+ * 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 qualifier, 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);
+ debug("in parseQuery()");
+
+ 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 = parseTerm(qualifier, relation);
term = new CQLAndNode(term, term2);
- } else if (lexer.sval.equals("or")) {
- match(lexer.TT_WORD);
- CQLNode term2 = parse_term(qualifier, relation);
+ } else if (lexer.ttype == lexer.TT_OR) {
+ match(lexer.TT_OR);
+ CQLNode term2 = parseTerm(qualifier, relation);
term = new CQLOrNode(term, term2);
- } else if (lexer.sval.equals("not")) {
- match(lexer.TT_WORD);
- CQLNode term2 = parse_term(qualifier, relation);
+ } else if (lexer.ttype == lexer.TT_NOT) {
+ match(lexer.TT_NOT);
+ CQLNode term2 = parseTerm(qualifier, relation);
term = new CQLNotNode(term, term2);
- } else if (lexer.sval.equals("prox")) {
- // ### Handle "prox"
+ } else if (lexer.ttype == lexer.TT_PROX) {
+ match(lexer.TT_PROX);
+ CQLProxNode proxnode = new CQLProxNode(term);
+ gatherProxParameters(proxnode);
+ CQLNode term2 = parseTerm(qualifier, relation);
+ proxnode.addSecondSubterm(term2);
+ term = (CQLNode) proxnode;
} else {
- throw new CQLParseException("unrecognised boolean: '" +
- lexer.sval + "'");
+ throw new CQLParseException("expected boolean, got " +
+ lexer.render());
}
}
@@ -75,43 +91,145 @@ public class CQLParser {
return term;
}
- private CQLNode parse_term(String qualifier, String 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 != '"') {
- throw new CQLParseException("expected qualifier or term, " +
- "got " + lexer.render());
+ } else if (lexer.ttype == '>') {
+ match('>');
+ return parsePrefix(qualifier, relation);
}
debug("non-parenthesised term");
- word = lexer.sval;
- match(lexer.ttype);
- if (!isRelation())
+ word = matchSymbol("qualifier or term");
+ if (!isBaseRelation())
break;
qualifier = word;
- relation = lexer.render(false);
- debug("got relation '" + relation + "'");
+ relation = new CQLRelation(lexer.render(lexer.ttype, false));
match(lexer.ttype);
- debug("qualifier='" + qualifier + ", relation='" + relation + "'");
+
+ while (lexer.ttype == '/') {
+ match('/');
+ if (lexer.ttype != lexer.TT_RELEVANT &&
+ lexer.ttype != lexer.TT_FUZZY &&
+ lexer.ttype != lexer.TT_STEM &&
+ lexer.ttype != lexer.TT_PHONETIC)
+ throw new CQLParseException("expected relation modifier, "
+ + "got " + lexer.render());
+ relation.addModifier(lexer.sval.toLowerCase());
+ match(lexer.ttype);
+ }
+
+ debug("qualifier='" + qualifier + ", " +
+ "relation='" + relation.toCQL() + "'");
}
CQLTermNode node = new CQLTermNode(qualifier, relation, word);
- debug("made term node " + node);
+ debug("made term node " + node.toCQL());
return node;
}
- boolean isRelation() {
- // ### Handle any, all and exact
+ 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++) {
+ 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 0: gatherProxRelation(node); break;
+ case 1: gatherProxDistance(node); break;
+ case 2: gatherProxUnit(node); break;
+ case 3: gatherProxOrdering(node); break;
+ }
+ }
+ }
+ }
+
+ 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));
+ }
+
+ 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);
+ }
+
+ 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);
+ }
+
+ // 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 == '<' ||
lexer.ttype == '>' ||
lexer.ttype == '=' ||
@@ -122,166 +240,200 @@ public class CQLParser {
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
+ // 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());
}
- // 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) {
- 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("-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 [-c] [-p