[antlr-interest] Superclasses = best practise ...?

Hans-Juergen Rennau hrennau at yahoo.de
Tue May 24 14:04:09 PDT 2011

Hi Michael, Eric, Jim,

thank you very much for discussing the question of the use of superclasses for 
separating logic from grammar. Indeed, you have convinced me thoroughly that the 
superclass approach is in *most* cases very appropriate - as Jim elaborated, 
only small and simple grammars without the perspective of reuse would make the 
all-in-one alternative really attractive. 

A sidenote to Jim: the archetype does currently not introduce a superclass of 
the tree parser... I think the tree parser is not less of a candidate than the 
parser - so you might consider adding a superclass for it, too. Second sidenote: 
using the archetype, the plugin version 3.3 is proposed, but selecting it 
results in an error - this may confuse.

Kind regards,
-- Hans-Juergen

More information about the antlr-interest mailing list