DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=26980>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=26980

XMLStringPool is not useful as a base class





------- Additional Comments From [EMAIL PROTECTED]  2004-02-17 17:20 -------
I wasn't implying that source code compatibility is not binary, I was simply 
saying you could minimize the impact by creating a new abstract base class, and 
using that solely in the interface of XMLGrammarPool, and in the interfaces of 
the internal classes.  So, I wasn't pushing for a "getXMLStringPoolBase()" call 
on XMLGrammarPool, I simply was suggesting you could change all the references 
to XMLStringPool in XMLGrammarPool to XMLStringPoolBase.  That way, anyone 
using XMLStringPool would still be using a concrete class, not an abstract one.

Although I appreciate your desire to maintain source code compatibility, 
sometimes it _should_ be sacrificed to clean up design problems, because the 
advantage to the improvement of the design outweighs the small pain suffered 
but those who will be affected.  And I suspect that is a very small number of 
people.

XMLStringPool is not useful for my purposes because it requires IDs be assigned 
in sequential order, and that's a very serious limitation.  But making the data 
members protected might be useful for future users who can live with that 
restriction.

But, for the future, you should consider that making abstract base classes 
depend on concrete classes in this way can really limit how

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

Reply via email to