001: /*BEGIN_COPYRIGHT_BLOCK
002: *
003: * Copyright (c) 2001-2007, JavaPLT group at Rice University (javaplt@rice.edu)
004: * All rights reserved.
005: *
006: * Redistribution and use in source and binary forms, with or without
007: * modification, are permitted provided that the following conditions are met:
008: * * Redistributions of source code must retain the above copyright
009: * notice, this list of conditions and the following disclaimer.
010: * * Redistributions in binary form must reproduce the above copyright
011: * notice, this list of conditions and the following disclaimer in the
012: * documentation and/or other materials provided with the distribution.
013: * * Neither the names of DrJava, the JavaPLT group, Rice University, nor the
014: * names of its contributors may be used to endorse or promote products
015: * derived from this software without specific prior written permission.
016: *
017: * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
018: * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
019: * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
020: * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
021: * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
022: * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
023: * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
024: * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
025: * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
026: * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
027: * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
028: *
029: * This software is Open Source Initiative approved Open Source Software.
030: * Open Source Initative Approved is a trademark of the Open Source Initiative.
031: *
032: * This file is part of DrJava. Download the current version of this project
033: * from http://www.drjava.org/ or http://sourceforge.net/projects/drjava/
034: *
035: * END_COPYRIGHT_BLOCK*/
036:
037: package edu.rice.cs.util.sexp;
038:
039: import java.io.File;
040: import java.io.Reader;
041: import java.io.FileReader;
042: import java.io.StringReader;
043: import java.io.IOException;
044: import java.util.List;
045: import java.util.ArrayList;
046: import java.util.LinkedList;
047:
048: /**
049: * This parser is not meant to be instantiated. It has
050: * static methods that do all the work for you. These
051: * parse methods take in the data that is to be parsed
052: * and simply returns an s-expression abstract syntax.
053: * @author Jonathan Lugo, PLT Group
054: */
055: public class SExpParser {
056:
057: public static List<SEList> parse(File f) throws SExpParseException,
058: IOException {
059: return parse(new FileReader(f));
060: }
061:
062: public static List<SEList> parse(String s)
063: throws SExpParseException {
064: return parse(new StringReader(s));
065: }
066:
067: public static List<SEList> parse(Reader r)
068: throws SExpParseException {
069: try {
070: return new ParseHelper(r).parseMultiple();
071: } catch (LexingException e) {
072: throw new SExpParseException(e.getMessage());
073: } catch (PrivateParseException e) {
074: throw new SExpParseException(e.getMessage());
075: }
076: }
077:
078: /**
079: * A new helper is instantiated for each time
080: * the user wants to parse data. This is not
081: * reused. The instances of the ParseHelpers are
082: * handled solely in the outer class SExpParser.
083: */
084: private static class ParseHelper {
085:
086: private Lexer _lex;
087:
088: public ParseHelper(Reader r) {
089: _lex = new Lexer(r);
090: }
091:
092: /**
093: * Parse a forest of top-level s-expressions from {@link #parseTopLevelExp()}.
094: * @see #parseTopLevelExp()
095: */
096: public List<SEList> parseMultiple() {
097: ArrayList<SEList> l = new ArrayList<SEList>();
098: SEList exp;
099: while ((exp = parseTopLevelExp()) != null) {
100: l.add(exp);
101: }
102: return l;
103: }
104:
105: /**
106: * A top-level s-expression is simply a non-empty list. Our s-expression files
107: * can be a forest of several trees, but the Atomic values are not allowed
108: * at the top level, only lists.
109: * @return the top-level list s-expression
110: */
111: public SEList parseTopLevelExp() {
112: Tokens.SExpToken t = _lex.readToken();
113: if (t == Tokens.LeftParenToken.ONLY) {
114: return parseList();
115: } else if (t == null) {
116: return null;
117: } else {
118: throw new PrivateParseException(
119: "A top-level s-expression must be a list. "
120: + "Invalid start of list: " + t);
121: }
122: }
123:
124: /**
125: * Parses the next s-expression in the lexer's buffer.
126: * This may be either a cons or an atom
127: * @return the next s-expression in the read buffer.
128: */
129: public SExp parseExp() {
130: Tokens.SExpToken t = _lex.readToken();
131: assertNotEOF(t);
132: if (t == Tokens.LeftParenToken.ONLY) {
133: return parseList();
134: } else {
135: return parseAtom(t);
136: }
137: }
138:
139: /**
140: * The left paren has already been read. This starts
141: * building up the recursive list structure
142: * @return the parsed recursive s-expression list
143: */
144: private SEList parseList() {
145: LinkedList<SExp> list = new LinkedList<SExp>();
146: Tokens.SExpToken t = _lex.peek();
147: assertNotEOF(t);
148:
149: while (t != Tokens.RightParenToken.ONLY) {
150: list.addFirst(parseExp());
151: t = _lex.peek();
152: }
153:
154: // t has to be a Tokens.RightParenToken at this point.
155: // simply eat the token
156: _lex.readToken();
157:
158: // Compile the cons structure from the list of exps
159: SEList cons = Empty.ONLY;
160: for (SExp exp : list) {
161: cons = new Cons(exp, cons);
162: }
163: return cons;
164: }
165:
166: /**
167: * Parses an atom. The token was already read and
168: * found not to start a list, this method interprets
169: * what is given. This method chooses which type of
170: * atom the token represents and creates the atom.
171: * @param t the token to interpret
172: * @return the correct corresponding atom
173: */
174: private Atom parseAtom(Tokens.SExpToken t) {
175: if (t instanceof Tokens.BooleanToken) {
176: if (((Tokens.BooleanToken) t).getValue())
177: return BoolAtom.TRUE;
178: else
179: return BoolAtom.FALSE;
180: } else if (t instanceof Tokens.NumberToken) {
181: return new NumberAtom(((Tokens.NumberToken) t)
182: .getValue());
183: } else if (t instanceof Tokens.QuotedTextToken) {
184: return new QuotedTextAtom(t.getText());
185: } else {
186: return new TextAtom(t.getText());
187: }
188: }
189:
190: /**
191: * Throws the EOF exception if the given token is the end of file
192: * @param t the token to check
193: */
194: private void assertNotEOF(Tokens.SExpToken t) {
195: if (t == null) {
196: throw new PrivateParseException(
197: "Unexpected <EOF> at line " + _lex.lineno());
198: }
199: }
200: }
201:
202: /**
203: * This runtime exception makes it easier to write the parser since
204: * the methods of the helper class won't need to constantly declare
205: * the SExpParseException to be thrown.
206: */
207: private static class PrivateParseException extends RuntimeException {
208: /**
209: * Creates a runtime exception with the message that is desired for
210: * the eventual checked exception
211: * @param msg the message to display
212: */
213: public PrivateParseException(String msg) {
214: super(msg);
215: }
216: }
217: }
|