X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;f=src%2Forg%2Fz3950%2Fzing%2Fcql%2FCQLParser.java;h=9ccb1398a89764fe857836332a6370a9654f2dd2;hb=0cf2302a18eacaa327f40743c4ac59cf6c630183;hp=96e67f018dd13ac2faed4481ce1f4976d08c063b;hpb=c77bd4c9ead8ea2dc08e1fadd801a5d01db84a9a;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 96e67f0..9ccb139 100644 --- a/src/org/z3950/zing/cql/CQLParser.java +++ b/src/org/z3950/zing/cql/CQLParser.java @@ -1,67 +1,142 @@ -// $Id: CQLParser.java,v 1.13 2002-11-02 01:24:14 mike Exp $ +// $Id: CQLParser.java,v 1.38 2007-07-03 16:40:41 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.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 1.13 2002-11-02 01:24:14 mike Exp $ + * @version $Id: CQLParser.java,v 1.38 2007-07-03 16:40:41 mike Exp $ * @see http://zing.z3950.org/cql/index.html */ public class CQLParser { private CQLLexer lexer; + private int compat; // When false, implement CQL 1.2 + public static int V1POINT1 = 12368; + public static int V1POINT2 = 12369; + public static int V1POINT1SORT = 12370; + static private boolean DEBUG = false; static private boolean LEXDEBUG = false; + /** + * The new parser implements a dialect of CQL specified by the + * compat argument: + * + */ + CQLParser(int compat) { + this.compat = compat; + } + + /** + * The new parser implements CQL 1.2 + */ + CQLParser() { + this.compat = V1POINT2; + } + 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", new CQLRelation("=")); + debug("about to parseQuery()"); + CQLNode root = parseTopLevelPrefixes("cql.serverChoice", + new CQLRelation(compat == V1POINT2 ? "=" : "scr")); if (lexer.ttype != lexer.TT_EOF) throw new CQLParseException("junk after end: " + lexer.render()); return root; } - private CQLNode parse_query(String qualifier, CQLRelation relation) + private CQLNode parseTopLevelPrefixes(String index, CQLRelation relation) throws CQLParseException, IOException { - debug("in parse_query()"); + debug("top-level prefix mapping"); + + if (lexer.ttype == '>') { + return parsePrefix(index, relation, true); + } + + CQLNode node = parseQuery(index, relation); + if ((compat == V1POINT2 || compat == V1POINT1SORT) && + lexer.ttype == lexer.TT_SORTBY) { + match(lexer.ttype); + debug("sortspec"); + + CQLSortNode sortnode = new CQLSortNode(node); + while (lexer.ttype != lexer.TT_EOF) { + String sortindex = matchSymbol("sort index"); + ModifierSet ms = gatherModifiers(sortindex); + sortnode.addSortIndex(ms); + } - CQLNode term = parse_term(qualifier, relation); + if (sortnode.keys.size() == 0) { + throw new CQLParseException("no sort keys"); + } + + node = sortnode; + } + + return node; + } + + private CQLNode parseQuery(String index, CQLRelation relation) + throws CQLParseException, IOException { + debug("in parseQuery()"); + + 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 = parse_term(qualifier, relation); - term = new CQLAndNode(term, term2); - } else if (lexer.ttype == lexer.TT_OR) { - match(lexer.TT_OR); - CQLNode term2 = parse_term(qualifier, relation); - term = new CQLOrNode(term, term2); - } else if (lexer.ttype == lexer.TT_NOT) { - match(lexer.TT_NOT); - CQLNode term2 = parse_term(qualifier, 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 = parse_term(qualifier, relation); - proxnode.addSecondSubterm(term2); - term = (CQLNode) proxnode; + lexer.ttype != ')' && + lexer.ttype != lexer.TT_SORTBY) { + 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("expected boolean, got " + lexer.render()); @@ -72,140 +147,101 @@ public class CQLParser { return term; } - private CQLNode parse_term(String qualifier, CQLRelation relation) + private ModifierSet gatherModifiers(String base) throws CQLParseException, IOException { - debug("in parse_term()"); + 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 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 != lexer.TT_NUMBER && - lexer.ttype != '"') { - throw new CQLParseException("expected qualifier or term, " + - "got " + lexer.render()); + } else if (lexer.ttype == '>') { + return parsePrefix(index, relation, false); } debug("non-parenthesised term"); - if (lexer.ttype == lexer.TT_NUMBER) { - word = lexer.render(); - } else { - word = lexer.sval; - } - match(lexer.ttype); - 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; + String relstr = (lexer.ttype == lexer.TT_WORD ? + lexer.sval : lexer.render(lexer.ttype, false)); + relation = new CQLRelation(relstr); match(lexer.ttype); - - while (lexer.ttype == '/') { - match('/'); - if (lexer.ttype != lexer.TT_RELEVANT && - lexer.ttype != lexer.TT_FUZZY && - lexer.ttype != lexer.TT_STEM) - throw new CQLParseException("expected relation modifier, " - + "got " + lexer.render()); - relation.addModifier(lexer.sval); - match(lexer.ttype); - } - - debug("qualifier='" + qualifier + ", " + + ModifierSet ms = gatherModifiers(relstr); + relation.setModifiers(ms); + 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 void gatherProxParameters(CQLProxNode node) + private CQLNode parsePrefix(String index, CQLRelation relation, + boolean topLevel) throws CQLParseException, IOException { - for (int i = 0; i < 4; i++) { - if (lexer.ttype != '/') - return; // end of proximity parameters - - match('/'); - if (lexer.ttype != '/') { - // not an omitted default - switch (i) { - // Order should be: relation/distance/unit/ordering - // For now, use MA's: unit/relation/distance/ordering - case 1: gatherProxRelation(node); break; - case 2: gatherProxDistance(node); break; - case 0: gatherProxUnit(node); break; - case 3: gatherProxOrdering(node); break; - } - } + debug("prefix mapping"); + + match('>'); + String name = null; + String identifier = matchSymbol("prefix-name"); + if (lexer.ttype == '=') { + match('='); + name = identifier; + identifier = matchSymbol("prefix-identifer"); } - } - - private void gatherProxRelation(CQLProxNode node) - throws CQLParseException, IOException { - if (!isProxRelation()) - throw new CQLParseException("expected proximity relation, got " + - lexer.render()); - node.addModifier("relation", lexer.render(lexer.ttype, false)); - match(lexer.ttype); - debug("gPR matched " + lexer.render(lexer.ttype, false)); - } + CQLNode node = topLevel ? + parseTopLevelPrefixes(index, relation) : + parseQuery(index, relation); - private void gatherProxDistance(CQLProxNode node) - throws CQLParseException, IOException { - if (lexer.ttype != lexer.TT_NUMBER) - throw new CQLParseException("expected proximity distance, got " + - lexer.render()); - node.addModifier("distance", lexer.render(lexer.ttype, false)); - match(lexer.ttype); - debug("gPD matched " + lexer.render(lexer.ttype, false)); - } - - private void gatherProxUnit(CQLProxNode node) - throws CQLParseException, IOException { - if (lexer.ttype != lexer.TT_pWORD && - lexer.ttype != lexer.TT_SENTENCE && - lexer.ttype != lexer.TT_PARAGRAPH && - lexer.ttype != lexer.TT_ELEMENT) - throw new CQLParseException("expected proximity unit, got " + - lexer.render()); - node.addModifier("unit", lexer.render()); - match(lexer.ttype); - } - - private void gatherProxOrdering(CQLProxNode node) - throws CQLParseException, IOException { - if (lexer.ttype != lexer.TT_ORDERED && - lexer.ttype != lexer.TT_UNORDERED) - throw new CQLParseException("expected proximity ordering, got " + - lexer.render()); - node.addModifier("ordering", lexer.render()); - match(lexer.ttype); + return new CQLPrefixNode(name, identifier, node); } - 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); - } - - boolean isProxRelation() { - debug("isProxRelation: checking ttype=" + lexer.ttype + + // 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.TT_LE || lexer.ttype == lexer.TT_GE || - lexer.ttype == lexer.TT_NE); + lexer.ttype == lexer.TT_NE || + lexer.ttype == lexer.TT_EQEQ); } private void match(int token) @@ -221,48 +257,126 @@ public class CQLParser { " (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_SORTBY) { + 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 -1 + * CQL version 1.1 (default version 1.2) + * @param -d + * Debug mode: extra output written to stderr. + * @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) { - boolean canonicalise = false; - Vector argv = new Vector(); + 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]); } + int compat = V1POINT2; + if (argv.size() > 0 && argv.get(0).equals("-1")) { + compat = V1POINT1; + argv.remove(0); + } + + if (argv.size() > 0 && argv.get(0).equals("-d")) { + DEBUG = true; + argv.remove(0); + } + if (argv.size() > 0 && argv.get(0).equals("-c")) { - canonicalise = true; + 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 [-c] []"); + System.err.println("Usage: CQLParser [-1] [-d] [-c] " + + "[-p []"); System.err.println("If unspecified, query is read from stdin"); System.exit(1); } @@ -275,29 +389,59 @@ public class CQLParser { try { // Read in the whole of standard input in one go int nbytes = System.in.read(bytes); - } catch (java.io.IOException ex) { + } catch (IOException ex) { System.err.println("Can't read query: " + ex.getMessage()); System.exit(2); } cql = new String(bytes); } - CQLParser parser = new CQLParser(); - CQLNode root; + CQLParser parser = new CQLParser(compat); + CQLNode root = null; try { root = parser.parse(cql); - debug("root='" + root + "'"); - if (canonicalise) { - System.out.println(root.toCQL()); - } else { - System.out.print(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); + + 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"); + } } }