<html>
<head>
<base href="https://bugs.webkit.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - No reliable way to get a snapshot of WKWebView (macOS)"
href="https://bugs.webkit.org/show_bug.cgi?id=161450#c26">Comment # 26</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - No reliable way to get a snapshot of WKWebView (macOS)"
href="https://bugs.webkit.org/show_bug.cgi?id=161450">bug 161450</a>
from <span class="vcard"><a class="email" href="mailto:dasau@microsoft.com" title="Dan Saunders <dasau@microsoft.com>"> <span class="fn">Dan Saunders</span></a>
</span></b>
<pre>(In reply to <a href="show_bug.cgi?id=161450#c14">comment #14</a>)
<span class="quote">> I think we need significantly more test coverage. The single test is better
> than nothing, but I’d like to see tests covering different types of content.</span >
I can look into adding additional tests. The problem with any test that is testing WebKit pixel contents is it is likely to unreliable when the rendering changes slightly. I see TestInvocation::dumpPixelsAndCompareWithExpected is already used in the test, perhaps this has been unreliable in the past. I don't want to make maintaining the test to be too much overhead. For Mac we are just calling into WebPage::takeSnapshot which is used to take a snapshot in other places such as thumbnails. The thumbnail tests do not compare pixel content and are more basic to be reliable. My unit tests are testing the code I added is functioning, but not testing all the edge cases of WebPage::takeSnapshot specifically. I am testing that I can retrieve a pixel that should deterministically be a certain color. WebPage::takeSnapshot should be reliable because it is used for core functionality of WebKit.</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>