org.antlr.codegen.Target Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of antlr-complete Show documentation
Complete distribution for ANTLR 3
The newest version!
/*
 * [The "BSD license"]
 *  Copyright (c) 2010 Terence Parr
 *  All rights reserved.
 *
 *  Redistribution and use in source and binary forms, with or without
 *  modification, are permitted provided that the following conditions
 *  are met:
 *  1. Redistributions of source code must retain the above copyright
 *      notice, this list of conditions and the following disclaimer.
 *  2. Redistributions in binary form must reproduce the above copyright
 *      notice, this list of conditions and the following disclaimer in the
 *      documentation and/or other materials provided with the distribution.
 *  3. The name of the author may not be used to endorse or promote products
 *      derived from this software without specific prior written permission.
 *
 *  THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
 *  IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
 *  OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
 *  IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
 *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
 *  NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 *  DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 *  THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
 *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */
package org.antlr.codegen;

import org.antlr.Tool;
import org.antlr.analysis.Label;
import org.antlr.runtime.Token;
import org.stringtemplate.v4.ST;
import org.antlr.tool.Grammar;

import java.io.IOException;
import java.util.List;

/** The code generator for ANTLR can usually be retargeted just by providing
 *  a new X.stg file for language X, however, sometimes the files that must
 *  be generated vary enough that some X-specific functionality is required.
 *  For example, in C, you must generate header files whereas in Java you do not.
 *  Other languages may want to keep DFA separate from the main
 *  generated recognizer file.
 *
 *  The notion of a Code Generator target abstracts out the creation
 *  of the various files.  As new language targets get added to the ANTLR
 *  system, this target class may have to be altered to handle more
 *  functionality.  Eventually, just about all language generation issues
 *  will be expressible in terms of these methods.
 *
 *  If org.antlr.codegen.XTarget class exists, it is used else
 *  Target base class is used.  I am using a superclass rather than an
 *  interface for this target concept because I can add functionality
 *  later without breaking previously written targets (extra interface
 *  methods would force adding dummy functions to all code generator
 *  target classes).
 *
 */
public class Target {

	/** For pure strings of Java 16-bit unicode char, how can we display
	 *  it in the target language as a literal.  Useful for dumping
	 *  predicates and such that may refer to chars that need to be escaped
	 *  when represented as strings.  Also, templates need to be escaped so
	 *  that the target language can hold them as a string.
	 *
	 *  I have defined (via the constructor) the set of typical escapes,
	 *  but your Target subclass is free to alter the translated chars or
	 *  add more definitions.  This is nonstatic so each target can have
	 *  a different set in memory at same time.
	 */
	protected String[] targetCharValueEscape = new String[255];

	public Target() {
		targetCharValueEscape['\n'] = "\\n";
		targetCharValueEscape['\r'] = "\\r";
		targetCharValueEscape['\t'] = "\\t";
		targetCharValueEscape['\b'] = "\\b";
		targetCharValueEscape['\f'] = "\\f";
		targetCharValueEscape['\\'] = "\\\\";
		targetCharValueEscape['\''] = "\\'";
		targetCharValueEscape['"'] = "\\\"";
	}

    public boolean useBaseTemplatesForSynPredFragments() {
        return true;
    }

	protected void genRecognizerFile(Tool tool,
									 CodeGenerator generator,
									 Grammar grammar,
									 ST outputFileST)
		throws IOException
	{
		String fileName =
			generator.getRecognizerFileName(grammar.name, grammar.type);
		generator.write(outputFileST, fileName);
	}

	protected void genRecognizerHeaderFile(Tool tool,
										   CodeGenerator generator,
										   Grammar grammar,
										   ST headerFileST,
										   String extName) // e.g., ".h"
		throws IOException
	{
		// no header file by default
	}

	protected void performGrammarAnalysis(CodeGenerator generator,
										  Grammar grammar)
	{
		// Build NFAs from the grammar AST
		grammar.buildNFA();

		// Create the DFA predictors for each decision
		grammar.createLookaheadDFAs();
	}

	/** Is scope in @scope::name {action} valid for this kind of grammar?
	 *  Targets like C++ may want to allow new scopes like headerfile or
	 *  some such.  The action names themselves are not policed at the
	 *  moment so targets can add template actions w/o having to recompile
	 *  ANTLR.
	 */
	public boolean isValidActionScope(int grammarType, String scope) {
		switch (grammarType) {
			case Grammar.LEXER :
				if ( scope.equals("lexer") ) {return true;}
				break;
			case Grammar.PARSER :
				if ( scope.equals("parser") ) {return true;}
				break;
			case Grammar.COMBINED :
				if ( scope.equals("parser") ) {return true;}
				if ( scope.equals("lexer") ) {return true;}
				break;
			case Grammar.TREE_PARSER :
				if ( scope.equals("treeparser") ) {return true;}
				break;
		}
		return false;
	}

	/** Target must be able to override the labels used for token types */
	public String getTokenTypeAsTargetLabel(CodeGenerator generator, int ttype) {
		String name = generator.grammar.getTokenDisplayName(ttype);
		// If name is a literal, return the token type instead
		if ( name.charAt(0)=='\'' ) {
			return String.valueOf(ttype);
		}
		return name;
	}

	/** Convert from an ANTLR char literal found in a grammar file to
	 *  an equivalent char literal in the target language.  For most
	 *  languages, this means leaving 'x' as 'x'.  Actually, we need
	 *  to escape '\u000A' so that it doesn't get converted to \n by
	 *  the compiler.  Convert the literal to the char value and then
	 *  to an appropriate target char literal.
	 *
	 *  Expect single quotes around the incoming literal.
	 */
	public String getTargetCharLiteralFromANTLRCharLiteral(
		CodeGenerator generator,
		String literal)
	{
		StringBuilder buf = new StringBuilder();
		buf.append('\'');
		int c = Grammar.getCharValueFromGrammarCharLiteral(literal);
		if ( c postProcessAction(List