I am pleased to announce the release of TextIndexNG 2.1.0 final: Changes in 2.1.0 (since latest 2.0.8): - **Minimum Python version:** Python 2.3.3 or higher - **Minimum Zope version:** Zope 2.7 - workaround for incorrect treatement of HTML entities in strip-o-gram (affected HTML document converter) - fixed entity handling in SGML and XML converters - misc. converter fixes - phrase and NEAR search should be working again - optimized space usage of the StandardStorage - added a new storage *StupidStorage* which uses less space but neither supports near, phrase nor ranking - the storage can be choosen at creation time - **Exisiting indexes should be deleted and re-created** - Fixed a bug reported by Simon Pamies where also unindexed documents appeared as indexed - fixed bug in reindexing behaviour in StupidStorage (patch by Dieter Maurer) - optimized implementation of StupidStorage and StandardStorage - fixed a list comprehension with a quadratic runtime behaviour - Better integration with Plone: TextIndexNG can now be installed through Plone Setup/Add+Remove programs. This will install a configlet under Plone Setup that allows you to convert the SearchableText index from ZCTextIndex to TextIndexNG2 - API enhancements: the 'extra' parameter of the constructor can now be a dictionary of parameters instead of being a record-like object Download it from http://www.zope.org/Members/ajung/TextIndexNG Documentation is http://zope.org/Members/ajung/TextIndexNG/wiki/TextIndexNG -aj