- From: Marcos Caceres <marcosc@opera.com>
- Date: Thu, 20 Aug 2009 14:26:45 +0200
- To: Scott Wilson <scott.bradley.wilson@gmail.com>
- Cc: public-webapps WG <public-webapps@w3.org>
On Thu, Aug 20, 2009 at 1:59 PM, Scott Wilson<scott.bradley.wilson@gmail.com> wrote: > The question is important as it affects conformance. Currently we only > support the first of the three methods you've stated, and there was > discussion on the list previously as to whether the second of the three was > even possible in a pure JavaScript implementation (search for "syntactic > sugar"). > > What is the actual conformance requirement? The way the Storage interface is to be implemented is defined by WebIDL and in ECMAScript (SomeECMAScript-Object implements Storage). The conformance requirement is to implement the Storage interface as defined in Web Storage (through the rules/behaviors of WebIDL and ECMAScript, when it is JavaScript; a Java Implementation would behave differently, I imagine), but with the additional constraints that some key/values are protected as defined by A&E. Again, preference is just an Object, as defined by ECMAScript. All javascript object's attributes can be accessed by "[]" or using ".whatever". If the Storage interface itself is screwy, then this needs to be taken up with the editor of the Web Storage spec, as it is not something we have control over. To me, personally, the behavior seems fine and has been implemented that way in at least two browsers (Safari and Firefox). ...seems a recent change to Web Storage may be even more serious. Someone just told me that the Web Storage spec has changed to allow storage of things other than strings. Need to confirm that. -- Marcos Caceres http://datadriven.com.au
Received on Thursday, 20 August 2009 12:27:44 UTC