note blokhead [ikegami]++ .. I recently tried to get a sane parse tree from a grammar which (like your example scenario) was the result of left-recursion elimination on left-associative grammar rules. I found it to be highly nontrivial. <p> I was at least aware of the first two approaches you presented (though it would have taken me a long time to get approach #3a working), but it had not occured to me to use an accumulation paradigm like you do in #3c. This is a standard trick in making recursive functions tail-recursive (pass forward the intermediate results as an argument, instead of passing back the intermediate results as a return value). <p> I probably wouldn't have thought to apply this trick to parsing, though. If you're a theorist like me and view parsing through the lens of formal languages, then passing information from parent rule to child rule is not a natural approach. Good thing there are still practitioners out there ;) <!-- Node text goes above. Div tags should contain sig only --> <div class="pmsig"><div class="pmsig-137386"> <p> blokhead </div></div> 553889 553889