← Back to context Comment by nly 10 years ago Interesting paper on left-recursion in recursive descent parsers like PEGs and Packratshttp://web.cs.ucla.edu/~todd/research/pub.php?id=pepm08 3 comments nly Reply ltratt 10 years ago If you're interested in left-recursion in PEGS then, at the risk of gross immodesty, you may be interested in http://tratt.net/laurie/research/pubs/html/tratt__direct_lef...With less risk of immodesty you may also find http://arxiv.org/pdf/1207.0443.pdf?ref=driverlayer.com/web interesting.There's probably more recent work than these two papers, but I'm a little out of date when it comes to the PEG world. nly 10 years ago Thanks, looks like you've put a lot of work in to it and I'll enjoy reading it. sklogic 10 years ago Yes, I implemented exactly this approach. But for the binary expressions I am using Pratt, it is faster and easier to construct left-associative nodes.
ltratt 10 years ago If you're interested in left-recursion in PEGS then, at the risk of gross immodesty, you may be interested in http://tratt.net/laurie/research/pubs/html/tratt__direct_lef...With less risk of immodesty you may also find http://arxiv.org/pdf/1207.0443.pdf?ref=driverlayer.com/web interesting.There's probably more recent work than these two papers, but I'm a little out of date when it comes to the PEG world. nly 10 years ago Thanks, looks like you've put a lot of work in to it and I'll enjoy reading it.
sklogic 10 years ago Yes, I implemented exactly this approach. But for the binary expressions I am using Pratt, it is faster and easier to construct left-associative nodes.
If you're interested in left-recursion in PEGS then, at the risk of gross immodesty, you may be interested in http://tratt.net/laurie/research/pubs/html/tratt__direct_lef...
With less risk of immodesty you may also find http://arxiv.org/pdf/1207.0443.pdf?ref=driverlayer.com/web interesting.
There's probably more recent work than these two papers, but I'm a little out of date when it comes to the PEG world.
Thanks, looks like you've put a lot of work in to it and I'll enjoy reading it.
Yes, I implemented exactly this approach. But for the binary expressions I am using Pratt, it is faster and easier to construct left-associative nodes.