|simple syntax analysis firstname.lastname@example.org (Julia Donawald) (2002-05-27)|
|Re: simple syntax analysis email@example.com (Joachim Durchholz) (2002-05-27)|
|Re: simple syntax analysis firstname.lastname@example.org (Andreas Gieriet) (2002-05-31)|
|Re: simple syntax analysis email@example.com (Walter Misar) (2002-06-07)|
|From:||"Joachim Durchholz" <firstname.lastname@example.org>|
|Date:||27 May 2002 21:08:44 -0400|
|Posted-Date:||27 May 2002 21:08:43 EDT|
Julia Donawald wrote:
> I have written the following pseudo-code to built up a parse tree of the
> following sequence: [A] [OR] [B] [AND] [C]. I read in some books about that
> theme and create the following grammar:
> expr := factor | factor OR expr | factor AND expr
> factor:= letter | (expr)
This grammar is ambiguous. Since you have a clear preference which of
the various possible parse trees is the one you want, the grammar
doesn't reflect your intentions: you need priorities.
Since you're hand-crafting the parser, the easy way out (just adding
operator precedence specifications to the grammar) won't work. So a
grammar rewrite is in order:
expr ::= expr1 | expr OR expr1
expr1 ::= expr2 | expr1 AND expr2
expr2 ::= letter | ( expr )
Note that you can control associativity by reversing the operands around
OR and AND, e.g. to get a right-associative OR, write "expr1 OR expr".
To get a nonassociative operator, write "expr1 OR expr1" (not very
useful for OR and AND, but could be helpful for "<" since that prevents
constructions like "a < b < c", which make sense but are difficult to
persuade into doing the right thing during parsing).
Return to the
Search the comp.compilers archives again.