1908512 - nsTextFrame could set the NS_HAS_FLOWLENGTH_PROPERTY flag on content even if the property setter failed

Status: RESOLVED FIXED

(Core :: Layout: Text and Fonts, defect)

Description

[Jonathan Kew [:jfkthame]]

Noticed in passing: in nsTextFrame::GetInFlowContentLength, we would end up setting the NS_HAS_FLOWLENGTH_PROPERTY flag even if the SetProperty call returned failure, which means the flag would be inconsistent with the actual presence (or not) of the property.

Fortunately, it doesn't look like this would lead to any visible change in behavior, or any security issue; it's just an internal inconsistency.

Comment 1

[Jonathan Kew [:jfkthame]]

Attachment: Bug 1908512 - Don't set the NS_HAS_FLOWLENGTH_PROPERTY flag if the property setter actually failed. r=#layout

Doesn't change observable behavior.

Comment 2

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/d2908fbd78d2
Don't set the NS_HAS_FLOWLENGTH_PROPERTY flag if the property setter actually failed. r=layout-reviewers,emilio

Comment 3

[Serban Stanca [:SerbanS]]

https://hg.mozilla.org/mozilla-central/rev/d2908fbd78d2

Comment 4

[Joel Maher ( :jmaher ) (UTC -8) (PTO back normal Nov 17)]

Attachment: Bug 1909512 - Do not add .list manifests to first chunk on backstop pushes. r=ahal

Comment 5

[Phabricator Automation]

Comment on attachment 9418583 [details]
Bug 1909512 - Do not add .list manifests to first chunk on backstop pushes. r=ahal

Revision D218959 was moved to bug 1909512. Setting attachment 9418583 [details] to obsolete.