[antlr-interest] need help writing a source code formatter

[Adrian Sandor]

Hello everybody

I'm a programmer and I'm very determined to write a source 
beautifier/formatter for C++. First I wanted to do it by myself, and 
started writing a grammar, but I had some difficulties when I wanted 
to write the parser (I didn't want to use CYK, I wanted to try LL1 
but I'm afraid C++ is not LL1).
Therefore I looked for help on the web, and I found antlr, which 
seems to be far greater than my expectations. So I gave up the idea 
of writing a parser/translator by myself, and I decided to write an 
antlr grammar instead.
The problem is I really have no idea how to do it.. I have some fair 
knowledge about formal languages but I'm a total newbie to antlr. I 
learnt the hard way about the differences between the lexer and 
parser rules, but I'm still confused.
Here are my questions:
- how do I decide which rules belong to the parser and which rules 
are tokens that belong to the lexer?
- what is the meaning of "warning:nondeterminism upon k==1:SOME_TOKEN 
between alt 1 and exit branch of block" ?
- could somebody have some mercy and write a simple example, like 
parsing a string of correctly matched parentheses with spaces between 
them and stripping all spaces and adding a '+' after each '(' and a '-
' before each ')' ? e.g. "( ()( (  ))  )" -> "(+(+-)(+(+-)-)-)"
- where can I find some simple and helpful documentation for what I 
want to do?

thanks a lot

Adrian


 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/