Support term extraction

I think this is superceeded by the mlt functionality

pwolanin - actually, getting the terms from any body of text is different than finding similar text. This would be useful in the current autocomplete effort, for example. It would also be useful for auto-tagging content coming in via FeedAPI. Or for doing term suggestions in autocomplete taxonomy boxes.

The TermComponent of Solr 1.4 doesn't support any filtering. Therefore it's somewhat useful for getting autocomplete suggestions, or seeing how many times a word is in the index, but nothing like what I thought. Postponing.

Taking out of postponed status, as I reserve that status for bugs that are waiting on other issues to move forward.

Nothing is blocking this feature request, so setting to active in 7.x.

I'm confused as to what the feature request is. I'm not sure there is a clear idea here.

Microsoft FAST search has this in core (http://sharepoint.microsoft.com/en-us/product/Pages/Feature-Details.aspx...) and I think this would be a great add-on to SOLR. The use case as I see it:

- Ad admin selects one or more vocabularies that will be used as a base to match terms on.
- When indexing, Drupal scans the search index (the already extracted words from the body and CKK fields) for matching terms and, when found, connects these terms with the nodes. However, these terms should imho only be used for filtering in the faceted search, not as normal term. So maybe these terms should be saved not in term_node but in term_auto_node or so?

Is this viable?

Solr doesn't work this way, afaikt.