:17 PST

PR to sync partially with bug - https://github.com/WebKit/WebKit/pull/20696

As for 'Comment 02', it is bit more involved change, will create separate bug for it.

@Ryosuke - thanks for your input. For some reason, we don't have 'idlharness' expectation file but I will run above PR as draft via EWS to see, what this change leads and then will make it 'Open' once everything looks good.

Comment 5 EWS 2023-11-17 23:38:01 PST

Committed 270933@main (6b9b667d2d27): <https://commits.webkit.org/270933@main>

Reviewed commits have been landed. Closing PR #20696 and removing active labels.

Comment 6 Radar WebKit Bug Importer 2023-11-17 23:39:15 PST

<rdar://problem/118594710>