RESOLVED FIXED 212476
[Apple Pay] Buttons render with a corner radius of PKApplePayButtonDefaultCornerRadius even when explicitly specifying "border-radius: 0px"
https://bugs.webkit.org/show_bug.cgi?id=212476
Summary [Apple Pay] Buttons render with a corner radius of PKApplePayButtonDefaultCor...
Andy Estes
Reported 2020-05-28 10:32:32 PDT
[Apple Pay] Buttons render with a corner radius of PKApplePayButtonDefaultCornerRadius even when explicitly specifying "border-radius: 0px"
Attachments
Patch (28.44 KB, patch)
2020-05-28 10:56 PDT, Andy Estes
no flags
Patch (28.99 KB, patch)
2020-05-28 18:39 PDT, Andy Estes
no flags
Andy Estes
Comment 1 2020-05-28 10:33:21 PDT
Andy Estes
Comment 2 2020-05-28 10:56:12 PDT
Simon Fraser (smfr)
Comment 3 2020-05-28 12:09:16 PDT
How does this interact with "all: unset" etc?
Antti Koivisto
Comment 4 2020-05-28 12:28:11 PDT
Comment on attachment 400481 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=400481&action=review > Source/WebCore/style/StyleBuilderCustom.h:127 > + // FIXME: Teach makeprop.pl to generate setters for hasExplicitlySet* flags Yes please. Doing this manually is bit silly. Also we might want to track the cascade level instead (so property can be set on UA sheet without being considered 'explicitly set').
Andy Estes
Comment 5 2020-05-28 14:15:42 PDT
(In reply to Simon Fraser (smfr) from comment #3) > How does this interact with "all: unset" etc? "all: unset" or "all: initial" would set border-radius back to its initial value, which would set hasExplicitlySetBorderRadius to false, and we would render at the default corner radius. "all: revert" would roll back to the User or UA origin. As Antti pointed out in comment #4, if border-radius is set in one of those origins then we'd consider it "explicitly set" and use it instead of the default corner radius for Apple Pay buttons. In practice, our UA stylesheet does not set border-radius on any elements that authors would style with '-webkit-appearance: -apple-pay-button', so I think this is ok for now. I'll do a few things as follow-up: 1. Update my tests to cover the 'all' shorthand. 2. File a bug for Antti's suggestion to track cascade level (and about the makeprop.pl improvement).
Andy Estes
Comment 6 2020-05-28 18:39:17 PDT
EWS
Comment 7 2020-05-28 21:08:21 PDT
Committed r262279: <https://trac.webkit.org/changeset/262279> All reviewed patches have been landed. Closing bug and clearing flags on attachment 400536 [details].
Note You need to log in before you can comment on or make changes to this bug.