[jira] Commented: (SOLR-199) N-gram

["Hoss Man (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SOLR-199?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12492947 ] 

Hoss Man commented on SOLR-199:
-------------------------------

NGramTokenizerFactory is refering to constants from, and constructing an instance of, EdgeNGramTokenizer

I'm also not crazy about some of the utilities being added to BaseTokenizerFactory .. at a minimum they need better names (like getStringArg) but i'm not really clear on what this is suppose to mean at all...

           protected int getInt(String name, int defaultVal, boolean useDefault)

...if i don't want to use the default, then what am i suppose to pass as the defaultVal?

how about if we don't make any changes to BaseTokenizerFactory and just let subclasses that want convenience methods for dealing with args use MapSolrParams and the methods it supports?

> N-gram
> ------
>
>                 Key: SOLR-199
>                 URL: https://issues.apache.org/jira/browse/SOLR-199
>             Project: Solr
>          Issue Type: New Feature
>          Components: search
>            Reporter: Adam Hiatt
>            Priority: Trivial
>         Attachments: SOLR-81-ngram.patch
>
>
> This tracks the creation of a patch that adds the n-gram/edge n-gram tokenizing functionality that was initially part of SOLR-81 (spell checking). This was taken out b/c the lucene SpellChecker class removed this dependency. None-the-less, I think this is useful functionality and the addition is trivial. How does everyone feel about such an addition?

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.