X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;f=src%2Forg%2Fz3950%2Fzing%2Fcql%2FCQLParser.java;h=85ee63ba22abc249985f7dab27f939e5f79b8ed2;hb=08b7134bda92b6a33c57c7c6c333dacf1461c7c7;hp=63291469fae8f2f87a1e83528a824af15978fea5;hpb=77e96a476683020357b2c6e6fa661024698a20d8;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 6329146..85ee63b 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 1.19 2002-11-08 16:38:47 mike Exp $
+// $Id: CQLParser.java,v 1.22 2002-11-20 01:15:15 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 1.19 2002-11-08 16:38:47 mike Exp $
+ * @version $Id: CQLParser.java,v 1.22 2002-11-20 01:15:15 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
@@ -45,39 +47,38 @@ public class CQLParser {
lexer = new CQLLexer(cql, LEXDEBUG);
lexer.nextToken();
- debug("about to parse_query()");
- CQLNode root = parse_query("srw.serverChoice", new CQLRelation("scr"));
- // ### "scr" above should arguably be "="
+ 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, CQLRelation relation)
+ private CQLNode parseQuery(String qualifier, CQLRelation relation)
throws CQLParseException, IOException {
- debug("in parse_query()");
+ debug("in parseQuery()");
- CQLNode term = parse_term(qualifier, relation);
+ CQLNode term = parseTerm(qualifier, 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);
+ CQLNode term2 = parseTerm(qualifier, relation);
term = new CQLAndNode(term, term2);
} else if (lexer.ttype == lexer.TT_OR) {
match(lexer.TT_OR);
- CQLNode term2 = parse_term(qualifier, relation);
+ CQLNode term2 = parseTerm(qualifier, relation);
term = new CQLOrNode(term, term2);
} else if (lexer.ttype == lexer.TT_NOT) {
match(lexer.TT_NOT);
- CQLNode term2 = parse_term(qualifier, relation);
+ CQLNode term2 = parseTerm(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);
+ CQLNode term2 = parseTerm(qualifier, relation);
proxnode.addSecondSubterm(term2);
term = (CQLNode) proxnode;
} else {
@@ -90,32 +91,25 @@ public class CQLParser {
return term;
}
- private CQLNode parse_term(String qualifier, CQLRelation relation)
+ private CQLNode parseTerm(String qualifier, 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(qualifier, 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 == '>') {
+ match('>');
+ return parsePrefix(qualifier, relation);
}
debug("non-parenthesised term");
- if (lexer.ttype == lexer.TT_NUMBER) {
- word = lexer.render();
- } else {
- word = lexer.sval;
- }
- match(lexer.ttype);
+ word = matchSymbol("qualifier or term");
if (!isBaseRelation())
break;
@@ -127,7 +121,8 @@ public class CQLParser {
match('/');
if (lexer.ttype != lexer.TT_RELEVANT &&
lexer.ttype != lexer.TT_FUZZY &&
- lexer.ttype != lexer.TT_STEM)
+ lexer.ttype != lexer.TT_STEM &&
+ lexer.ttype != lexer.TT_PHONETIC)
throw new CQLParseException("expected relation modifier, "
+ "got " + lexer.render());
relation.addModifier(lexer.sval.toLowerCase());
@@ -143,6 +138,21 @@ public class CQLParser {
return node;
}
+ private CQLNode parsePrefix(String qualifier, 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(qualifier, relation);
+ return new CQLPrefixNode(name, identifier, term);
+ }
+
private void gatherProxParameters(CQLProxNode node)
throws CQLParseException, IOException {
for (int i = 0; i < 4; i++) {
@@ -212,9 +222,11 @@ public class CQLParser {
return (isProxRelation() ||
lexer.ttype == lexer.TT_ANY ||
lexer.ttype == lexer.TT_ALL ||
- lexer.ttype == lexer.TT_EXACT);
+ lexer.ttype == lexer.TT_EXACT ||
+ lexer.ttype == lexer.TT_SCR);
}
+ // Checks for a relation that may be used inside a prox operator
private boolean isProxRelation() {
debug("isProxRelation: checking ttype=" + lexer.ttype +
" (" + lexer.render() + ")");
@@ -239,6 +251,46 @@ 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
+ // qualifiers, 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_PHONETIC) {
+ String symbol = (lexer.ttype == lexer.TT_NUMBER) ?
+ lexer.render() : lexer.sval;
+ match(lexer.ttype);
+ return symbol;
+ }
+
+ throw new CQLParseException("expected " + expected + ", " +
+ "got " + lexer.render());
+ }
+
/**
* Simple test-harness for the CQLParser class.