Re: Byte ranges & Document title (question)
Gavin Nicol (gtn@ebt.com)
Thu, 18 Jan 1996 11:53:47 -0500
>Is there any hope that there could be at least a particular
>solution, instead of saying "It shouldn't be done at all, because in
>complex documents it becomes infeasible" ?
Nope, it's absolutely impossible to do reliably for HTML in any
SGML-conformant manner.
Go back through the archives, and you'll find that I've already
outlined these problems. Basically, it amounts to this: you have to
keep the previous peices around, and when you download a new peice,
tack it on the end, and reparse the whole document. You also need
very robust error handling for the cases where the document is
incomplete.