$Id: Changes,v 1.13 2002-11-17 23:29:02 mike Exp $

Revision history for "cql-java"
See the bottom of this file for a list of things still to do.

0.4  (IN PROGRESS)
	- Add support for the new "phonetic" relation modifier,
	  including its mapping in "etc/pqf.properties".
	- Make prefix-maps bind loosely, so that ``>dc=x a and b''
	  applies the dc=x binding to both the ``a'' and ``b''
	  sub-queries.
	- Change build process so that javadoc documentation is built
	  by "make" in the "docs" directory.
	- Write javadoc comments for CQLRelation and ModifierSet.
	- Other improvements to javadoc comments (e.g. describe the
	  `config' parameter to the toPQF() method.)

0.3  Fri Nov 15 12:04:04 2002
	- Allow keywords to be used unquoted as search terms.
	- Add support for serverChoiceRelation (scr).
	- Add support for prefix-mapping, as in
		>dc="http://dublincore.org/ dc.title=fish
	  and
		>"http://dublincore.org/ title=fish
	  The XCQL generated corresponds to Adam's suggested format
	  rather than Rob's, not so much because I prefer it (although
	  I do) as because it's what fell out when I just Coded What
	  Comes Natur'ly.  That may need to change, depending on the
	  result of the ZNG list's deliberations.
	- Fix the parser to normalise relation modifiers to lower case.
	- Fix the CQLParser test harness not to emit an extraneous
	  blank line at end of XCQL output.
	- Fix CQLNode documentation to contain a link to YAZ's
	  documentation of Prefix Query Format (PQF) rather than
	  containing a rather unhelpful chunk of BNF.
	- Change the test/regression Makefile so that "make clean" now
	  does what "make distclean" used to do - the distinction
	  between them is pointless.
	- Fix a few typos in the documentation.
	- Move the README file's old "THINGS TO DO" section to the end
	  of this file, the new "Still to do" section.

0.2  Wed Nov  6 23:05:54 2002
	- Fix the order of proximity parameters in accordance with the
	  updated official grammar, which now specifies proximity
	  operators of the form
		prox/<relation>/<distance>/<unit>/<ordering>
	  as in ``foo prox/<=/3/sentence bar''.
	- Make the necessary parts of the CQLNode class and its
	  subclasses public.  This means that client code can now walk
	  through parse-trees and so implement its own back-end
	  (e.g. to build BER-friendly data structures using whatever
	  Z39.50 toolkit is preferred.)
	- Add the toPQF(Properties p) method to CQLNode and its
	  subclasses.  This produces a query in YAZ-style Prefix Query
	  Format, which can be trivially translated into a Z39.50
	  Type-1 query (see, for example, JZKit's code to do so).
	- Add etc/pqf.properties to configure the toPQF() method.
	- Add "-p <props-file>" option to the CQLParser test-harness,
	  indicating that the parsed tree is to be rendered to PQF.
	- Add PQFTranslationException and its subclasses
	  UnknownQualifierException, UnknownRelationException,
	  UnknownRelationModifierException and
	  UnknownPositionException.
	- Rename ParameterMissingException to MissingParameterException.
	- Add javadoc comments for CQLNode and its subclasses.

0.1  Sun Nov  3 20:58:27 2002
	- First public release.

--

### Still to do
	- Fix the bug where "9x" is parsed as two tokens, a TT_NUMBER
	  followed by a TT_WORD.  The problem here is that I don't
	  think it's actually possible to fix this without throwing
	  out StreamTokenizer and rolling our own, which we absolutely
	  _don't_ want to do.
	- Find out what the ZNG group finally settled on as the
	  correct XCQL representation of prefix-mapping, and tweak my
	  toXCQL() output to match that.
	- Add srw.resultSet support to the toPFQ() method.
	- Write "overview" file for the javadoc documentation.
	- Some niceties for the cql-decompiling back-end:
	  * Don't emit redundant parentheses.
	  * Don't put spaces around relations that don't need them.
	- Consider the utility of yet another back-end that translates
	  a CQLNode tree into JZKit's representation of a Type-1 query
	  tree.  That would be nice so that CQL could become a JZKit
	  query-type; but you could achieve the same effect by
	  generating PQF, and running that through JZKit's existing
	  PQN-to-Type-1 compiler.
	- Many refinements to the random query generator:
	  * Generate relation modifiers
	  * Proximity support
	  * Don't always generate qualifier/relation for terms
	  * Better selection of qualifier (configurable?)
	  * Better selection of terms (from a dictionary file?)
	  * Introduce wildcard characters into generated terms
	  * Generate multi-word terms