Cart
[Home] [By Thread] [By Date] [Recent Entries]
> I suppose you may think that this should be > in the realm of schema, probbaly yes, because a schema validator has many more options on what to do on failure. XPath's don't ever really fail, they just don't always select any nodes, so if the author is expecting there is only one head and goes /html/!head/!title and then in the document there are two <head>. One might hope to get an error condition, but (presumably) in XPath there would be no error, but the above XPath would just fail to select any nodes, just as if there were not a <head> element at all. I'm not sure if this would be the expected behaviour. This is assuming the semantics of !head are (baring optimisation concerns) the same as head[position()=1 and last()=1] David
|
