Re: [xml-dev] Unicode BOM as document separator [was: RE:[xml-dev] "Intr

[John Cowan <cowan@mercury.ccil.org>]

Jim DeLaHunt scripsit:

> I'm not sure how important this is to your usage, but The Unicode
> Standard already defines the meaning of a Byte Order Mark (BOM) code
> point in the midst of data. Up until Unicode 3.2, the BOM code point
> U+FEFF had the Byte Order Mark semantics at the start of a text
> stream, and the Zero-Width Non-Breaking Space (ZWNBS) semantics
> within a text stream. As such, your "<data>" element could validly
> include a U+FEFF codoe point.

That's true, but a U+FEFF cannot appear outside the root element, where
only PIs, comments, and whitespace are valid, never character content.
However, using a control character is easier on the recipient, who can
split the documents before parsing them.

-- 
Being understandable rather than obscurantist poses certain
risks, in that one's opinions are clear and therefore     | John Cowan
falsifiable in the light of new data, but it has the      | cowan@ccil.org
advantage of encouraging feedback from others.  --James A. Matisoff