chore: modified the file structure to implement the new way of antlr parsing

on-behalf-of: @Mermaid-Chart <[email protected]>

Author: omkarht

.esbuild/server.ts

@@ -4,6 +4,7 @@ import cors from 'cors';
 import { context } from 'esbuild';
 import type { Request, Response } from 'express';
 import express from 'express';
+import { execSync } from 'child_process';
 import { packageOptions } from '../.build/common.js';
 import { generateLangium } from '../.build/generateLangium.js';
 import { defaultOptions, getBuildConfig } from './util.js';
@@ -64,6 +65,28 @@ function eventsHandler(request: Request, response: Response) {
 }
 
 let timeoutID: NodeJS.Timeout | undefined = undefined;
+let isGeneratingAntlr = false;
+
+/**
+ * Generate ANTLR parser files from grammar files
+ */
+function generateAntlr(): void {
+  if (isGeneratingAntlr) {
+    console.log('⏳ ANTLR generation already in progress, skipping...');
+    return;
+  }
+
+  try {
+    isGeneratingAntlr = true;
+    console.log('🎯 ANTLR: Generating parser files...');
+    execSync('tsx scripts/antlr-generate.mts', { stdio: 'inherit' });
+    console.log('✅ ANTLR: Parser files generated successfully\n');
+  } catch (error) {
+    console.error('❌ ANTLR: Failed to generate parser files:', error);
+  } finally {
+    isGeneratingAntlr = false;
+  }
+}
 
 /**
  * Debounce file change events to avoid rebuilding multiple times.
@@ -89,7 +112,7 @@ async function createServer() {
   handleFileChange();
   const app = express();
   chokidar
-    .watch('**/src/**/*.{js,ts,langium,yaml,json}', {
+    .watch('**/src/**/*.{js,ts,g4,langium,yaml,json}', {
       ignoreInitial: true,
       ignored: [/node_modules/, /dist/, /docs/, /coverage/],
     })
@@ -103,6 +126,9 @@ async function createServer() {
       if (path.endsWith('.langium')) {
         await generateLangium();
       }
+      if (path.endsWith('.g4')) {
+        generateAntlr();
+      }
       handleFileChange();
     });
 

package.json

@@ -15,8 +15,10 @@
     "git graph"
   ],
   "scripts": {
-    "build": "pnpm build:esbuild && pnpm build:types",
+    "build": "pnpm antlr:generate && pnpm build:esbuild && pnpm build:types",
     "build:esbuild": "pnpm run -r clean && tsx .esbuild/build.ts",
+    "antlr:generate": "tsx scripts/antlr-generate.mts",
+    "antlr:watch": "tsx scripts/antlr-watch.mts",
     "build:mermaid": "pnpm build:esbuild --mermaid",
     "build:viz": "pnpm build:esbuild --visualize",
     "build:types": "pnpm --filter mermaid types:build-config && tsx .build/types.ts",

packages/mermaid/package.json

@@ -34,6 +34,8 @@
   "scripts": {
     "clean": "rimraf dist",
     "dev": "pnpm -w dev",
+    "antlr:generate": "tsx ../../scripts/antlr-generate.mts",
+    "antlr:watch": "tsx ../../scripts/antlr-watch.mts",
     "docs:code": "typedoc src/defaultConfig.ts src/config.ts src/mermaid.ts && prettier --write ./src/docs/config/setup",
     "docs:build": "rimraf ../../docs && pnpm docs:code && pnpm docs:spellcheck && tsx scripts/docs.cli.mts",
     "docs:verify": "pnpm docs:code && pnpm docs:spellcheck && tsx scripts/docs.cli.mts --verify",
@@ -71,6 +73,7 @@
     "@iconify/utils": "^3.0.2",
     "@mermaid-js/parser": "workspace:^",
     "@types/d3": "^7.4.3",
+    "antlr4ng": "^3.0.7",
     "cytoscape": "^3.33.1",
     "cytoscape-cose-bilkent": "^4.1.0",
     "cytoscape-fcose": "^2.2.0",

packages/mermaid/src/diagrams/usecase/parser/antlr/UsecaseErrorListener.ts

@@ -0,0 +1,90 @@
+import { BaseErrorListener } from 'antlr4ng';
+import type { RecognitionException, Recognizer } from 'antlr4ng';
+
+/**
+ * Custom error listener for ANTLR usecase parser
+ * Captures syntax errors and provides detailed error messages
+ */
+export class UsecaseErrorListener extends BaseErrorListener {
+  private errors: { line: number; column: number; message: string; offendingSymbol?: any }[] = [];
+
+  syntaxError(
+    _recognizer: Recognizer<any>,
+    offendingSymbol: any,
+    line: number,
+    charPositionInLine: number,
+    message: string,
+    _e: RecognitionException | null
+  ): void {
+    this.errors.push({
+      line,
+      column: charPositionInLine,
+      message,
+      offendingSymbol,
+    });
+  }
+
+  reportAmbiguity(): void {
+    // Optional: handle ambiguity reports
+  }
+
+  reportAttemptingFullContext(): void {
+    // Optional: handle full context attempts
+  }
+
+  reportContextSensitivity(): void {
+    // Optional: handle context sensitivity reports
+  }
+
+  getErrors(): { line: number; column: number; message: string; offendingSymbol?: any }[] {
+    return this.errors;
+  }
+
+  hasErrors(): boolean {
+    return this.errors.length > 0;
+  }
+
+  clear(): void {
+    this.errors = [];
+  }
+
+  /**
+   * Create a detailed error with JISON-compatible hash property
+   */
+  createDetailedError(): Error {
+    if (this.errors.length === 0) {
+      return new Error('Unknown parsing error');
+    }
+
+    const firstError = this.errors[0];
+    const message = `Parse error on line ${firstError.line}: ${firstError.message}`;
+    const error = new Error(message);
+
+    // Add hash property for JISON compatibility
+    Object.assign(error, {
+      hash: {
+        line: firstError.line,
+        loc: {
+          first_line: firstError.line,
+          last_line: firstError.line,
+          first_column: firstError.column,
+          last_column: firstError.column,
+        },
+        text: firstError.offendingSymbol?.text ?? '',
+        token: firstError.offendingSymbol?.text ?? '',
+        expected: [],
+      },
+    });
+
+    return error;
+  }
+
+  /**
+   * Get all error messages as a single string
+   */
+  getErrorMessages(): string {
+    return this.errors
+      .map((error) => `Line ${error.line}:${error.column} - ${error.message}`)
+      .join('\n');
+  }
+}

packages/mermaid/src/diagrams/usecase/parser/antlr/UsecaseLexer.g4

@@ -0,0 +1,65 @@
+lexer grammar UsecaseLexer;
+
+// Keywords
+ACTOR: 'actor';
+SYSTEM_BOUNDARY: 'systemBoundary';
+END: 'end';
+DIRECTION: 'direction';
+CLASS_DEF: 'classDef';
+CLASS: 'class';
+STYLE: 'style';
+USECASE: 'usecase';
+
+// Direction keywords
+TB: 'TB';
+TD: 'TD';
+BT: 'BT';
+RL: 'RL';
+LR: 'LR';
+
+// System boundary types
+PACKAGE: 'package';
+RECT: 'rect';
+TYPE: 'type';
+
+// Arrow types (order matters - longer patterns first)
+SOLID_ARROW: '-->';
+BACK_ARROW: '<--';
+CIRCLE_ARROW: '--o';
+CIRCLE_ARROW_REVERSED: 'o--';
+CROSS_ARROW: '--x';
+CROSS_ARROW_REVERSED: 'x--';
+LINE_SOLID: '--';
+
+// Symbols
+COMMA: ',';
+AT: '@';
+LBRACE: '{';
+RBRACE: '}';
+COLON: ':';
+LPAREN: '(';
+RPAREN: ')';
+CLASS_SEPARATOR: ':::';
+
+// Hash color (must come before HASH to avoid conflicts)
+HASH_COLOR: '#' [a-fA-F0-9]+;
+
+// Number with optional unit
+NUMBER: [0-9]+ ('.' [0-9]+)? ([a-zA-Z]+)?;
+
+// Identifier
+IDENTIFIER: [a-zA-Z_][a-zA-Z0-9_]*;
+
+// String literals
+STRING: '"' (~["\r\n])* '"' | '\'' (~['\r\n])* '\'';
+
+// These tokens are defined last so they have lowest priority
+// This ensures arrow tokens like '-->' are matched before DASH
+DASH: '-';
+DOT: '.';
+PERCENT: '%';
+
+// Whitespace and newlines
+NEWLINE: [\r\n]+;
+WS: [ \t]+ -> skip;
+