Richard Tobin wrote: >> I see (responding to your point about it not being related to ancestor >> elements). But I was trying to find somewhere where the issue was made >> explicit as to how xml:base is not only based on the parent, since there >> are these following lines (and basically only these lines on the topic) >> which only refer to "parent element" and not clearly to an indefinite >> number of levels of inheritance of within the document (and regardless >> of whether xml:base was explicitly present on the parent or not, etc.): >> >> "The base URI of an element is...the base URI of the element's parent >> element within the document or external entity, if one exists..." >> >> "The base URI for a URI reference appearing in an |xml:base| attribute >> is the base URI of the parent element of the element bearing the >> |xml:base| attribute, if one exists within the document entity or >> external entity, otherwise the base URI of the document entity or >> external entity containing the element." >> >> Maybe that is made clear in other referenced specs, but since this seems >> to be such a central point of the spec, I would think clarifying that >> would be helpful. >> > > The answer follows from the rules you quote above. Given > > <a xml:base="http://example.com"> > <b> > <c> ... > > the base URI of the <c> element is (by 4.2 part 2) the base URI > of the <b> element. Similarly, the base URI of the <b> element > is the base URI of the <a> element. And by 4.2 part 1, the base > URI of the <a> element is "http://example.com". > Yes, I am aware that this is the intention, but by specifying "parent" (without reference to recursion), it raises the question of whether it is constrained to this and whether a base URI must be explicit except for an immediate child. With the phrase "if one exists" in these rules (while I know its intention), I think this also raises the question about whether this might refer to whether an explicit definition exists. I know that the answer can be derived, but I think some extra note or qualification could help make it more explicit. Brett BrettReceived on Monday, 3 November 2008 23:19:32 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 08:39:46 GMT