[antlr-interest] best way for mass tree parsing

[Friedhelm Hellerhoff]

Hi all,
with ANTLR 3 (and C#), I have set up a lexer, parser
and treeparser for parsing some special syntax. 
I have to evaluate some millions of records against 10
different AST with the treeparser. With ANTLR 2.7, I
first run the syntax through the lexer and parser and
then put the 10 resulting AST into a  list of AST.
Then in the loop over the million records, this list
was used for evaluating the millions of records with
just one instance of the treeparser.
What about ANTLR 3, what is the best way to separate
the lexing,parsing and treeparsing? I think creating
new treeparser instances for every record would not be
the sweetest way...?
Thanks in advance for any answer,
  Friedhelm


      __________________________________ Wissenswertes zum Thema PC, Zubehör oder Programme. BE A BETTER INTERNET-GURU!  www.yahoo.de/clever