fail - measured over 24137 pixels carrying ink in either image.
These results are not meaningful. This check renders a single static frame, and nothing in a test document records which moment in time its reference image was captured at - so a correct SMIL implementation and a broken one are equally likely to mismatch. Animation behaviour is covered by dedicated unit tests instead; see ctgnz/foxglove#208 for a real measurement methodology, still to be worked out.
W3C reference (context only, see #200)
Foxglove
Difference
In the difference panel, red marks pixels that
differ beyond tolerance and grey marks matching content. The tinted band, where present, is the
<text id="revision"> legend every test carries: it renders in a fallback font this
library cannot match, so it is excluded from the comparison entirely rather than failing every test.
What this test checks
This test contains four cases. The cases must produce the following results for the test to pass. Case 1: On a mouseover event on the top square, all four squares must turn blue. Case 2: On a mouseover event on the top middle square, all four squares must turn blue and a black stroke must appear on the referencing square (element). Case 3: On a mouseover event on the bottom middle square, all four squares must turn blue and a black stroke must appear on the referencing square (element). Case 4: On a mouseover event on the bottom square, all four squares must turn blue, and on a mousedown event a black stroke must appear on the referencing square (element).
