You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I don't think browsers implement this (Firefox definitely doesn't). I think this is both very hard to test, and inconsistent with other mutations. That'd cause things like img.srcset = img.srcset trigger a load event or not depending on whether it was partially available.
That condition would make more sense, IMO, if it also included fully available images. Firefox does have a code-path for some mutations where we don't reload the source if the effective source hasn't changed (e.g. when re-evaluating sources as a consequence of a window resize). Is that the intention?
In any case, even with that tweak that doesn't quite match current browser behavior (though I think it'd be nice if we could make that change). E.g., changing a sizes attribute, or adding a <source> that doesn't change the URI currently fires a load event consistently, which is rather unfortunate and I think we should try to avoid if we can...
What is the issue with the HTML Standard?
https://html.spec.whatwg.org/multipage/images.html#updating-the-image-data has:
I don't think browsers implement this (Firefox definitely doesn't). I think this is both very hard to test, and inconsistent with other mutations. That'd cause things like
img.srcset = img.srcset
trigger a load event or not depending on whether it was partially available.That condition would make more sense, IMO, if it also included fully available images. Firefox does have a code-path for some mutations where we don't reload the source if the effective source hasn't changed (e.g. when re-evaluating sources as a consequence of a window resize). Is that the intention?
In any case, even with that tweak that doesn't quite match current browser behavior (though I think it'd be nice if we could make that change). E.g., changing a
sizes
attribute, or adding a<source>
that doesn't change the URI currently fires aload
event consistently, which is rather unfortunate and I think we should try to avoid if we can...https://bug1934222.bmoattachments.org/attachment.cgi?id=9440726 has a test-case.
cc @zcorpan @tnikkel @annevk
The text was updated successfully, but these errors were encountered: