Thanks for sharing the library with the community. It looks like a lot of work has been put into it, and the documentation is excellent.
My only quibble, is that I wouldn't call the approach unique. While I am unaware of what is available in Java there are many similar libraries in other languages. For example I used a similar technique in C# in the Heron language (http://code.google.com/p/heron-language/source/browse/trunk/HeronEngine/HeronGrammar.cs). The only difference is that I used operator overloading for sequence and priortized choice operators. - Christopher Diggins On Thu, Nov 12, 2009 at 11:10 AM, Mathias <math...@parboiled.org> wrote: > Hello everyone, > > I thought the Java guys among you might be interested in a new, open > source Java 1.5+ library supporting PEG parsing in a pretty unique way. > http://www.parboiled.org was born out of frustration over existing > parser generators with their external grammar files that lack IDE > support and are often just a little bit too hard to integrate in real- > world projects. > > parboiled let's you specify your PEG grammar rules right in Java source. > > Small example: > Consider the following grammar (a classical calculator example): > > Expression ← Term ((‘+’ / ‘-’) Term)* > Term ← Factor ((‘*’ / ‘/’) Factor)* > Factor ← Number / ‘(’ Expression ‘)’ > Number ← [0-9]+ > > > A parboiled parser definition, complete and in ready-to-compile Java > code would look like this: > > > public class CalculatorParser extends BaseParser<Object, > Actions<Object>> { > > public Rule expression() { > return sequence( > term(), > zeroOrMore( > sequence( > firstOf('+', '-'), > term() > ) > ) > ); > } > > public Rule term() { > return sequence( > factor(), > zeroOrMore( > sequence( > firstOf('*', '/'), > factor() > ) > ) > ); > } > > public Rule factor() { > return firstOf( > number(), > sequence('(', expression(), ')') > ); > } > > public Rule number() { > return oneOrMore(charRange('0', '9')); > } > } > > Your rule definitions can also include direct calls to action methods > which will behave as expected, they are called at the right time > during the parsing run. > parboiled supports AST construction in a very flexible way and has a > few other, quite neat features (like proper parse error recovery, > white space gobbling and so on). It currently implements a recursive > descent parser, but could be extended to support packrat parsing as > well without too much trouble. > > The source and a rather full-blown documentation currently lives at > github, which you are forwarded to from http://parboiled.org. > > Maybe you like the approach and are willing to provide feedback... I'd > be happy to receive any. > > Cheers, > Mathias > > --- > math...@parboiled.org > http://www.parboiled.org > > > _______________________________________________ > PEG mailing list > PEG@lists.csail.mit.edu > https://lists.csail.mit.edu/mailman/listinfo/peg > _______________________________________________ PEG mailing list PEG@lists.csail.mit.edu https://lists.csail.mit.edu/mailman/listinfo/peg
