lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Robert Muir (JIRA)" <>
Subject [jira] [Updated] (LUCENE-5483) hunspell inaccuracies
Date Sun, 02 Mar 2014 01:32:19 GMT


Robert Muir updated LUCENE-5483:

    Attachment: LUCENE-5483.patch

added additional test for COMPLEXPREFIXES, fixed mark/reset for proper line numbers in error
messages (needed for stuff like hungarian with tons of stuff before the SET), and i did a
bunch of manual testing vs hunspell -m with various real dictionaries (no problems).

This is ready to go.

> hunspell inaccuracies
> ---------------------
>                 Key: LUCENE-5483
>                 URL:
>             Project: Lucene - Core
>          Issue Type: Bug
>            Reporter: Robert Muir
>         Attachments: LUCENE-5483.patch, LUCENE-5483.patch
> I added a lot of tests and greatly refined this algorithm to match correct hunspell behavior.
there were many bugs:
> * recursionLimit was a hack: this is actually specified by the dictionary to be twofold
suffix + one prefix, or if COMPLEXPREFIXES is specified, twofold prefix + one suffix. This
patch removes cursion limit
> * recursion didn't work correctly: it didnt validate multi-level continuation classes
> * add COMPLEXPREFIXES support.
> * probably other minor bugs fixed in the process.
> I validated all testing against hunspell

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message