X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;f=src%2Forg%2Fz3950%2Fzing%2Fcql%2FCQLParser.java;h=94e8b87190a6783998ad77363ba67c9f442e7d54;hb=b53fe47e0bd5f0825b0d8ceeb33538e714bf9e7e;hp=eadedefd8571df176225017f3d5ae0abbd2823fa;hpb=64483d3d31c182adfb96fd75f8be10ff9f374d34;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 eadedef..94e8b87 100644 --- a/src/org/z3950/zing/cql/CQLParser.java +++ b/src/org/z3950/zing/cql/CQLParser.java @@ -1,4 +1,4 @@ -// $Id: CQLParser.java,v 1.20 2002-11-14 22:04:16 mike Exp $ +// $Id: CQLParser.java,v 1.30 2007-06-28 00:24:48 mike Exp $ package org.z3950.zing.cql; import java.io.IOException; @@ -12,7 +12,7 @@ import java.io.FileNotFoundException; /** * Compiles CQL strings into parse trees of CQLNode subtypes. * - * @version $Id: CQLParser.java,v 1.20 2002-11-14 22:04:16 mike Exp $ + * @version $Id: CQLParser.java,v 1.30 2007-06-28 00:24:48 mike Exp $ * @see http://zing.z3950.org/cql/index.html */ @@ -34,8 +34,10 @@ public class CQLParser { * 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; and toXCQL() returns an - * XML snippet representing the query. + * 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 @@ -53,30 +55,30 @@ public class CQLParser { return root; } - private CQLNode parseQuery(String qualifier, CQLRelation relation) + private CQLNode parseQuery(String index, CQLRelation relation) throws CQLParseException, IOException { debug("in parseQuery()"); - CQLNode term = parseTerm(qualifier, relation); + CQLNode term = parseTerm(index, relation); while (lexer.ttype != lexer.TT_EOF && lexer.ttype != ')') { if (lexer.ttype == lexer.TT_AND) { match(lexer.TT_AND); - CQLNode term2 = parseTerm(qualifier, relation); + CQLNode term2 = parseTerm(index, relation); term = new CQLAndNode(term, term2); } else if (lexer.ttype == lexer.TT_OR) { match(lexer.TT_OR); - CQLNode term2 = parseTerm(qualifier, relation); + CQLNode term2 = parseTerm(index, relation); term = new CQLOrNode(term, term2); } else if (lexer.ttype == lexer.TT_NOT) { match(lexer.TT_NOT); - CQLNode term2 = parseTerm(qualifier, relation); + CQLNode term2 = parseTerm(index, relation); term = new CQLNotNode(term, term2); } else if (lexer.ttype == lexer.TT_PROX) { match(lexer.TT_PROX); CQLProxNode proxnode = new CQLProxNode(term); gatherProxParameters(proxnode); - CQLNode term2 = parseTerm(qualifier, relation); + CQLNode term2 = parseTerm(index, relation); proxnode.addSecondSubterm(term2); term = (CQLNode) proxnode; } else { @@ -89,7 +91,7 @@ public class CQLParser { return term; } - private CQLNode parseTerm(String qualifier, CQLRelation relation) + private CQLNode parseTerm(String index, CQLRelation relation) throws CQLParseException, IOException { debug("in parseTerm()"); @@ -98,44 +100,61 @@ public class CQLParser { if (lexer.ttype == '(') { debug("parenthesised term"); match('('); - CQLNode expr = parseQuery(qualifier, relation); + CQLNode expr = parseQuery(index, relation); match(')'); return expr; } else if (lexer.ttype == '>') { match('>'); - return parsePrefix(qualifier, relation); + return parsePrefix(index, relation); } debug("non-parenthesised term"); - word = matchSymbol("qualifier or term"); - if (!isBaseRelation()) + word = matchSymbol("index or term"); + if (!isRelation() && lexer.ttype != lexer.TT_WORD) break; - qualifier = word; - relation = new CQLRelation(lexer.render(lexer.ttype, false)); + index = word; + relation = new CQLRelation(lexer.ttype == lexer.TT_WORD ? + lexer.sval : + lexer.render(lexer.ttype, false)); match(lexer.ttype); while (lexer.ttype == '/') { match('/'); - if (lexer.ttype != lexer.TT_RELEVANT && - lexer.ttype != lexer.TT_FUZZY && - lexer.ttype != lexer.TT_STEM) + if (lexer.ttype != lexer.TT_WORD) throw new CQLParseException("expected relation modifier, " + "got " + lexer.render()); - relation.addModifier(lexer.sval.toLowerCase()); + String type = lexer.sval.toLowerCase(); match(lexer.ttype); + if (!isRelation()) { + // It's a simple modifier consisting of type only + relation.addModifier(type); + } else { + // It's a complex modifier of the form type=value + String comparision = lexer.render(lexer.ttype, false); + match(lexer.ttype); + + // Yuck + String value = lexer.ttype == lexer.TT_WORD ? lexer.sval : + (double) lexer.nval == (int) lexer.nval ? + new Integer((int) lexer.nval).toString() : + new Double((double) lexer.nval).toString(); + + matchSymbol("relation-modifier value"); + relation.addModifier(type, comparision, value); + } } - debug("qualifier='" + qualifier + ", " + + debug("index='" + index + ", " + "relation='" + relation.toCQL() + "'"); } - CQLTermNode node = new CQLTermNode(qualifier, relation, word); + CQLTermNode node = new CQLTermNode(index, relation, word); debug("made term node " + node.toCQL()); return node; } - private CQLNode parsePrefix(String qualifier, CQLRelation relation) + private CQLNode parsePrefix(String index, CQLRelation relation) throws CQLParseException, IOException { debug("prefix mapping"); @@ -146,7 +165,7 @@ public class CQLParser { name = identifier; identifier = matchSymbol("prefix-identifer"); } - CQLNode term = parseTerm(qualifier, relation); + CQLNode term = parseQuery(index, relation); return new CQLPrefixNode(name, identifier, term); } @@ -173,10 +192,10 @@ public class CQLParser { private void gatherProxRelation(CQLProxNode node) throws CQLParseException, IOException { - if (!isProxRelation()) + if (!isRelation()) throw new CQLParseException("expected proximity relation, got " + lexer.render()); - node.addModifier("relation", lexer.render(lexer.ttype, false)); + node.addModifier("relation", null, lexer.render(lexer.ttype, false)); match(lexer.ttype); debug("gPR matched " + lexer.render(lexer.ttype, false)); } @@ -186,7 +205,7 @@ public class CQLParser { if (lexer.ttype != lexer.TT_NUMBER) throw new CQLParseException("expected proximity distance, got " + lexer.render()); - node.addModifier("distance", lexer.render(lexer.ttype, false)); + node.addModifier("distance", null, lexer.render(lexer.ttype, false)); match(lexer.ttype); debug("gPD matched " + lexer.render(lexer.ttype, false)); } @@ -199,7 +218,7 @@ public class CQLParser { lexer.ttype != lexer.TT_ELEMENT) throw new CQLParseException("expected proximity unit, got " + lexer.render()); - node.addModifier("unit", lexer.render()); + node.addModifier("unit", null, lexer.render()); match(lexer.ttype); } @@ -209,22 +228,13 @@ public class CQLParser { lexer.ttype != lexer.TT_UNORDERED) throw new CQLParseException("expected proximity ordering, got " + lexer.render()); - node.addModifier("ordering", lexer.render()); + node.addModifier("ordering", null, lexer.render()); match(lexer.ttype); } - private boolean isBaseRelation() { - debug("isBaseRelation: checking ttype=" + lexer.ttype + - " (" + lexer.render() + ")"); - return (isProxRelation() || - lexer.ttype == lexer.TT_ANY || - lexer.ttype == lexer.TT_ALL || - lexer.ttype == lexer.TT_EXACT || - lexer.ttype == lexer.TT_SCR); - } - - private boolean isProxRelation() { - debug("isProxRelation: checking ttype=" + lexer.ttype + + // Checks for a relation that may be used inside a prox operator + private boolean isRelation() { + debug("isRelation: checking ttype=" + lexer.ttype + " (" + lexer.render() + ")"); return (lexer.ttype == '<' || lexer.ttype == '>' || @@ -256,24 +266,19 @@ public class CQLParser { lexer.ttype == '"' || // The following is a complete list of keywords. Because // they're listed here, they can be used unquoted as - // qualifiers, terms, prefix names and prefix identifiers. + // 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_ANY || - lexer.ttype == lexer.TT_ALL || - lexer.ttype == lexer.TT_EXACT || 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 || - lexer.ttype == lexer.TT_RELEVANT || - lexer.ttype == lexer.TT_FUZZY || - lexer.ttype == lexer.TT_STEM || - lexer.ttype == lexer.TT_SCR) { + lexer.ttype == lexer.TT_UNORDERED) { String symbol = (lexer.ttype == lexer.TT_NUMBER) ? lexer.render() : lexer.sval; match(lexer.ttype); @@ -344,11 +349,16 @@ public class CQLParser { char mode = 'x'; // x=XCQL, c=CQL, p=PQF String pfile = null; - Vector argv = new Vector(); + 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); @@ -360,8 +370,7 @@ public class CQLParser { } if (argv.size() > 1) { - System.err.println( - "Usage: CQLParser [-c] [-p []"); + System.err.println("Usage: CQLParser [-d] [-c] [-p []"); System.err.println("If unspecified, query is read from stdin"); System.exit(1); } @@ -411,8 +420,8 @@ public class CQLParser { } catch (IOException ex) { System.err.println("Can't render query: " + ex.getMessage()); System.exit(5); - } catch (UnknownQualifierException ex) { - System.err.println("Unknown qualifier: " + ex.getMessage()); + } catch (UnknownIndexException ex) { + System.err.println("Unknown index: " + ex.getMessage()); System.exit(6); } catch (UnknownRelationException ex) { System.err.println("Unknown relation: " + ex.getMessage());