Core language

FORMKL is a markup language used to annotate form model, sections and fields. It simply contains a set of grammar rules and a parser.

Grammar

The grammar is defined in EBNF.

An simple example of EBNF:

bash

# BNF is a set of rules that express the possible
# valid cases of a set of keywords (tokens)

Statement
  = Indefinite Subject # a girl, a boy, the spider man
  | Adjective Subject # beautiful girl, awesome boy
  | Indefinite Adjective Subject # a beautiful girl, the amazing spider man
  ;

Subject
  = "girl"
  | "boy"
  | "spider man"
  ;

Adjective
  = "beautiful"
  | "amazing"
  | "awesome"
  ;

Indefinite
  = "a"
  | "an"
  | "the"
  ;

Another simple example for programming language:

bash


# Accept a full function declaration
#  function foo () {}
#  function foo (a, b) {}
#  function foo (a, b) {
#    var c = a + b;
#    return c;
# }
FunctionDeclaration
  = "function" Identifier "(" ParamList ")" BlockStatement
  ;

# Accept a list of parameters recursively or no parameter (null)
ParamList
  = null
  | Param
  | ParamList "," Param
  ;

Param
  = "somethinghere"
  ;

# Accept multiple lines of statements and wrapped in {  }
BlockStatement
  = "{" StatementList "}"
  ;

# Accept multiple lines of statements
StatementList
  = null
  | Statement
  | StatementList Statement
  ;

::: Please aware that the grammar is over-simplified for the sake of example. :::

Tokenizer

The Tokenizer allows us to include the valid tokens for the syntax, this will help us validate the syntax and give a clear SyntaxError messages.

SyntaxError: Invalid token { at 12:34.

The tokens are characters, symbols, keywords, operators, etc. that are used to build the syntax.

Parser

The parser uses the tokenizer to check for the syntax's grammar and return the parsed value as a usable JSON object.

typescript

class Parser {
  constructor() {
    // Initialize the tokenizer
    this._tokenizer = new Tokenizer();
  }

  parse(string: string): Formkl {
    // Parse the input string
    return this.FormBlock();
  }

  /**
   * Main entry point.
   *
   * FormBlock
   *  = SectionBlockList
   *  ;
   */
  private FormBlock(): Formkl {}

  /**
   * SectionBlockList
   *  = (SectionBlock)*
   *  ;
   */
  private SectionBlockList(): Section[] {}

  private SectionBlock(): Section {}

  // ...
}

Please aware that this code is over-simplified for the sake of example.

Core language #

Grammar #

Tokenizer #

Parser #

Core language

Grammar

Tokenizer

Parser