Here's how you can use Janet's built-in Parsing Expression Grammar (PEG) module to define a grammar capable of parsing valid JSON into an Abstract Syntax Tree in only 13 lines of code (one of which is just the "define" keyword and a symbol name):
(def json-parser
~{:null (/ (<- "null") ,|[$ :null])
:bool (/ (<- (+ "true" "false")) ,|[$ :bool])
:number (/ (<- (* (? "-") :d+ (? (* "." :d+)))) ,|[$ :number])
:string (/ (* "\"" (<- (to (* (> -1 (not "\\")) "\"")))
(* (> -1 (not "\\")) "\"")) ,|[$ :string])
:array (/ (* "[" :value (any (* :s* "," :value)) "]") ,|[$& :array])
:object (/ (* "{" :s* :string :s* ":" :value
(any (* :s* "," :s* :string :s* ":" :value))
"}") ,|[(from-pairs (partition 2 $&)) :object])
:value (* :s* (+ :null :bool :number :string :array :object) :s*)
:unmatched (/ (<- (some 1)) ,|[$ :unmatched])
:main (some (+ :value "\n" :unmatched))})
[WIP]
You would then use this grammar in an actual program like this:
(peg/match json-parser "{\"key\": \"value\"}")
=> @[(@{("key" :string) ("value" :string)} :object)]
(NB: You'd probably also want to call (peg/compile) on the grammar object before using it a lot, because that will speed up the (peg/match) function considerably.)
In what follows, I'll re-create this parsing grammar step-by-step from scratch as a follow-along introduction to Janet's Parsing Expression Grammars. The best way to come along on this journey with me is to download and install Janet, then create a project, pop open your favorite Text Editor and a Janet REPL, and try everything yourself as you read along!