+ * 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 = 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 index, CQLRelation relation)
throws CQLParseException, IOException {
- debug("in parse_query()");
+ debug("in parseQuery()");
- CQLNode term = parse_term(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 = parse_term(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 = parse_term(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 = parse_term(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 = parse_term(qualifier, relation);
+ CQLNode term2 = parseTerm(index, relation);
proxnode.addSecondSubterm(term2);
term = (CQLNode) proxnode;
} else {
@@ -72,59 +91,77 @@ public class CQLParser {
return term;
}
- private CQLNode parse_term(String qualifier, CQLRelation relation)
+ 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 != lexer.TT_NUMBER &&
- 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");
- if (lexer.ttype == lexer.TT_NUMBER) {
- word = lexer.render();
- } else {
- word = lexer.sval;
- }
- match(lexer.ttype);
+ word = matchSymbol("index or term");
if (!isBaseRelation())
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)
+ lexer.ttype != lexer.TT_STEM &&
+ lexer.ttype != lexer.TT_PHONETIC &&
+ lexer.ttype != lexer.TT_WORD)
throw new CQLParseException("expected relation modifier, "
+ "got " + lexer.render());
- relation.addModifier(lexer.sval);
+ if (lexer.ttype == lexer.TT_WORD &&
+ lexer.sval.indexOf('.') == -1)
+ throw new CQLParseException("unknown first-class " +
+ "relation modifier: " +
+ lexer.sval);
+
+ relation.addModifier(lexer.sval.toLowerCase());
match(lexer.ttype);
}
- 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 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);
+ }
+
private void gatherProxParameters(CQLProxNode node)
throws CQLParseException, IOException {
for (int i = 0; i < 4; i++) {
@@ -137,9 +174,9 @@ public class CQLParser {
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 0: gatherProxRelation(node); break;
+ case 1: gatherProxDistance(node); break;
+ case 2: gatherProxUnit(node); break;
case 3: gatherProxOrdering(node); break;
}
}
@@ -151,7 +188,7 @@ public class CQLParser {
if (!isProxRelation())
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));
}
@@ -161,7 +198,7 @@ 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));
}
@@ -174,7 +211,7 @@ 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);
}
@@ -184,20 +221,30 @@ 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);
}
- boolean isBaseRelation() {
+ private boolean isBaseRelation()
+ throws CQLParseException {
debug("isBaseRelation: checking ttype=" + lexer.ttype +
" (" + lexer.render() + ")");
+
+ if (lexer.ttype == lexer.TT_WORD &&
+ lexer.sval.indexOf('.') == -1)
+ throw new CQLParseException("unknown first-class relation: " +
+ lexer.sval);
+
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 ||
+ lexer.ttype == lexer.TT_WORD);
}
- boolean isProxRelation() {
+ // Checks for a relation that may be used inside a prox operator
+ private boolean isProxRelation() {
debug("isProxRelation: checking ttype=" + lexer.ttype +
" (" + lexer.render() + ")");
return (lexer.ttype == '<' ||
@@ -221,48 +268,128 @@ 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_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());
+ }
+
- // Test harness.
- //
- // e.g. echo '(au=Kerninghan or au=Ritchie) and ti=Unix' |
- // java org.z3950.zing.cql.CQLParser
- // yields:
- //
+ * 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) {
- boolean canonicalise = false;
- Vector argv = new Vector();
+ char mode = 'x'; // x=XCQL, c=CQL, p=PQF
+ String pfile = null;
+
+ Vector