[gvpr] Add new grammar gvpr (fix for #3803) (#3809)

* Fix for #3803

* Add tests from graphviz source code. Reformat grammar.

* Add testing.

* Rename to avoid symbol conflict.

* Update grammar, targets.

* Remove ambiguity.

* Update statement_list and action_list to not be empty.

* Update grammar with prec/assoc, types, semicolon decls.

* Split grammar because semantic predicate needed. Added tests.

* Update errors file.

* Adding Java target.

* Fix java target.

* Add Dart target.

* Add JavaScript target.

* Add TypeScript port.

* Add Python3 port.

* Add Cpp port.

* Add Go port.

* Add new tests from graphviz source code (renamed with .gvpr extension). Fix grammar a little.

* Fix parse errors in new tests.

* Move DigitSquence down in grammar, and change to fragment (it should not be a token).

* Add test, update escape char set.

Author: kaby76

gvpr/CSharp/GvprLexerBase.cs

@@ -0,0 +1,29 @@
+using System;
+using System.IO;
+using System.Reflection;
+using Antlr4.Runtime;
+using Antlr4.Runtime.Misc;
+
+public class GvprLexerBase : Lexer
+{
+    protected GvprLexerBase(ICharStream input, TextWriter output, TextWriter errorOutput)
+        : base(input, output, errorOutput)
+    {
+    }
+
+    public GvprLexerBase(ICharStream input)
+        : base(input)
+    {
+    }
+
+    public override string[] RuleNames => throw new NotImplementedException();
+
+    public override IVocabulary Vocabulary => throw new NotImplementedException();
+
+    public override string GrammarFileName => throw new NotImplementedException();
+
+    public bool IsColumnZero()
+    {
+        return this.Column == 1;
+    }
+}

gvpr/CSharp/GvprParserBase.cs

@@ -0,0 +1,26 @@
+using Antlr4.Runtime;
+using System.IO;
+
+public abstract class GvprParserBase : Parser {
+    private readonly ITokenStream _input;
+
+    protected GvprParserBase(ITokenStream input, TextWriter output, TextWriter errorOutput)
+        : base(input, output, errorOutput)
+    {
+        _input = input;
+    }
+
+    public bool IsSemiRequired()
+    {
+        var c = (this.InputStream as CommonTokenStream).LT(-1);
+        var d = (this.InputStream as CommonTokenStream).LT(1);
+        return c.Type != gvprParser.CCBC;
+    }
+
+    public bool IsSemiNotRequired()
+    {
+        var c = (this.InputStream as CommonTokenStream).LT(-1);
+        var d = (this.InputStream as CommonTokenStream).LT(1);
+        return c.Type == gvprParser.CCBC;
+    }
+}

gvpr/Cpp/GvprLexerBase.cpp

@@ -0,0 +1,13 @@
+#include "antlr4-runtime.h"
+#include "GvprLexerBase.h"
+#include "gvprLexer.h"
+
+GvprLexerBase::GvprLexerBase(antlr4::CharStream * input) : antlr4::Lexer(input)
+{
+    _input = input;
+}
+
+bool GvprLexerBase::IsColumnZero()
+{
+    return this->getCharPositionInLine() == 1;
+}

gvpr/Cpp/GvprLexerBase.h

@@ -0,0 +1,9 @@
+#pragma once
+#include "antlr4-runtime.h"
+
+class GvprLexerBase : public antlr4::Lexer
+{
+    public:
+        GvprLexerBase(antlr4::CharStream * input);
+        bool IsColumnZero();
+};

gvpr/Cpp/GvprParserBase.cpp

@@ -0,0 +1,17 @@
+#include "gvprParser.h"
+
+using namespace antlr4;
+
+bool GvprParserBase::IsSemiRequired()
+{
+	auto c = this->getTokenStream()->LT(-1);
+	auto d = this->getTokenStream()->LT(1);
+	return c->getType() != gvprParser::CCBC;
+}
+
+bool GvprParserBase::IsSemiNotRequired()
+{
+	auto c = this->getTokenStream()->LT(-1);
+	auto d = this->getTokenStream()->LT(1);
+	return c->getType() == gvprParser::CCBC;
+}

gvpr/Cpp/GvprParserBase.h

@@ -0,0 +1,10 @@
+#pragma once
+
+#include "antlr4-runtime.h"
+
+class GvprParserBase : public antlr4::Parser {
+public:
+    GvprParserBase(antlr4::TokenStream *input) : Parser(input) { }
+    bool IsSemiRequired();
+    bool IsSemiNotRequired();
+};

gvpr/Cpp/transformGrammar.py

@@ -0,0 +1,34 @@
+import sys, os, re, shutil
+from glob import glob
+from pathlib import Path
+
+def main(argv):
+    for file in glob("./*.g4"):
+        fix(file)
+
+def fix(file_path):
+    print("Altering " + file_path)
+    if not os.path.exists(file_path):
+        print(f"Could not find file: {file_path}")
+        sys.exit(1)
+    parts = os.path.split(file_path)
+    file_name = parts[-1]
+    shutil.move(file_path, file_path + ".bak")
+    input_file = open(file_path + ".bak",'r')
+    output_file = open(file_path, 'w')
+    for x in input_file:
+        if '// Insert here @header for C++ lexer.' in x:
+            x = x.replace('// Insert here @header for C++ lexer.', '@header {#include "GvprLexerBase.h"}')
+        if '// Insert here @header for C++ parser.' in x:
+            x = x.replace('// Insert here @header for C++ parser.', '@header {#include "GvprParserBase.h"}')
+        if 'this.' in x:
+            x = x.replace('this.', 'this->')
+        output_file.write(x)
+        output_file.flush()
+
+    print("Writing ...")
+    input_file.close()
+    output_file.close()
+
+if __name__ == '__main__':
+    main(sys.argv)

gvpr/Dart/GvprLexerBase.dart

@@ -0,0 +1,18 @@
+import 'package:antlr4/antlr4.dart';
+import 'dart:io';
+import 'dart:core';
+import 'dart:convert';
+import 'dart:collection';
+import 'gvprLexer.dart';
+
+abstract class GvprLexerBase extends Lexer
+{
+    GvprLexerBase(CharStream input) : super(input)
+    {
+    }
+
+    bool IsColumnZero()
+    {
+        return this.charPositionInLine == 1;
+    }
+}

gvpr/Dart/GvprParserBase.dart

@@ -0,0 +1,24 @@
+import 'package:antlr4/antlr4.dart';
+import 'dart:io';
+import 'dart:convert';
+import 'gvprParser.dart';
+
+abstract class GvprParserBase extends Parser {
+    GvprParserBase(TokenStream input) : super(input)
+    {
+    }
+
+    bool IsSemiRequired()
+    {
+        var c = this.tokenStream.LT(-1);
+        var d = this.tokenStream.LT(1);
+        return c?.type != gvprParser.TOKEN_CCBC;
+    }
+
+    bool IsSemiNotRequired()
+    {
+        var c = this.tokenStream.LT(-1);
+        var d = this.tokenStream.LT(1);
+        return c?.type == gvprParser.TOKEN_CCBC;
+    }
+}

gvpr/Go/gvpr_lexer_base.go

@@ -0,0 +1,13 @@
+package parser
+
+import (
+    "github.com/antlr4-go/antlr/v4"
+)
+
+type GvprLexerBase struct {
+    *antlr.BaseLexer
+}
+
+func (l *GvprLexerBase) IsColumnZero() bool {
+    return l.GetCharPositionInLine() == 1
+}