+ * 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 {
+ return parse(new StringReader(cql));
+ }
/**
* Compiles a CQL query.
@@ -71,16 +116,18 @@ public class CQLParser {
* @param cql The query
* @return A CQLNode object which is the root of a parse
* tree representing the query. */
- public CQLNode parse(String cql)
+ public CQLNode parse(Reader cql)
throws CQLParseException, IOException {
- lexer = new CQLLexer(cql, LEXDEBUG);
+ par = new PositionAwareReader(cql);
+ lexer = new CQLLexer(par, LEXDEBUG);
lexer.nextToken();
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());
+ if (lexer.ttype != CQLLexer.TT_EOF)
+ throw new CQLParseException("junk after end: " + lexer.render(),
+ par.getPosition());
return root;
}
@@ -95,19 +142,19 @@ public class CQLParser {
CQLNode node = parseQuery(index, relation);
if ((compat == V1POINT2 || compat == V1POINT1SORT) &&
- lexer.ttype == lexer.TT_SORTBY) {
+ lexer.ttype == CQLLexer.TT_SORTBY) {
match(lexer.ttype);
debug("sortspec");
CQLSortNode sortnode = new CQLSortNode(node);
- while (lexer.ttype != lexer.TT_EOF) {
+ while (lexer.ttype != CQLLexer.TT_EOF) {
String sortindex = matchSymbol("sort index");
ModifierSet ms = gatherModifiers(sortindex);
sortnode.addSortIndex(ms);
}
if (sortnode.keys.size() == 0) {
- throw new CQLParseException("no sort keys");
+ throw new CQLParseException("no sort keys", par.getPosition());
}
node = sortnode;
@@ -121,25 +168,25 @@ public class CQLParser {
debug("in parseQuery()");
CQLNode term = parseTerm(index, relation);
- while (lexer.ttype != lexer.TT_EOF &&
+ while (lexer.ttype != CQLLexer.TT_EOF &&
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) {
+ lexer.ttype != CQLLexer.TT_SORTBY) {
+ if (lexer.ttype == CQLLexer.TT_AND ||
+ lexer.ttype == CQLLexer.TT_OR ||
+ lexer.ttype == CQLLexer.TT_NOT ||
+ lexer.ttype == CQLLexer.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) :
+ term = ((type == CQLLexer.TT_AND) ? new CQLAndNode(term, term2, ms) :
+ (type == CQLLexer.TT_OR) ? new CQLOrNode (term, term2, ms) :
+ (type == CQLLexer.TT_NOT) ? new CQLNotNode(term, term2, ms) :
new CQLProxNode(term, term2, ms));
} else {
throw new CQLParseException("expected boolean, got " +
- lexer.render());
+ lexer.render(), par.getPosition());
}
}
@@ -154,12 +201,13 @@ public class CQLParser {
ModifierSet ms = new ModifierSet(base);
while (lexer.ttype == '/') {
match('/');
- if (lexer.ttype != lexer.TT_WORD)
+ if (lexer.ttype != CQLLexer.TT_WORD)
throw new CQLParseException("expected modifier, "
- + "got " + lexer.render());
+ + "got " + lexer.render(),
+ par.getPosition());
String type = lexer.sval.toLowerCase();
match(lexer.ttype);
- if (!isRelation()) {
+ if (!isSymbolicRelation()) {
// It's a simple modifier consisting of type only
ms.addModifier(type);
} else {
@@ -192,16 +240,21 @@ public class CQLParser {
debug("non-parenthesised term");
word = matchSymbol("index or term");
- if (!isRelation() && lexer.ttype != lexer.TT_WORD)
+ while (lexer.ttype == CQLLexer.TT_WORD && !isRelation()) {
+ word = word + " " + lexer.sval;
+ match(CQLLexer.TT_WORD);
+ }
+
+ if (!isRelation())
break;
index = word;
- String relstr = (lexer.ttype == lexer.TT_WORD ?
+ String relstr = (lexer.ttype == CQLLexer.TT_WORD ?
lexer.sval : lexer.render(lexer.ttype, false));
relation = new CQLRelation(relstr);
match(lexer.ttype);
ModifierSet ms = gatherModifiers(relstr);
- relation.setModifiers(ms);
+ relation.ms = ms;
debug("index='" + index + ", " +
"relation='" + relation.toCQL() + "'");
}
@@ -231,17 +284,34 @@ public class CQLParser {
return new CQLPrefixNode(name, identifier, node);
}
- // Checks for a relation
private boolean isRelation() {
debug("isRelation: checking ttype=" + lexer.ttype +
" (" + lexer.render() + ")");
+ if (lexer.ttype == CQLLexer.TT_WORD &&
+ (lexer.sval.indexOf('.') >= 0 ||
+ lexer.sval.equals("any") ||
+ lexer.sval.equals("all") ||
+ lexer.sval.equals("within") ||
+ lexer.sval.equals("encloses") ||
+ (lexer.sval.equals("exact") && compat != V1POINT2) ||
+ (lexer.sval.equals("scr") && compat != V1POINT2) ||
+ (lexer.sval.equals("adj") && compat == V1POINT2) ||
+ customRelations.contains(lexer.sval)))
+ return true;
+
+ return isSymbolicRelation();
+ }
+
+ private boolean isSymbolicRelation() {
+ debug("isSymbolicRelation: 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_EQEQ);
+ lexer.ttype == CQLLexer.TT_LE ||
+ lexer.ttype == CQLLexer.TT_GE ||
+ lexer.ttype == CQLLexer.TT_NE ||
+ lexer.ttype == CQLLexer.TT_EQEQ);
}
private void match(int token)
@@ -250,7 +320,8 @@ public class CQLParser {
if (lexer.ttype != token)
throw new CQLParseException("expected " +
lexer.render(token, true) +
- ", " + "got " + lexer.render());
+ ", " + "got " + lexer.render(),
+ par.getPosition());
int tmp = lexer.nextToken();
debug("match() got token=" + lexer.ttype + ", " +
"nval=" + lexer.nval + ", sval='" + lexer.sval + "'" +
@@ -261,27 +332,27 @@ public class CQLParser {
throws CQLParseException, IOException {
debug("in matchSymbol()");
- if (lexer.ttype == lexer.TT_WORD ||
- lexer.ttype == lexer.TT_NUMBER ||
+ if (lexer.ttype == CQLLexer.TT_WORD ||
+ lexer.ttype == CQLLexer.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.ttype == CQLLexer.TT_AND ||
+ lexer.ttype == CQLLexer.TT_OR ||
+ lexer.ttype == CQLLexer.TT_NOT ||
+ lexer.ttype == CQLLexer.TT_PROX ||
+ lexer.ttype == CQLLexer.TT_SORTBY) {
+ String symbol = (lexer.ttype == CQLLexer.TT_NUMBER) ?
lexer.render() : lexer.sval;
match(lexer.ttype);
return symbol;
}
throw new CQLParseException("expected " + expected + ", " +
- "got " + lexer.render());
+ "got " + lexer.render(), par.getPosition());
}
@@ -348,7 +419,7 @@ public class CQLParser {
char mode = 'x'; // x=XCQL, c=CQL, p=PQF
String pfile = null;
- Vector