Cart
[Home] [By Thread] [By Date] [Recent Entries]
At 04:02 PM 7/23/2001 -0400, Elliotte Rusty Harold wrote: >It's trivial. The parser has a boolean flag that says whether or not a >Blueberry character has been seen which is initialized to false when the >parser starts reading a document. If the parser get to the end of the >document, and that flag is still false, it reports a well-formedness >error, in whatever way well-formedness errors are normally reported by the >parser. If you've gone to all the trouble of parsing it, where is there then some sort of savings in then throwing it all away and reporting an error? You've already done the majority of the work. I'm open to being pointed toward examples, but this strikes me as a very odd way of dealing with versioning, simply to avoid letting a member of a sub-set use the superset designation. Your argument is beginning to sound like you just don't want any progression on XML, and therefore want to make it difficult to use any new version -- please correct me if that impression is wrong. Ann Ann Navarro, WebGeek Inc. http://www.webgeek.com/ Now in print! Effective Web Design, 2nd Edition http://www.webgeek.com/books/ What's on my mind? http://www.snorf.net/blog/
|
