Last active
September 1, 2018 17:23
-
-
Save aengelberg/6faf3bafae52145990ba735874e400ac to your computer and use it in GitHub Desktop.
*slaps roof of parser*
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
(ns alex.csbnf | |
"Framework for writing parsers using CSBNF (Car Salesman Backus-Naur Form) notation." | |
(:require | |
[instaparse.combinators :as c] | |
[instaparse.core :as insta])) | |
(def csbnf-parser | |
(insta/parser | |
"grammar = rule+ | |
<rule> = (<'*slaps roof of'> rule-name <'*' 'this bad boy'> | |
| <'this'> rule-name) | |
<'can fit so'> | |
<'many' | 'much'> | |
<'fucking'?> <'goddamned'?> | |
alt | |
<'in it'> | |
rule-name = #'[a-zA-Z0-9_-]*[a-rt-zA-Z0-9_-]' | |
alt = cat (<'or'> cat)* | |
cat = token (<'followed by'> token)* | |
<token> = string | rhs-rule | <'('> alt <')'> | |
string = ( <'\"'> #'([^\"]|\\\\\")*' <'\"'> | |
| <\"'\"> #\"([^']|\\\\')*\" <\"'\">) | |
plural | |
rhs-rule = rule-name plural | |
<plural> = <'s' | \"'s\">?" | |
:auto-whitespace :standard | |
:string-ci true)) | |
(defn csbnf | |
[grammar] | |
(->> grammar | |
(csbnf-parser) | |
(insta/transform | |
{:grammar vector | |
:cat c/cat | |
:alt c/alt | |
:string c/string | |
:rule-name keyword | |
:rhs-rule c/nt}) | |
(insta/parser) | |
(comp not insta/failure?))) | |
(def balanced-paren-parser | |
(csbnf | |
"*slaps roof of sexpr* this bad boy can fit so many | |
left-parens followed by content followed by right-parens in it | |
this left-paren can fit so many '('s in it | |
this right-paren can fit so many fucking ')'s in it | |
this content can fit so much goddamned '' or sexprs followed by content in it")) | |
(balanced-paren-parser "(()((()))())") | |
;; => true | |
(balanced-paren-parser "(((") | |
;; => false |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment