<rdar://problem/105509452>

Created attachment 465009 [details]
Patch

rdar://105470972

Comment on attachment 465009 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=465009&action=review

> Source/WebKit/WebProcess/Plugins/PDF/PDFPlugin.mm:451
> +    callOnMainRunLoopAndWait([&] {

do we need to wait for this?

(In reply to chris fleizach from comment #4)
> Comment on attachment 465009 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=465009&action=review
> 
> > Source/WebKit/WebProcess/Plugins/PDF/PDFPlugin.mm:451
> > +    callOnMainRunLoopAndWait([&] {
> 
> do we need to wait for this?
Perhaps not, but this function is called from non-accessibility codepaths too, and I thought it might be safer to preserve the synchronous behavior in case some caller depends on that.

Created attachment 465011 [details]
Patch

Actually I believe these other non-accessibility codepaths will be going through the main runloop, meaning `callOnMainRunLoop` (without the andWait) would be sync. Uploaded a new patch using callOnMainRunLoop, please re-review when you get a moment.

Created attachment 465012 [details]
Patch

Committed 260451@main (83013c6d19eb): <https://commits.webkit.org/260451@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 465012 [details].