X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;f=src%2Forg%2Fz3950%2Fzing%2Fcql%2FCQLParser.java;h=6d49c56830a88b99478c8147a9ef7a5a0af0d9da;hb=8e35f984081a29de9b53af7c1b2fbf981c474d88;hp=16d9bd6f52c348134c5b64bc44f8b91e12779a34;hpb=9b7bbaf04b9492a1071d10fe5a1ce3f1ee60a0fc;p=cql-java-moved-to-github.git diff --git a/src/org/z3950/zing/cql/CQLParser.java b/src/org/z3950/zing/cql/CQLParser.java index 16d9bd6..6d49c56 100644 --- a/src/org/z3950/zing/cql/CQLParser.java +++ b/src/org/z3950/zing/cql/CQLParser.java @@ -1,73 +1,82 @@ -// $Id: CQLParser.java,v 1.8 2002-10-27 00:46:25 mike Exp $ +// $Id: CQLParser.java,v 1.33 2007-06-29 12:27:08 mike Exp $ package org.z3950.zing.cql; +import java.io.IOException; +import java.util.Vector; import java.util.Properties; import java.io.InputStream; -import java.io.IOException; -import java.io.StringReader; -import java.io.StreamTokenizer; +import java.io.FileInputStream; +import java.io.FileNotFoundException; /** - * Compiles a CQL string into a parse tree ... - * ### + * Compiles CQL strings into parse trees of CQLNode subtypes. * - * @version $Id: CQLParser.java,v 1.8 2002-10-27 00:46:25 mike Exp $ + * @version $Id: CQLParser.java,v 1.33 2007-06-29 12:27:08 mike Exp $ * @see http://zing.z3950.org/cql/index.html */ public class CQLParser { private CQLLexer lexer; - static private boolean PARSEDEBUG = false; + static private boolean DEBUG = false; static private boolean LEXDEBUG = false; - private class CQLParseException extends Exception { - CQLParseException(String s) { super(s); } - } - - static void debug(String str) { - if (PARSEDEBUG) + private static void debug(String str) { + if (DEBUG) System.err.println("PARSEDEBUG: " + str); } + /** + * Compiles a CQL query. + *

+ * The resulting parse tree may be further processed by hand (see + * the individual node-types' documentation for details on the + * data structure) or, more often, simply rendered out in the + * desired form using one of the back-ends. toCQL() + * returns a decompiled CQL query equivalent to the one that was + * compiled in the first place; toXCQL() returns an + * XML snippet representing the query; and toPQF() + * returns the query rendered in Index Data's Prefix Query + * Format. + * + * @param cql The query + * @return A CQLNode object which is the root of a parse + * tree representing the query. */ public CQLNode parse(String cql) throws CQLParseException, IOException { lexer = new CQLLexer(cql, LEXDEBUG); lexer.nextToken(); - debug("about to parse_query()"); - CQLNode root = parse_query("srw.serverChoice", "="); + debug("about to parseQuery()"); + CQLNode root = parseQuery("srw.serverChoice", new CQLRelation("scr")); if (lexer.ttype != lexer.TT_EOF) throw new CQLParseException("junk after end: " + lexer.render()); return root; } - private CQLNode parse_query(String qualifier, String relation) + private CQLNode parseQuery(String index, CQLRelation relation) throws CQLParseException, IOException { - debug("in parse_query()"); - - CQLNode term = parse_term(qualifier, relation); - while (lexer.ttype == lexer.TT_WORD) { - String op = lexer.sval.toLowerCase(); - debug("checking op '" + op + "'"); - if (lexer.sval.equals("and")) { - match(lexer.TT_WORD); - CQLNode term2 = parse_term(qualifier, relation); - term = new CQLAndNode(term, term2); - } else if (lexer.sval.equals("or")) { - match(lexer.TT_WORD); - CQLNode term2 = parse_term(qualifier, relation); - term = new CQLOrNode(term, term2); - } else if (lexer.sval.equals("not")) { - match(lexer.TT_WORD); - CQLNode term2 = parse_term(qualifier, relation); - term = new CQLNotNode(term, term2); - } else if (lexer.sval.equals("prox")) { - // ### Handle "prox" + debug("in parseQuery()"); + + CQLNode term = parseTerm(index, relation); + while (lexer.ttype != lexer.TT_EOF && lexer.ttype != ')') { + if (lexer.ttype == lexer.TT_AND || + lexer.ttype == lexer.TT_OR || + lexer.ttype == lexer.TT_NOT || + lexer.ttype == lexer.TT_PROX) { + int type = lexer.ttype; + String val = lexer.sval; + match(type); + ModifierSet ms = gatherModifiers(val); + CQLNode term2 = parseTerm(index, relation); + term = ((type == lexer.TT_AND) ? new CQLAndNode(term, term2, ms) : + (type == lexer.TT_OR) ? new CQLOrNode (term, term2, ms) : + (type == lexer.TT_NOT) ? new CQLNotNode(term, term2, ms) : + new CQLProxNode(term, term2, ms)); } else { - throw new CQLParseException("unrecognised boolean: '" + - lexer.sval + "'"); + throw new CQLParseException("expected boolean, got " + + lexer.render()); } } @@ -75,172 +84,290 @@ public class CQLParser { return term; } - private CQLNode parse_term(String qualifier, String relation) + private ModifierSet gatherModifiers(String base) + throws CQLParseException, IOException { + debug("in gatherModifiers()"); + + ModifierSet ms = new ModifierSet(base); + while (lexer.ttype == '/') { + match('/'); + if (lexer.ttype != lexer.TT_WORD) + throw new CQLParseException("expected modifier, " + + "got " + lexer.render()); + String type = lexer.sval.toLowerCase(); + match(lexer.ttype); + if (!isRelation()) { + // It's a simple modifier consisting of type only + ms.addModifier(type); + } else { + // It's a complex modifier of the form type=value + String comparision = lexer.render(lexer.ttype, false); + match(lexer.ttype); + String value = matchSymbol("modifier value"); + ms.addModifier(type, comparision, value); + } + } + + return ms; + } + + private CQLNode parseTerm(String index, CQLRelation relation) throws CQLParseException, IOException { - debug("in parse_term()"); + debug("in parseTerm()"); String word; while (true) { if (lexer.ttype == '(') { debug("parenthesised term"); match('('); - CQLNode expr = parse_query(qualifier, relation); + CQLNode expr = parseQuery(index, relation); match(')'); return expr; - } else if (lexer.ttype != lexer.TT_WORD && lexer.ttype != '"') { - throw new CQLParseException("expected qualifier or term, " + - "got " + lexer.render()); + } else if (lexer.ttype == '>') { + match('>'); + return parsePrefix(index, relation); } debug("non-parenthesised term"); - word = lexer.sval; - match(lexer.ttype); - if (!isRelation()) + word = matchSymbol("index or term"); + if (!isRelation() && lexer.ttype != lexer.TT_WORD) break; - qualifier = word; - relation = lexer.render(false); + index = word; + String relstr = (lexer.ttype == lexer.TT_WORD ? + lexer.sval : lexer.render(lexer.ttype, false)); + relation = new CQLRelation(relstr); match(lexer.ttype); - debug("qualifier='" + qualifier + ", relation='" + relation + "'"); + ModifierSet ms = gatherModifiers(relstr); + relation.setModifiers(ms); + debug("index='" + index + ", " + + "relation='" + relation.toCQL() + "'"); } - CQLTermNode node = new CQLTermNode(qualifier, relation, word); - debug("made term node " + node); + CQLTermNode node = new CQLTermNode(index, relation, word); + debug("made term node " + node.toCQL()); return node; } - boolean isRelation() { - // ### Also need to handle <=, >=, <> + private CQLNode parsePrefix(String index, CQLRelation relation) + throws CQLParseException, IOException { + debug("prefix mapping"); + + String name = null; + String identifier = matchSymbol("prefix-name"); + if (lexer.ttype == '=') { + match('='); + name = identifier; + identifier = matchSymbol("prefix-identifer"); + } + CQLNode term = parseQuery(index, relation); + return new CQLPrefixNode(name, identifier, term); + } + + // Checks for a relation + private boolean isRelation() { + debug("isRelation: checking ttype=" + lexer.ttype + + " (" + lexer.render() + ")"); return (lexer.ttype == '<' || lexer.ttype == '>' || - lexer.ttype == '='); + lexer.ttype == '=' || + lexer.ttype == lexer.TT_LE || + lexer.ttype == lexer.TT_GE || + lexer.ttype == lexer.TT_NE); } private void match(int token) throws CQLParseException, IOException { - debug("in match(" + lexer.render(token, null, true) + ")"); + debug("in match(" + lexer.render(token, true) + ")"); if (lexer.ttype != token) throw new CQLParseException("expected " + - lexer.render(token, null, true) + + lexer.render(token, true) + ", " + "got " + lexer.render()); - lexer.nextToken(); + int tmp = lexer.nextToken(); + debug("match() got token=" + lexer.ttype + ", " + + "nval=" + lexer.nval + ", sval='" + lexer.sval + "'" + + " (tmp=" + tmp + ")"); } + private String matchSymbol(String expected) + throws CQLParseException, IOException { + + debug("in matchSymbol()"); + if (lexer.ttype == lexer.TT_WORD || + lexer.ttype == lexer.TT_NUMBER || + lexer.ttype == '"' || + // The following is a complete list of keywords. Because + // they're listed here, they can be used unquoted as + // indexes, terms, prefix names and prefix identifiers. + // ### Instead, we should ask the lexer whether what we + // have is a keyword, and let the knowledge reside there. + lexer.ttype == lexer.TT_AND || + lexer.ttype == lexer.TT_OR || + lexer.ttype == lexer.TT_NOT || + lexer.ttype == lexer.TT_PROX || + lexer.ttype == lexer.TT_pWORD || + lexer.ttype == lexer.TT_SENTENCE || + lexer.ttype == lexer.TT_PARAGRAPH || + lexer.ttype == lexer.TT_ELEMENT || + lexer.ttype == lexer.TT_ORDERED || + lexer.ttype == lexer.TT_UNORDERED) { + String symbol = (lexer.ttype == lexer.TT_NUMBER) ? + lexer.render() : lexer.sval; + match(lexer.ttype); + return symbol; + } + + throw new CQLParseException("expected " + expected + ", " + + "got " + lexer.render()); + } - // Test harness. - // - // e.g. echo '(au=Kerninghan or au=Ritchie) and ti=Unix' | - // java org.z3950.zing.cql.CQLParser - // yields: - // - // and - // - // or - // - // au - // = - // Kerninghan - // - // - // au - // = - // Ritchie - // - // - // - // ti - // = - // Unix - // - // - // + + /** + * Simple test-harness for the CQLParser class. + *

+ * Reads a CQL query either from its command-line argument, if + * there is one, or standard input otherwise. So these two + * invocations are equivalent: + *

+     *  CQLParser 'au=(Kerninghan or Ritchie) and ti=Unix'
+     *  echo au=(Kerninghan or Ritchie) and ti=Unix | CQLParser
+     * 
+ * The test-harness parses the supplied query and renders is as + * XCQL, so that both of the invocations above produce the + * following output: + *
+     *	<triple>
+     *	  <boolean>
+     *	    <value>and</value>
+     *	  </boolean>
+     *	  <triple>
+     *	    <boolean>
+     *	      <value>or</value>
+     *	    </boolean>
+     *	    <searchClause>
+     *	      <index>au</index>
+     *	      <relation>
+     *	        <value>=</value>
+     *	      </relation>
+     *	      <term>Kerninghan</term>
+     *	    </searchClause>
+     *	    <searchClause>
+     *	      <index>au</index>
+     *	      <relation>
+     *	        <value>=</value>
+     *	      </relation>
+     *	      <term>Ritchie</term>
+     *	    </searchClause>
+     *	  </triple>
+     *	  <searchClause>
+     *	    <index>ti</index>
+     *	    <relation>
+     *	      <value>=</value>
+     *	    </relation>
+     *	    <term>Unix</term>
+     *	  </searchClause>
+     *	</triple>
+     * 
+ *

+ * @param -c + * Causes the output to be written in CQL rather than XCQL - that + * is, a query equivalent to that which was input, is output. In + * effect, the test harness acts as a query canonicaliser. + * @return + * The input query, either as XCQL [default] or CQL [if the + * -c option is supplied]. + */ public static void main (String[] args) { - if (args.length != 0) { - System.err.println("Usage: " + args[0]); + char mode = 'x'; // x=XCQL, c=CQL, p=PQF + String pfile = null; + + Vector argv = new Vector(); + for (int i = 0; i < args.length; i++) { + argv.add(args[i]); + } + + if (argv.size() > 0 && argv.get(0).equals("-d")) { + DEBUG = true; + argv.remove(0); + } + + if (argv.size() > 0 && argv.get(0).equals("-c")) { + mode = 'c'; + argv.remove(0); + } else if (argv.size() > 1 && argv.get(0).equals("-p")) { + mode = 'p'; + argv.remove(0); + pfile = (String) argv.get(0); + argv.remove(0); + } + + if (argv.size() > 1) { + System.err.println("Usage: CQLParser [-d] [-c] [-p []"); + System.err.println("If unspecified, query is read from stdin"); System.exit(1); } - byte[] bytes = new byte[10000]; - try { - // Read in the whole of standard input in one go - int nbytes = System.in.read(bytes); - } catch (java.io.IOException ex) { - System.err.println("Can't read query: " + ex.getMessage()); - System.exit(2); + String cql; + if (argv.size() == 1) { + cql = (String) argv.get(0); + } else { + byte[] bytes = new byte[10000]; + try { + // Read in the whole of standard input in one go + int nbytes = System.in.read(bytes); + } catch (IOException ex) { + System.err.println("Can't read query: " + ex.getMessage()); + System.exit(2); + } + cql = new String(bytes); } - String cql = new String(bytes); + CQLParser parser = new CQLParser(); - CQLNode root; + CQLNode root = null; try { root = parser.parse(cql); - debug("root='" + root + "'"); - System.out.println(root.toXCQL(0)); } catch (CQLParseException ex) { System.err.println("Syntax error: " + ex.getMessage()); System.exit(3); - } catch (java.io.IOException ex) { + } catch (IOException ex) { System.err.println("Can't compile query: " + ex.getMessage()); System.exit(4); } - } -} + try { + if (mode == 'c') { + System.out.println(root.toCQL()); + } else if (mode == 'p') { + InputStream f = new FileInputStream(pfile); + if (f == null) + throw new FileNotFoundException(pfile); -// This is a trivial subclass for java.io.StreamTokenizer which knows -// about the multi-character tokens "<=", ">=" and "<>", and included -// a render() method. Used only by CQLParser. -// -class CQLLexer extends StreamTokenizer { - private static boolean lexdebug; - - CQLLexer(String cql, boolean lexdebug) { - super(new StringReader(cql)); - this.ordinaryChar('='); - this.ordinaryChar('<'); - this.ordinaryChar('>'); - this.ordinaryChar('/'); - this.ordinaryChar('('); - this.ordinaryChar(')'); - this.wordChars('\'', '\''); // prevent this from introducing strings - this.lexdebug = lexdebug; - } - - public int nextToken() throws java.io.IOException { - int token = super.nextToken(); - if (lexdebug) - System.out.println("LEXDEBUG: " + - "token=" + token + ", " + - "nval=" + this.nval + ", " + - "sval=" + this.sval); - - return token; - } - - String render() { - return this.render(this.ttype, null, true); - } - - String render(boolean quoteChars) { - return this.render(this.ttype, null, quoteChars); - } - - String render(int token, String str, boolean quoteChars) { - String ret; - - if (token == this.TT_EOF) { - return "EOF"; - } else if (token == this.TT_EOL) { - return "EOL"; - } else if (token == this.TT_NUMBER) { - return "number: " + this.nval; - } else if (token == this.TT_WORD) { - return "word: \"" + this.sval + "\""; - } else if (token == '"') { - return "string: \"" + this.sval + "\""; + Properties config = new Properties(); + config.load(f); + f.close(); + System.out.println(root.toPQF(config)); + } else { + System.out.print(root.toXCQL(0)); + } + } catch (IOException ex) { + System.err.println("Can't render query: " + ex.getMessage()); + System.exit(5); + } catch (UnknownIndexException ex) { + System.err.println("Unknown index: " + ex.getMessage()); + System.exit(6); + } catch (UnknownRelationException ex) { + System.err.println("Unknown relation: " + ex.getMessage()); + System.exit(7); + } catch (UnknownRelationModifierException ex) { + System.err.println("Unknown relation modifier: " + + ex.getMessage()); + System.exit(8); + } catch (UnknownPositionException ex) { + System.err.println("Unknown position: " + ex.getMessage()); + System.exit(9); + } catch (PQFTranslationException ex) { + // We catch all of this class's subclasses, so -- + throw new Error("can't get a PQFTranslationException"); } - - String res = String.valueOf((char) token); - if (quoteChars) res = "'" + res + "'"; - return res; } }