WIP: use an "unknown" rule that matches everything

Author: kasbah

js/bin/electro-grammar.js

@@ -1,5 +1,5 @@
 const cli = require('cli');
-const lib = require('../lib/index');
+const lib = require('../lib/lax_parser');
 
 const args = cli.parse({
   parser: [ 'p', 'Parser to use (strict | lax)', 'string', 'strict' ],

js/lib/lax_parser.js

@@ -1,51 +1,46 @@
 const {ElectroGrammarLexer} = require('./ElectroGrammarLexer');
 const {ElectroGrammarParser} = require('./ElectroGrammarParser');
+const {ElectroGrammarListener} = require('./ElectroGrammarListener');
 const {ToObjectListener} = require('./to_object_listener');
 const antlr4 = require('antlr4');
 
-class LexerIgnoreListener extends antlr4.error.ErrorListener {
+class IgnoreListener extends ElectroGrammarListener {
   constructor() {
     super();
-    this.ignored = '';
+    this.obj = '';
   }
-  syntaxError(lexer, offendingSymbol, line, char, err) {
-    console.error('lexer:', err);
-    // hopefully there is a better way to get the input, but I haven't found it
-    let input = err.split(':')[1].trim();
-    input = input.substring(1, input.length - 1);
-    this.ignored += input;
+  exitIgnored(ctx) {
+    ctx.UNKNOWN().forEach(s => {
+      this.obj += s.getText();
+    });
   }
 }
 
-class ParserIgnoreListener extends antlr4.error.ErrorListener {
-  constructor() {
-    super();
-    this.ignored = '';
-  }
-  syntaxError(lexer, offendingSymbol, line, char, err) {
-    console.error('parser:', err);
-    //this.ignored += input;
+function get_parser(start_rule, listener) {
+  function parse(input) {
+    const chars = new antlr4.InputStream(input);
+    const lexer = new ElectroGrammarLexer(chars);
+    const tokens = new antlr4.CommonTokenStream(lexer);
+    const parser = new ElectroGrammarParser(tokens);
+    parser.buildParseTrees = true;
+
+    const tree = parser[start_rule]();
+    antlr4.tree.ParseTreeWalker.DEFAULT.walk(listener, tree);
+    return listener.obj;
   }
+  return parse;
 }
 
 function parse(input) {
-  const chars = new antlr4.InputStream(input);
-  const lexer = new ElectroGrammarLexer(chars);
-  const lexerIgnorer = new LexerIgnoreListener();
-  lexer.removeErrorListeners();
-  lexer.addErrorListener(lexerIgnorer);
-  const tokens = new antlr4.CommonTokenStream(lexer);
-  const parser = new ElectroGrammarParser(tokens);
-  const parserIgnorer = new ParserIgnoreListener();
-  parser.buildParseTrees = true;
-  parser.removeErrorListeners();
-  parser.addErrorListener(parserIgnorer);
+  return get_parser('electro_grammar')(input);
+}
 
-  const tree = parser.electro_grammar();
-  const listener = new ToObjectListener();
-  const walker = antlr4.tree.ParseTreeWalker.DEFAULT.walk(listener, tree);
-  console.log({lexer: lexerIgnorer.ignored, parser: parserIgnorer.ignored});
-  return {component: listener.obj, ignored: lexerIgnorer.ignored};
+function parse(input) {
+  const parseComponent = get_parser('electro_grammar', new ToObjectListener());
+  const component = parseComponent(input);
+  const parseIgnored = get_parser('ignored', new IgnoreListener());
+  const ignored = parseIgnored(input);
+  return {component, ignored};
 }
 
 module.exports = {parse};

src/ElectroGrammar.g4

@@ -1,10 +1,8 @@
-/*
- * Lexer Rules
- */
-
 grammar ElectroGrammar;
-import Passive, Semi;
+import Passive;
+
+electro_grammar: passive EOF;
 
-electro_grammar: (passive | semi) EOF;
+ignored: UNKNOWN* EOF;
 
-WHITESPACE: [\p{White_Space}] -> skip;
+UNKNOWN: .+?;

src/Passive.g4

@@ -1,22 +1,6 @@
 grammar Passive;
-import Dielectric, Package, Units;
+import Units;
 
-passive: resistor | capacitor | inductor | oscillator;
+passive: resistor;
 
-passive: resistor | capacitor | inductor | oscillator;
-resistor: resistance rspec*;
-rspec: rtype | rpackage | power;
-rtype: POT;
-rpackage: package_chip;
-POT: P O T | P O T E N T I O M E T E R;
-
-capacitor: capacitance cspec*;
-cspec: dielectric | cpackage | voltage;
-cpackage: package_chip;
-
-inductor: inductance lspec*;
-lspec: lpackage | current;
-lpackage: package_chip;
-
-oscillator: frequency ospec*;
-ospec: capacitance;
+resistor: resistance;

src/Units.g4

@@ -48,8 +48,8 @@ aprefix: MILI | MICRO | NANO | PICO;
 power: NUMBER pprefix? WATT tolerance?;
 pprefix: MILI;
 
-resistance: NUMBER (rprefix | OHM) tolerance?;
-rprefix: MEGA | KILO | MILI;
+resistance: NUMBER (RPREFIX | OHM);
+RPREFIX: MEGA | KILO | MILI;
 
 capacitance: NUMBER cprefix FARAD tolerance?;
 cprefix: MICRO | NANO | PICO;