Unicode aware lexicon for ZCTextIndex
The standard ZCTextIndex lexicon only deals with 8-bit strings (and only if you get the zope.conf locale setting right). It does not handle Unicode or UTF-8. UnicodeLexicon fills this gap.
This product adds a ZCTextIndex Unicode Lexicon type to Zope. The lexicon comes with word splitters, stop word removers, a case normalizer, and two accent normalizers.
If you have GenericSetup installed, you can use the included extension profile to create a UnicodeLexicon in your portal_catalog and update the Title, Description, and SearchableText ZCTextIndexes.
There is no upgrade path from UnicodeLexicon 1.0. If you have 1.0 on your system, you have to delete and recreate the lexicon.
The splitter works with all languages that separate words with whitespace characters.
The stop word remover knows about English language stop words only.
The accent normalizer comes in two flavors. There is a normalizer for Latin and Western European text (fr, es, pt, it, en, nl), and one for German and Scandinavian text (de, dk, no, se, fi, is). The latter keeps the umlaut characters ä, ö, and ü in tact.
The lexicon assumes either Unicode or UTF-8. If your application uses a different encoding, you must change the enc constant in UnicodeSplitter.py accordingly.
2.0 - 2011-01-21
Add an ordered PipelineFactory. [stefan]
Add an accent-normalizing pipeline element originally contributed by Marc-Auréle Darche. [stefan]
Release as Python egg. [stefan]
1.0 - 2006-08-14
Initial release. [stefan]
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.