Janiczek/elm-grammar

(WIP) Parse your strings with BNF grammars in Elm

Janiczek/elm-grammar

Context free grammar parsers

TODO

• the whole documentation
• mention we can't deal with left recursion
  • mention a failed experiment with instaparse's GLL in the gll branch

breakfast -> protein ( <" with "> breakfast <" on the side"> )?
breakfast -> bread

protein   -> ("really" <" ">)+ "crispy" <" "> "bacon"
protein   -> "sausage"
protein   -> cooked <" "> "eggs" 

cooked    -> "scrambled" | "poached" | "fried"

bread     -> "toast" | "biscuits" | "English muffin" 

• TODO the arithmetic expression example?

• TODO the Lox language example?

• TODO: when running

<foo> -> "bar"
foo -> "baz"

on "bar", don't show the "foo" tag, but when running it on "baz", show the "foo" tag.

• TODO: mention that it's enough to only hide one tag to hide all its occurences

• TODO: mention that the starting tag cannot be hidden

• TODO: mention the ::= synonym for ->

• TODO: mention the = synonym for ->

• TODO: mention the ; optional ending character

• TODO: mention that ; also serves as a line comment

• TODO: mention -- line comments

• TODO: mention / as alternative syntax for |

• TODO: mention , as whitespace (ignored character)

Implemented:

- Literal:       s  -> "a"
- Regex:         s  -> /\d+/
- Tag:           s  -> a
- Concatenation: s  -> a b
- Alternation:   s  -> a | b
- Grouping:      s  -> (a)
- Hiding:        s  -> <a>
- Optional:      s  -> a?
- Zero or more:  s  -> a*
- One or more:   s  -> a+
- Lookahead:     s  -> &a
- Comment:       s  -> (* hello *) "world"
- Hiding tags:  <s> -> "abc"
- EOF:           s  -> EOF