[webkit-reviews] review denied: [Bug 97426] Unexpected reftest passes are only reported when pixel testing is enabled in results.html as well : [Attachment 165313] Patch

[bugzilla-daemon at webkit.org]

Ojan Vafai <ojan at chromium.org> has denied Zan Dobersek
<zandobersek at gmail.com>'s request for review:
Bug 97426: Unexpected reftest passes are only reported when pixel testing is
enabled in results.html as well
https://bugs.webkit.org/show_bug.cgi?id=97426

Attachment 165313: Patch
https://bugs.webkit.org/attachment.cgi?id=165313&action=review

------- Additional Comments from Ojan Vafai <ojan at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=165313&action=review


The python change here makes it so that *every* test is treated as a reftest.
Not sure off the top of my head what the correct code is, but we should only
mark reftests as reftests.

I'd prefer if we also keep is_reftest only for matching reftests. It just
complicates the if-statement on the JS side to need to check both is_reftest
and is_mismatch_reftest.

Also, can you please write a test for the results.html change?

> LayoutTests/fast/harness/results.html:515
> +	   if (expected != 'IMAGE' || globalState().results.pixel_tests_enabled
|| testObject.is_reftest) {

This should be (expected != 'IMAGE' ||
(globalState().results.pixel_tests_enabled || testObject.is_reftest)), no?