public ParserRuleContext parse(IEnumerable <SyntaxToken> tokens, Scope scope, int offset)
{
var text = RoslynCompiler.TokensToString(tokens);
AntlrInputStream stream = new AntlrInputStream(text);
ITokenSource lexer = new RLexer(stream);
ITokenStream tokenStream = new CommonTokenStream(lexer);
RParser parser = new RParser(tokenStream);
return(parser.prog());
}
public void ParseR()
{
var inputStream = new AntlrInputStream(@"
x <- 1:3
y <- 2:4
z <- call('plot', x, y)
eval(z)
");
var lexer = new RLexer(inputStream);
var commonTokenStream = new CommonTokenStream(lexer);
var parser = new RParser(commonTokenStream);
var visitor = new CstBuilderForAntlr4(parser);
visitor.Visit(parser.prog());
Console.WriteLine(visitor.FinishParsing());
}
/// <summary> This method filter a R file with "RFilter" and parse it with "R"</summary>
/// <param name="loc"> The R file.</param>
/// <returns> The number of errors in file (NumberOfSyntaxErrors) at filtering and parsing. </returns>
static int Parse(string loc)
{
// Reads characters from a byte stream in a particular encoding.
StreamReader reader = new StreamReader(loc);
// Vacuum all input from a Reader/InputStream and then treat it like a char[] buffer.
// Can also pass in a String or char[] to use.
AntlrInputStream input = new AntlrInputStream(reader);
/* A lexer is recognizer that draws input symbols from a character stream. lexer grammars result in a subclass of this object.
* A Lexer object uses simplified match() and error recovery mechanisms in the interest of speed.
*/
RLexer lexer = new RLexer(input);
/* This class extends BufferedTokenStream with functionality to filter token streams to tokens on a particular channel(tokens
* where Token.getChannel() returns a particular value). This token stream provides access to all tokens by index or when calling
* methods like BufferedTokenStream.getText(). The channel filtering is only used for code accessing tokens via the lookahead
* methods BufferedTokenStream.LA(int), LT(int), and LB(int). By default, tokens are placed on the default channel(Token.DEFAULT_CHANNEL),
* but may be reassigned by using the->channel(HIDDEN) lexer command, or by using an embedded action to call Lexer.setChannel(int).
* Note: lexer rules which use the->skip lexer command or call Lexer.skip() do not produce tokens at all, so input text matched by such a
* rule will not be available as part of the token stream, regardless of channel.
*
* A collection of all tokens fetched from the token source. The list is considered a complete view of the input once fetchedEOF is set to true.
*/
CommonTokenStream tokens = new CommonTokenStream(lexer);
// Print tokens BEFORE filtering.
//tokens.Fill(); // Get all tokens from lexer until EOF
//Console.WriteLine("BEFORE");
//foreach (IToken tok in tokens.GetTokens())
//{
// Console.WriteLine(tok);
// //Console.WriteLine(tok.Text);
//}
RFilter filter = new RFilter(tokens); // Parse with filter.
filter.stream(); // Call start rule: stream .
tokens.Reset(); // Reset all the token (actually use seek(0) for reuse)
//Print tokens AFTER filtering.
//Console.WriteLine("AFTER");
//foreach (IToken tok in tokens.GetTokens())
//{
// Console.WriteLine(tok);
// Console.WriteLine(tok.Text);
//}
RParser parser = new RParser(tokens); // Parse with RParser.
IParseTree tree = parser.prog(); // Call start rule: prog .
PTPrinter PTvisitor = new PTPrinter(loc);
PTvisitor.Visit(tree);
ASTGenerator ast = new ASTGenerator();
ast.Visit(tree);
ASTPrinter ASTVisitor = new ASTPrinter(loc);
ASTVisitor.Visit(ast.M_Root);
ASTScopeVisitor astScopeVisitor = new ASTScopeVisitor(loc);
astScopeVisitor.Visit(ast.M_Root);
return(parser.NumberOfSyntaxErrors + filter.NumberOfSyntaxErrors);
}
public void ParseR() {
var inputStream = new AntlrInputStream(@"
x <- 1:3
y <- 2:4
z <- call('plot', x, y)
eval(z)
");
var lexer = new RLexer(inputStream);
var commonTokenStream = new CommonTokenStream(lexer);
var parser = new RParser(commonTokenStream);
var visitor = new CstBuilderForAntlr4(parser);
visitor.Visit(parser.prog());
Console.WriteLine(visitor.FinishParsing());
}
