[
Lists Home |
Date Index |
Thread Index
]
In article <425AE296.6010408@metalab.unc.edu> you write:
>By that reasoning, no true infosets have any unexpanded entity reference
>information items.
That's true for valid documents, and for invalid documents that don't
have any undefined entities.
>It seems that a document can have multiple infosets depending on how one
>processes it (with or without entity expansion).
If you're going to talk about a document, it makes sense to say that
there is one infoset that is *the* infoset of that document. There
are various incomplete infosets that processors may produce if they
don't read the whole document. This seems much clearer than saying
that the document itself has many infosets.
-- Richard
|
