Test case: http://dvcs.w3.org/hg/editing/raw-file/6aaa4b8455c9/selecttest/unload.html Mozilla bug: https://bugzilla.mozilla.org/show_bug.cgi?id=717339 The editing spec was just changed to require this. Neither Mozilla nor WebKit do this currently, but IE/Opera do and I think it makes sense.
Safari 5.1 says Pass. When did this regress?
After discussion with Boris Zbarsky and Ian Hickson: actually, the Selection object is supposed to be replaced with a new one. This is specified in HTML as part of the document.open() algorithm: """ Replace the Document's singleton objects with new instances of those objects. (This includes in particular the Window, Location, History, ApplicationCache, and Navigator, objects, the various BarProp objects, the two Storage objects, the various HTMLCollection objects, and objects defined by other specifications, like Selection and the document's UndoManager. It also includes all the Web IDL prototypes in the JavaScript binding, including the Document object's prototype.) "" http://www.whatwg.org/specs/web-apps/current-work/multipage/elements.html#dom-document-open New test: http://dvcs.w3.org/hg/editing/raw-file/56c83bbaa44c/selecttest/Document-open.html The new test shows that Gecko does replace the selection with a new object, but that object has the same ranges as the old one. WebKit (Chrome 17 dev) is the same object as before. Opera clears the ranges but I can't tell if it's the same object, because getSelection() != getSelection() in Opera. IE returns a different object and clears the ranges, so it's the only one that passes the new test. This is specific to document.open(), not a general unloading thing.
Regressed in <http://trac.webkit.org/changeset/96257>.