[antlr-interest] your Or ast question

[Randall R Schulz]

On Thursday 02 October 2008 17:54, you wrote:
> Mr. Schulz :-

I'm sorry. My father isn't here right now...


> Sorry for this disconnected reply to your last message to me.

It's OK. I hope you don't mind my sending the reply back to the list.

I'm surprised this seems so difficult (not yet conceding 
impossibility!), so I think it would be generally interesting to have 
the whole list membership observe the attempts.


> ...
>
> recapping from memory, you want
>
> p | q | r | s
>
> to be represented, essentially, as an ast of
>
> (| p q r s)

Precisely.


> ...
>
> maybe try this
>
> o : (a->a) ( ('|' a)+ -> ^('|' a+) )? ;

The result is the same:

	(Or p (Or q (Or r (Or s))))


> and also sorry, because i have forgotten your op7xxFormula rule
> naming.

Well, it's idiosyncratic, to be sure, but I just chose it based on the 
language description I was using to write the grammar. There is nothing 
even approximating a BNF for this language, so there's a lot of 
interpretive leeway. That would be fine, but the software I'm trying to 
emulate has a long history and thus I have to replicate it's grammar 
pretty closely, lest I fail to successfully and properly transliterate 
it.


> too many Sorries; but hope this helps...

Really, there's not need to apologize.


>    -jbb

Thanks!

Randall Schulz