Re: CDATA sections in W3C XML Infoset

["Martin v. Loewis" <martin@loewis.home.cs.tu-berlin.de>]

> As I just noted, normalize() shouldn't be guaranteed to succeed.

And you gave out-of-memory as a potential reason for failure. I hope
it should not be allowed to fail silently, though - if it fails for
some reason, it ought to report the failure.

> Even so, the example above is simplistic... if I choose to, I
> can break the code above any number of ways.

One way you are proposing is to use entity references; I guess another
is to put a comment inside the string I'm searching.

> At the end of the day, almost every application has some number of
> constraints on the actual use of XML. If that is not the case, it's
> better to normalize, then process.

That seems to be the conclusion. However, I then would like to see
more powerful normalization operations in the DOM itself - assuming
that my example falls into a class of common applications.

Regards,
Martin