Hello Mark,
I have few suggestions:
1. I was going through the tutorial and comparing EBNF and ABNF grammars.
ABNF adds Bounded Repetition 3*5 A. Is there any chance of adding it also
to EBNF? I don't know, how the syntax would be, but it can be useful at
times.
Repetition of characters can be done inside regular expression. Repetition
of rules can be done also - 3*5 A would be:
S = A A A A? A?
Repetition of 100*120 A is not practical to encode like this anymore. I
know about the combinators:
{:S (rep 100 120 (nt :A))}
and mixing of grammars:
(abnf "S = 100*130 A")
but still, it would be nice to be able to do it inside the EBNF string.
Reading through the:
http://en.wikipedia.org/wiki/Ebnf
2. I found that EBNF implements an exception operator -
http://en.wikipedia.org/wiki/Ebnf#Table_of_symbols
used for example like this:
string = '"' , { all characters - '"' }, '"' ;
Negative lookahead can be used instead
string = '"' , { !'"' all characters }, '"' ;
Just wanted to let you know about this difference. I can imagine that there
many modifications of EBNF grammar out there :-)
3. Returning back to the point 1. and reading
http://en.wikipedia.org/wiki/Ebnf#Conventions
There actually is a way, how to express repetitions in EBNF:
aa = "A";
bb = 3 * aa, "B";
cc = 3 * [aa], "C";
dd = {aa}, "D";
ee = aa, {aa}, "E";
ff = 3 * aa, 3 * [aa], "F";
gg = {3 * aa}, "G";
Again, don't know if the syntax is optimal. ABNF and regular expressions
define it better, in my opinion.
Thank you for your good work!
Frantisek
On Tuesday, May 14, 2013 10:13:52 AM UTC+2, puzzler wrote:
>
> Instaparse is an easy-to-use, feature-rich parser generator for Clojure.
> The big idea behind instaparse is to make it simple to convert grammars to
> parsers without needing to know the idiosyncrasies of LL1, LALR, and other
> esoteric grammar restrictions imposed by most parser generators.
>
> When I announced instaparse a little over a month ago, I imagined that I
> had packed so many features into it, that surely there was nothing more
> that anyone could possibly want. Oh, how naive I was :) . Within a few
> days, there were a half-dozen great enhancement ideas posted on the github
> site.
>
> Here are the highlights of the new 1.1.0 release:
>
> 1. Support for comments in the grammar. (This was by far the most popular
> request.)
>
> 2. A new front-end for ABNF grammars. ABNF is a syntax popular for
> carefully defining protocols in formal specs. Instaparse's support for
> ABNF means that it is a simple copy-paste exercise to turn these
> specifications into an executable parser.
>
> 3. The ability to convert EBNF and ABNF *fragments* into Clojure data
> structures that can be easily merged with one another and with data
> structures built by the combinator library.
>
> https://github.com/Engelberg/instaparse for full feature list and
> extensive tutorial.
> https://github.com/Engelberg/instaparse/blob/master/CHANGES.md for a list
> of changes since the last version.
> https://github.com/Engelberg/instaparse/blob/master/docs/ABNF.md for
> detailed docs about the new ABNF syntax.
>
> It seems that whenever instaparse comes up, there is an outcry from people
> who have a visceral reaction against the notion of building a parser from a
> *string specification* of a grammar. So I want to be clear up front that
> instaparse supports *three *input formats that you can freely
> mix-and-match: EBNF strings, ABNF strings, and Clojure data structures.
> String input is particularly handy since most grammars are already written
> down somewhere in one of these two notations, but the data structures are
> an option if you need them. Instaparse also supports two output formats:
> hiccup and enlive.
>
> This release focused mostly on adding new features; the next version will
> primarily be another round of performance tuning. If you want to follow
> along and help test as I try out new optimization strategies, keep an eye
> on the 1.2.0-SNAPSHOT branch:
> https://github.com/Engelberg/instaparse/tree/v1.2
>
> Special thanks to Alex Engelberg who implemented the new ABNF input
> format, and David Powell and Peter Monks who suggested the feature and
> helped test it.
>
--
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
---
You received this message because you are subscribed to the Google Groups
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to clojure+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
