| Release NotesJanuary 14, 2008: Qizx/db is
now a XMLmind productPlease visit http://www.xmlmind.com/qizx/. Version 0.2p7 -Sept 23, 2007- Bug fix:
- Serialization: The <?xml declaration was appearing in HTML,
even with the option omit-xml-declaration equal to true.
Version 0.2p6 -June 11, 2007- Bug fix:
- Serialization: characters not supported by the encoding are now
output as a character entity reference. in HTML mode, symbolic
character entities defined in HTML4 are used if the encoding cannot
represent them: for example Unicode character U+8482 (TM sign) is
output as ™
- Java API: the QName passed as argument of the "endElement" event
in SAXEventHandler and derived classes (eg NodeXMLReader) was not
correct.
Version 0.2p5 -Feb 17, 2007- Bug fixing:
Java binding issue with overloaded methods: the basic problem
is that XQuery does not support function overloading, therefore
calling overloaded Java methods bound as XQuery functions should not
be supported by Qizx in principle. This feature was however
partially supported in Qizx/open 1.0 but the changes in type
handling in 1.1 have made it nearly unusable. Qizx/db 0.2p5
attempts to improve on the current situation: the types of arguments
(declared or inferred) are used to select the best matching
overloaded method. This should solve most cases. However there are
still issues when argument types map to xdt:object. See the
documentation for more details and for examples.
Version 0.2p4 -Nov 13, 2006- Bug fixing:
- The "export results" command now saves all items of the sequence
(not only the displayed items)
- indexing: in a collection containing more than 4096 documents,
the first documents were invisible.
- Some query result sequences were not rewinding properly (for
example //X[@attr > 1])
- The Java API documentation (javadoc) has been withdrawn from the web
site. Available in the distribution package.
Version 0.2p3 -Oct 1st, 2006Bug fixing and
conformance release: passes 99.9% of all 'MinimalConformance' tests from the
latest instalment (v1.0) of the XQuery Test Suite. Passes also all the tests
of the optional feature "Modules". Remaining issues: - Functions id/idref. Implemented for parsed documents, but not yet
for database documents.
- Advanced Unicode features: surrogates and normalizations. Low
priority, as very few users are likely to be concerned.
Other bug fixes: - Compilation bug in complex FLWOR expressions [B. Haible/ILOG]
Version 0.2p2 - Sep 3, 2006Bug fixing and
conformance release: passes more than 99.5% of all 'MinimalConformance'
tests from the latest instalment (v1.0) of the XQuery Test Suite. Passes
also all the tests of the optional feature "Modules". Remaining issues
(0.5%) mainly concern: - Functions id/idref. Will be implemented, but not in the near
future.
- Advanced Unicode features: surrogates and normalizations. Low
priority, as very few users are likely to be concerned.
New features: - The collection() function now accepts a list of document URI's
separated by commas or semicolons.
Version 0.2p1 - June 29, 2006Bug fixing
version: - Slight improvements in conformance: 96.5% of XQuery Test Suite 0.9
(95% of XQTS 0.9.4)
- Fixed blocking bug in namespace handling (wrong "duplicate
attribute" error).
- Introduced 3 new function names in SQL Connectivity, due to an issue
with function overloading: sqlx:execStatement, sqlx:rawExecStatement,
sqlx:execUpdateStatement.
Known bugs, not yet fixed: - Java binding: trying to call overloaded methods or constructors in a
Java class may fail in a platform dependent way. The issue was already
existing in 0.1 but has been worsened by type-checking changes in
0.2.
Version 0.2 (alpha) - May 29, 2006Initial alpha release Version 0.1 - April 10,
2006This version is already fairly fast and scalable, but may suffer
some functional instability, hence the term "alpha".
© 2004-2008 Axyana software.
Site update
2008-01-15
using Qizx.
Java and all Java-based marks are trademarks or
registered trademarks of Sun Microsystems, Inc.
in the U.S. and other countries.
Acrobat is a trademark of Adobe Systems Incorporated. |