Element.toggleAttribute provides a simple boolean interface the same as classList.toggle for Element. See https://github.com/whatwg/dom/issues/461 Firefox implementation bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1469592 Other browsers are willing to implement: https://github.com/whatwg/dom/issues/461#issuecomment-398206390
Created attachment 344202 [details] Patch
Comment on attachment 344202 [details] Patch Clearing flags on attachment: 344202 Committed r233475: <https://trac.webkit.org/changeset/233475>
All reviewed patches have been landed. Closing bug.
<rdar://problem/41790122>
Comment on attachment 344202 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=344202&action=review > Source/WebCore/dom/Element.h:117 > + ExceptionOr<bool> toggleAttribute(const AtomicString& name, std::optional<bool> force); Should this be localName instead of name? > Source/WebCore/dom/Element.idl:47 > + [CEReactions, MayThrowException] boolean toggleAttribute(DOMString qualifiedName, optional boolean force); Should this be localName instead of qualifiedName?
(In reply to Darin Adler from comment #5) > Comment on attachment 344202 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=344202&action=review > > > Source/WebCore/dom/Element.h:117 > > + ExceptionOr<bool> toggleAttribute(const AtomicString& name, std::optional<bool> force); > > Should this be localName instead of name? > > > Source/WebCore/dom/Element.idl:47 > > + [CEReactions, MayThrowException] boolean toggleAttribute(DOMString qualifiedName, optional boolean force); > > Should this be localName instead of qualifiedName? We try to keep the IDL in sync with the spec and the spec says "qualifiedName". I will check what makes sense for the implementation though.
(In reply to Chris Dumez from comment #6) > (In reply to Darin Adler from comment #5) > > Comment on attachment 344202 [details] > > Patch > > > > View in context: > > https://bugs.webkit.org/attachment.cgi?id=344202&action=review > > > > > Source/WebCore/dom/Element.h:117 > > > + ExceptionOr<bool> toggleAttribute(const AtomicString& name, std::optional<bool> force); > > > > Should this be localName instead of name? > > > > > Source/WebCore/dom/Element.idl:47 > > > + [CEReactions, MayThrowException] boolean toggleAttribute(DOMString qualifiedName, optional boolean force); > > > > Should this be localName instead of qualifiedName? > > We try to keep the IDL in sync with the spec and the spec says > "qualifiedName". > I will check what makes sense for the implementation though. At the very least, the new method is consistent with the others: WEBCORE_EXPORT const AtomicString& getAttribute(const AtomicString& name) const; WEBCORE_EXPORT ExceptionOr<void> setAttribute(const AtomicString& name, const AtomicString& value);
Comment on attachment 344202 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=344202&action=review > Source/WebCore/dom/Element.cpp:1284 > + unsigned index = elementData() ? elementData()->findAttributeIndexByName(caseAdjustedLocalName, false) : ElementData::attributeNotFound; Specification says the input is a qualifiedName so I should probably be using the findAttributeIndexByName(QualifiedName) overload here. I'll look into this further. > Source/WebCore/dom/Element.cpp:1287 > + setAttributeInternal(index, QualifiedName { nullAtom(), caseAdjustedLocalName, nullAtom() }, emptyString(), NotInSynchronizationOfLazyAttribute); Here is correct though because the spec says: "If force is not given or is true, create an attribute whose local name is qualifiedName"
Comment on attachment 344202 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=344202&action=review > LayoutTests/imported/w3c/web-platform-tests/dom/nodes/attributes.html:134 > + el.toggleAttribute("foo:bar"); Actually the behavior for qualified name is tested here and we pass this test. findAttributeIndexByName(const AtomicString&) does the right thing when the attribute has a prefix: attribute.name().toString() == caseAdjustedName Based on this, I believe I should rename localName / name in the implementation to qualifiedName, to match the IDL and the spec.
(In reply to Chris Dumez from comment #9) > Comment on attachment 344202 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=344202&action=review > > > LayoutTests/imported/w3c/web-platform-tests/dom/nodes/attributes.html:134 > > + el.toggleAttribute("foo:bar"); > > Actually the behavior for qualified name is tested here and we pass this > test. findAttributeIndexByName(const AtomicString&) does the right thing > when the attribute has a prefix: > attribute.name().toString() == caseAdjustedName > > Based on this, I believe I should rename localName / name in the > implementation to qualifiedName, to match the IDL and the spec. https://bugs.webkit.org/show_bug.cgi?id=187347