> From: "Glenn Marcy" <[EMAIL PROTECTED]>
>
> Issue 1:
> 
>   The JavaDoc for DTDHandler has an @link to
> org.xml.sax.ext.LexicalHandler.
>   This would seem to imply that there would be a broken link if there
>   were a core-only SAX2 package with docs.  

Javadoc warnings, no more.  Just the same as when
building javadoc for a core-only package without the
helpers.   (Yep, go back to the SAX1 overview and
notice it describes them as "optional" ... :)

This seems to me like a straightforward "design for the
typical case" tradeoff ... and tweaking @see/@link is
easy for anyone else.


> Issue 2b:

Resolved along with issue 2a, in the r2pre3 release.


> Issue 3b:
> 
> The comment about "standard processing of both byte and character streams"
> is nothing of the sort.  Assigning some sort of legitimacy to such behavior
> is ridiculous at this point.

No it isn't.  I just ran the attached program, which reports that in fact
it *IS* the standard processing for every SAX2 parser I have handy.
Including AElfred2 (many versions), Crimson (1.1.3), Xerces (1.4.3,
despite your comment about not closing :) and XP (0.5 still :) all do
what's described there as "standard processing".

Data stream handling was originally underspecified, so portable
applications couldn't rely on being able to reuse streams after giving
them to a parser.  (The InputSource isn't modified by closing one
of the streams it wraps.)  After that spec update ... they still can't!

So it's not as if anything changed.  Closing still isn't a solid requirement,
since that change was purely clarification.  Parsers that for some strange
reason don't close() won't immediately become nonconformant.

- Dave


Attachment: close.java
Description: Binary data

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to