[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On 09/06/11 09:16, tom wrote: > > I have got a work around to do what I wanted to (elem.type="text" IIRC) > but proving IE8 is not dom compliant would be fun but I'm having trouble > locating what might be the standard. http://www.w3.org/TR/2000/WD-DOM-Level-1-20000929/level-one-html.html#ID-6043025 DOM-Level-1 defines this as a read-only element. Microsoft note it can be written once in IE5 before it is inserted into the DOM, which is consistent with the DOM-Level-1 spec. In DOM-level-2 spec the attribute "type" of HTMLinputElement is no longer specified as read only. http://www.w3.org/TR/DOM-Level-2-HTML/html.html#ID-6043025 The same spec also notes that the preferred way of setting attributes is via a generic mechanism (I read that as preferring "element.attribute["type"]="text", the reasoning is that not all attributes of XML could be referred to as "element.attribute.name" in valid ECMAscript. So yes IE8 is almost certainly falling short of the DOM-Level-2-HTML specification in this matter. On the other hand as a programmer I write 'e.type="text"' rather than 'e.attribute["type"]="text"' because it is 13 characters shorter, maybe I should get VIM to change them as I go (except for "type" on InputElements - because it'll break in IE8). Personally I marvel of the fact that diverse programs can implement specs this large, for such complex behaviour and not have more security and more compatibility issues than they already do. Sometimes I think it would be wise to ignore it all, and just start again - maybe some evil trick of writing you new browser in Java, and stepping away from the whole "web" thing, and doing whatever one likes. -- The Mailing List for the Devon & Cornwall LUG http://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/listfaq