Hi All,
Digester 1.6 was released on 9-sep-2004 (7 months ago).
Since then there have been a handful of small bugfixes or improvements.
There's nothing particularly exciting happening here; Digester 1.x
appears to have reached an extremely stable form.
I've updated the RELEASE-NOTES.txt file with the list of all the changes
since 1.6 (see attachment).
Normally the few items that have accumulated since the last release
wouldn't justify a release at this time. However it doesn't appear as if
there will be any more changes to Digester 1.x coming along in the near
future, and there are a few items checked in that will be useful to
digester users.
So it seems to me we may as well make what we have available now.
Bugzilla currently contains 4 enhancement requests, none of which appear
to have anyone willing to work on them - or to use them if they should
ever be implemented. I've certainly got little enthusiasm for addressing
any of these enhancements myself (though I would be willing to review
patches if any are put forward).
There's also one trivial enhancement/bug entry to do with MD5 sums for
jar files that can be addressed during the release procedure.
I'm happy to be the release manager for this one if that's ok with
y'all. So what do you think? Should we work towards a release 1.7 or
not?
Cheers,
Simon
$Id: RELEASE-NOTES.txt 155412 2005-02-26 12:58:36Z dirkv $
Commons Digester Package
Version 1.7
Release Notes
INTRODUCTION
============
This is a minor bugfix and maintenance release. A few small features have been
added.
New projects are encouraged to use this release of digester, but there is no
urgency
for existing projects to upgrade; Digester 1.6 has proven to be a stable
release.
This release is 100% binary and source compatible with the previous release.
RELEASE TODO
============
* update xdocs/index.xml
* xdocs/navigation.xml
* project.xml
* build.xml
* src/conf/MANIFEST.MF
* verify package.html is correct html.
* decide on javadoc license text in footer
* run clirr to check for changes between releases.
IMPORTANT NOTES
===============
* The jakarta commons project has migrated to the Subversion version control
system
(previously, CVS was used). There should be no effect on users of the Digester
library, but obviously the process of examining the latest source code, and of
creating patches for Digester has now changed. Please see the jakarta commons
website for details (http://jakarta.apache.org/commons).
Dependencies
=============
Release 1.7 has the same dependencies as release 1.6.
Compatible Dependency Sets:
Digester 1.7 + Logging 1.0.x + BeanUtils 1.x + Collections 2.x
Digester 1.7 + Logging 1.0.x + BeanUtils 1.x + Collections 3.x
Digester 1.7 + Logging 1.0.x + BeanUtils 1.7
NEW FEATURES
=============
Improved Documentation
----------------------
As usual, documentation has improved in this release.
Minor Javadoc improvements occur in the following classes:
Rule, RulesBase, ExtendedBaseRules,
NodeCreateRule, CallMethodRule, CallParamRule, SetNextRule
The javadoc package documentation (package.html) has also had minor
updates to the following topics:
* How Digester can be used as a SAX content handler.
* How wildcard rules are ignored if non-wildcard matches
are available.
Digester
--------
Named stacks are now cleared by the clear() method. Note that it is recommended
that a new Digester instance be created for each document parsed, hence this
should not be necessary.
Method resetRoot has been added. Again, this should only be relevant for
programs
that attempt to reuse a single Digester instance to process multiple documents
(which is not recommended).
Method peek(String stackname, int n) has been added for consistency, to allow
access to arbitrary objects on named stacks. Thanks to Brian Hanafee for the
suggestion (bugzilla #33873).
SetNestedPropertiesRule
------------------------
The toString method has been improved, for better logging diagnostics.
Patch provided by Wendy Smoak.
The addressbook sample now demonstrates use of SetNestedPropertiesRule.
SetPropertiesRule
-----------------
A new ignoreMissingProperty flag can be set false to cause
an exception to be generated when the xml contains an
attribute not available on the target bean. Patch contributed
by Gabriele Carcassi.
Xmlrules Enhancements
----------------------
The xmlrules module has had a number of minor updates to provide access
to functionality that was previously accessable only via the digester
API:
-- add set-nested-properties-rule tag. Much of this
patch provided by Wendy Smoak.
-- add targetoffset attribute to call-method-rule tag,
to allow the target object whose method is invoked
to be any object on the digester stack. Patch by
Wendy Smoak (bugzilla #33550).
-- add stack-index attribute to call-param-rule tag,
to allow passing any object on the digester stack
as a parameter to a target method. Patch by
Kris Nuttycombe, inspired by Blake Meike.
The digester-rules.dtd file has also been slightly cleaned up to
remove unnecessary duplication.
Plugins Enhancements
--------------------
PluginCreateRule can now have an associated "factory" object for the Rules
object created at each "plugin point". This allows Rules implementations
other than RulesBase to be used with plugins. It also gives user code the
chance to set up default Rule objects within the Rules object which will
apply to all plugins. This change affects classes PluginRules,
PluginCreateRule and (new class) RulesFactory.
BUGS FROM PREVIOUS RELEASE
==========================
Reports Addressed
------------------
#31393 SetNestedPropertiesRule causes StackOverflowError
-- thanks to James Pine
#28739 Problem with parse(File) on windows machines
-- thanks to John Taylor
DEPRECATIONS
============
None
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
