<html>
<head>
<base href="https://bugs.webkit.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - Notify client immediately if network session doesn't exist for a synchronous XHR load."
href="https://bugs.webkit.org/show_bug.cgi?id=158239#c6">Comment # 6</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - Notify client immediately if network session doesn't exist for a synchronous XHR load."
href="https://bugs.webkit.org/show_bug.cgi?id=158239">bug 158239</a>
from <span class="vcard"><a class="email" href="mailto:yongjun_zhang@apple.com" title="Yongjun Zhang <yongjun_zhang@apple.com>"> <span class="fn">Yongjun Zhang</span></a>
</span></b>
<pre>(In reply to <a href="show_bug.cgi?id=158239#c5">comment #5</a>)
<span class="quote">> (In reply to <a href="show_bug.cgi?id=158239#c4">comment #4</a>)
> > (In reply to <a href="show_bug.cgi?id=158239#c3">comment #3</a>)
> > > Comment on <span class=""><a href="attachment.cgi?id=280188&action=diff" name="attach_280188" title="Notify the client immediately when the network session doesn't exist for a sync XHR request.">attachment 280188</a> <a href="attachment.cgi?id=280188&action=edit" title="Notify the client immediately when the network session doesn't exist for a sync XHR request.">[details]</a></span>
> > > Notify the client immediately when the network session doesn't exist for a
> > > sync XHR request.
> > >
> > > Why is this only an issue for sync loads? It seems like it's an issue for
> > > async loads, too.
> > >
> > > Sync loads cause a larger issue because the WebProcess is purposefully
> > > hanging for them, but isn't it also bad for the WebProcess to think an async
> > > load is happening.... forever...? Even though it will never happen?
> >
> > I wasn't able to reproduce with async load. We could hit the point where the
> > network session is removed for an async load. However, since this only
> > happens when we closing a web page, Web Process will cancel the load
> > immediately and we won't have a long waiting WebProcess.
>
> That's *currently* the only time a Session can be removed, but I foresee
> that could change in the future.
>
> I don't see any downside to immediately signaling the error for all loads.
>
> You've looked at the code more recently - Do you foresee any downside?</span >
I don't see any downside either, and I agree signaling the error for all loads sounds more future-proof. I will submit another path. Thanks!</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the assignee for the bug.</li>
</ul>
</body>
</html>