Bug 220102

Summary:

Use WeakHashSet<Page> instead of HashSet<Page*>

Product:

WebKit

Reporter:

Alex Christensen <achristensen>

Component:

New Bugs

Assignee:

Alex Christensen <achristensen>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

darin, ews-watchlist, hi, joepeck, webkit-bug-importer

Priority:

Keywords:

InRadar

Version:

WebKit Nightly Build

Hardware:

Unspecified

OS:

Unspecified

Attachments:

Description	Flags
Patch	none
Patch	none

Description Alex Christensen 2020-12-22 13:30:53 PST

Use WeakHashSet<Page> instead of HashSet<Page*>

Comment 1 Alex Christensen 2020-12-22 13:32:29 PST

Created attachment 416687 [details]
Patch

Comment 2 Radar WebKit Bug Importer 2020-12-29 13:31:12 PST

<rdar://problem/72728030>

Comment 3 Darin Adler 2021-01-02 14:38:30 PST

Comment on attachment 416687 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=416687&action=review

> Source/WebKit/WebProcess/InjectedBundle/InjectedBundle.cpp:173
> +    for (const auto& page : PageGroup::pageGroup(pageGroup->identifier())->pages())

Not sure we should have the const here. For one thing, changing the settings for a page doesn’t seem very "const".

> Source/WebKit/WebProcess/InjectedBundle/InjectedBundle.cpp:289
> +    for (const auto& page : PageGroup::pageGroup(pageGroup->identifier())->pages())

Ditto.

> Source/WebKit/WebProcess/InjectedBundle/InjectedBundle.cpp:345
> +        for (const auto& page : PageGroup::pageGroup(pageGroup->identifier())->pages()) {
> +            for (const auto* frame = &page.mainFrame(); frame; frame = frame->tree().traverseNext()) {

Could omit const here too.

Comment 4 Alex Christensen 2021-01-05 10:07:58 PST

Created attachment 417016 [details]
Patch

Comment 5 Alex Christensen 2021-01-05 10:16:17 PST

http://trac.webkit.org/r271153