Elliotte pointed out a couple of problems in our PR. I have fixed them both. For the second, I added another sentence to the defn of acquired infoset in 4.2: "[Definition: xi:include elements in this infoset are recursively processed to create the acquired infoset. For an intra-document reference (via xpointer attribute) the source infoset is used as the acquired infoset.]" This bakes subresource identification into the process for determining which items are included, rather than relying on section 4.2.5 and the implications there to enforce this constraint. With this definition, the text in 4.2.5 can be softened a little: "Intra-document references within xi:include elements must be resolved against the source infoset. The effect of this is that the order in which xi:include elements are processed does not affect the result." I've replaced "must be" with "are", since the gist is now baked into the definition and the "must" doesn't add anything. Happy to back these out of they're controversial, but they seemed editorial in nature to me.
In section 4.1 "the acquited infoset" should be "the acquired infoset" -- Elliotte Rusty Harold
OK. The sentence in 4.5 "Intra-document references within xi:include elements must be resolved against the source infoset." means there isn't really a problem with same document references as I feared. However, that they are not treated the same as interdocument references is a little confusing. I suggest adding something to section 3.1 about this; e.g. "XPointers that point to the current docuemnt refer to the source infoset; all others point to the acquired infoset." Hmm, what happens if a same document reference is made with a URL in an href attribute? And does this mean a multi-URL'd document could have different results depending on which URL it was accessed from? Anyway, the sentence after this is, "The effect of this is that the order in which xi:include elements are processed does not affect the result." This isn't actually true. Obviously as my test case showed the order in which xi:include elements are processed does affect the results. This sentecne should be rewritten so that it's clear this only applies within one document; e.g. "The effect of this is that the order in which xi:include elements in the same document are processed does not affect the result." -- Elliotte Rusty Harold elharo@metalab.unc.edu
Received on Friday, 10 September 2004 20:45:15 UTC