Common event-based parser API

[David Megginson]

Mark L. Fussell writes:

 > A mild question would be: are you planning on being able to modify the
 > Dictionary after you have given it out, or can the client assume it is a
 > constant after the startElement? 

This is undecided right now.  My current test implementation gives you
a new Dictionary, so you can do what you want with it, but the common
interface might impose more restrictions (Tim almost certainly won't
want to allocate a new object each time).

 > > Note the use of the Dictionary abstract base class here.  Hashtable is
 > > derived from Dictionary, as is my Trie class that I released a
 > > couple of years ago.  
 > 
 > Just a heads-up in case anyone doesn't know. In JDK 1.2, Hashtable is
 > obsolete: 
 >    NOTE: This class is obsolete. New implementations should implement
 >    the Map interface, rather than extending this class. 
 >
 > The problem with Hashtable is that it is an abstract class instead of an
 > interface so although you can have different implementations they are
 > still pretty restricted in their implementation approach.  This has been a
 > known problem for a long-long time, and the 1.2 collections are finally
 > interface based.  The basic 'get', 'put' operations for Map are the same
 > though, so it is just a type-ing problem. 

(I have substituted "Dictionary" for "Hashtable" in the above
caution).  

Thanks for the warning -- I have always been annoyed by the fact that
java.util.Dictionary was an abstract base class instead of an
interface, so I am happy to see that they are finally getting around
to changing it.

That makes agreeing on a common event-based interface a little more
difficult, though.  

 >     Not that this should weigh very heavily if you are trying to support
 > 1.0 and 1.1 based browsers, but I suspect the 1.2 release will be migrated
 > to pretty rapidly (in the Spring->Summer). 

Yes, but many users haven't even upgraded to Netscape 3 yet, so it
will be years before we can count on a general user base that will be
able to handle this (including a local copy of Map.class is a clumsy
work-around, and it could sabotage other parts of an applet or
application).


All the best,


David

-- 
David Megginson                 ak117 at freenet.carleton.ca
Microstar Software Ltd.         dmeggins at microstar.com
      http://home.sprynet.com/sprynet/dmeggins/

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)