SSIA
<rdar://problem/64312450>
Created attachment 401792 [details] Patch
Comment on attachment 401792 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=401792&action=review > Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm:1932 > + return !WebKit::currentUserInterfaceIdiomIsPad(); This changes the behavior for read-only <input type=select> and <input type=color> on iPhone. They will now return true. Is this what we want?
Comment on attachment 401792 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=401792&action=review >> Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm:1932 >> + return !WebKit::currentUserInterfaceIdiomIsPad(); > > This changes the behavior for read-only <input type=select> and <input type=color> on iPhone. They will now return true. Is this what we want? (I assume you mean <select> and <input type=color>) This patch will make us match shipping behavior, but it does seem strange that we would allow the keyboard to be presented on iPhone for readonly color inputs. (Note that it doesn't change behavior for select elements, since the readonly attribute does not apply to select elements). We probably do want to avoid showing any input view for readonly color inputs on iPhone though, so while I'm here I'll preserve that behavior. But this does highlight another issue, which is that on iPad, the `readonly` attribute doesn't prevent us from mutating form controls that show a popover or context menu (i.e. date, time, color). We probably should address this in a separate followup.
Created attachment 401807 [details] v2
Thanks for the review! (I filed https://bugs.webkit.org/show_bug.cgi?id=213159 to track the followup I mentioned above)
Committed r262993: <https://trac.webkit.org/changeset/262993> All reviewed patches have been landed. Closing bug and clearing flags on attachment 401807 [details].
*** Bug 213835 has been marked as a duplicate of this bug. ***