X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;f=src%2Forg%2Fz3950%2Fzing%2Fcql%2FCQLParser.java;h=9ccb1398a89764fe857836332a6370a9654f2dd2;hb=0cf2302a18eacaa327f40743c4ac59cf6c630183;hp=4c507b680e53115405631c5d137164d9adb694a9;hpb=729cbe638bca2960fcb93a581026e0e4b9432977;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 4c507b6..9ccb139 100644 --- a/src/org/z3950/zing/cql/CQLParser.java +++ b/src/org/z3950/zing/cql/CQLParser.java @@ -1,207 +1,447 @@ -// $Header: /home/cvsroot/cql-java/src/org/z3950/zing/cql/CQLParser.java,v 1.2 2002-10-24 16:05:15 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; -import java.io.IOException; -import java.io.StringReader; -import java.io.StreamTokenizer; /** - * Compiles a CQL string into a parse tree ... - * ### + * Compiles CQL strings into parse trees of CQLNode subtypes. * - * @version $Id: CQLParser.java,v 1.2 2002-10-24 16:05:15 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 */ -class CQLCompiler { - private String cql; - private String qualset; - private Properties qualsetProperties; - private StreamTokenizer st; - - private class CQLParseException extends Exception { - CQLParseException(String s) { super(s); } +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: + *
+ * 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 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 parseTopLevelPrefixes(String index, CQLRelation relation) + throws CQLParseException, IOException { + debug("top-level prefix mapping"); + + if (lexer.ttype == '>') { + return parsePrefix(index, relation, true); } - if (st.ttype != st.TT_EOF) { - System.err.println("### Extra bits: " + render(st)); - return null; + 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); + } + + if (sortnode.keys.size() == 0) { + throw new CQLParseException("no sort keys"); + } + + node = sortnode; } - // Interpret attributes as BIB-1 unless otherwise specified - return "@attrset bib-1 " + ret; + return node; } - private String parse_expression() + private CQLNode parseQuery(String index, CQLRelation relation) throws CQLParseException, IOException { - String term = parse_term(); + debug("in parseQuery()"); - while (st.ttype == st.TT_WORD) { - String op = st.sval.toLowerCase(); - if (!st.sval.equals("and") && - !st.sval.equals("or") && - !st.sval.equals("not")) - break; - match(st.TT_WORD); - String term2 = parse_term(); - term = "@" + op + " " + term + " " + term2; + CQLNode term = parseTerm(index, relation); + while (lexer.ttype != lexer.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) { + 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()); + } } + debug("no more ops"); return term; } - private String parse_term() + private ModifierSet gatherModifiers(String base) throws CQLParseException, IOException { - if (st.ttype == '(') { - match('('); - String expr = parse_expression(); - match(')'); - return expr; + 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); + } } - String word = null; - String attrs = ""; + return ms; + } - // ### We treat ',' and '=' equivalently here, which isn't quite right. - while (st.ttype == st.TT_WORD) { - word = st.sval; - match(st.TT_WORD); - if (st.ttype != '=' && st.ttype != ',') { - // end of qualifer list - break; - } + private CQLNode parseTerm(String index, CQLRelation relation) + throws CQLParseException, IOException { + debug("in parseTerm()"); - String attr = qualsetProperties.getProperty(word); - if (attr == null) { - throw new CQLParseException("unrecognised qualifier: " + word); + String word; + while (true) { + if (lexer.ttype == '(') { + debug("parenthesised term"); + match('('); + CQLNode expr = parseQuery(index, relation); + match(')'); + return expr; + } else if (lexer.ttype == '>') { + return parsePrefix(index, relation, false); } - attrs = attrs + attr + " "; - match(st.ttype); - word = null; // mark as not-yet-read - } - if (word == null) { - // got to the end of a "foo,bar=" sequence - word = st.sval; - if (st.ttype != '\'' || st.ttype != '"') { - word = "\"" + word + "\""; - match(st.ttype); - } else { - match(st.TT_WORD); - } + debug("non-parenthesised term"); + word = matchSymbol("index or term"); + if (!isRelation() && lexer.ttype != lexer.TT_WORD) + break; + + index = word; + String relstr = (lexer.ttype == lexer.TT_WORD ? + lexer.sval : lexer.render(lexer.ttype, false)); + relation = new CQLRelation(relstr); + match(lexer.ttype); + ModifierSet ms = gatherModifiers(relstr); + relation.setModifiers(ms); + debug("index='" + index + ", " + + "relation='" + relation.toCQL() + "'"); } - return attrs + word; + CQLTermNode node = new CQLTermNode(index, relation, word); + debug("made term node " + node.toCQL()); + return node; } - private void match(int token) + private CQLNode parsePrefix(String index, CQLRelation relation, + boolean topLevel) throws CQLParseException, IOException { - if (st.ttype != token) - throw new CQLParseException("expected " + render(st, token, null) + - ", " + "got " + render(st)); - st.nextToken(); + debug("prefix mapping"); + + match('>'); + String name = null; + String identifier = matchSymbol("prefix-name"); + if (lexer.ttype == '=') { + match('='); + name = identifier; + identifier = matchSymbol("prefix-identifer"); + } + CQLNode node = topLevel ? + parseTopLevelPrefixes(index, relation) : + parseQuery(index, relation); + + return new CQLPrefixNode(name, identifier, node); + } + + // 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_EQEQ); } - // ### This utility should surely be a method of the StreamTokenizer class - private static String render(StreamTokenizer st) { - return render(st, st.ttype, null); + private void match(int token) + throws CQLParseException, IOException { + debug("in match(" + lexer.render(token, true) + ")"); + if (lexer.ttype != token) + throw new CQLParseException("expected " + + lexer.render(token, true) + + ", " + "got " + lexer.render()); + int tmp = lexer.nextToken(); + debug("match() got token=" + lexer.ttype + ", " + + "nval=" + lexer.nval + ", sval='" + lexer.sval + "'" + + " (tmp=" + tmp + ")"); } - private static String render(StreamTokenizer st, int token, String str) { - String ret; + private String matchSymbol(String expected) + throws CQLParseException, IOException { - switch (token) { - case st.TT_EOF: return "EOF"; - case st.TT_EOL: return "EOL"; - case st.TT_NUMBER: return "number"; - case st.TT_WORD: ret = "word"; break; - case '"': case '\'': ret = "string"; break; - default: return "'" + String.valueOf((char) token) + "'"; + 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; } - if (str != null) - ret += "(\"" + str + "\")"; - return ret; + throw new CQLParseException("expected " + expected + ", " + + "got " + lexer.render()); } - // ### Not really the right place for this test harness. - // - // e.g. java uk.org.miketaylor.zoom.CQLCompiler - // '(au=Kerninghan or au=Ritchie) and ti=Unix' qualset.properties - // yields: - // @and - // @or - // @attr 1=1 @attr 4=1 Kerninghan - // @attr 1=1 @attr 4=1 Ritchie - // @attr 1=4 @attr 4=1 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) {
- if (args.length != 2) {
- System.err.println("Usage: CQLQuery