I'm working on a site with the Search API Solr module.
I notice that, in fulltext searches, the system gives results only for entire word. If I've a title like "Beautiful", if I search "eautiful" it doesn't find something. Is there a way to solve this?
Set up Solr with the class EdgeNGramFilterFactory. This will not catch 'eautiful' instead of 'beautiful', but will give results on bea, beau, beaut and so on depending on minGramSize and maxGramSize. IMO this will be sufficient for most searches.
You could try adding 'beautiful' to the spelling suggestions for Solr. I believe there is a spellings.txt file in the Solr profile you are using for your site. It's probably in the apache-solr-1.4.1/example/solr/conf folder of your Solr installation. This is just from memory of setting this up several months ago, so the location could be wrong. The spelling suggestions will just notify the user that they may have misspelled a word and suggest what it thinks is the correct spelling. If you want to automatically search for the new spelling, you would probably have to write something yourself.
Some more details:
Did you try fuzzy search project? It's maybe a good solution...If you write beaut he will match beautiful
As described in the documentation
Each word in a node is split into 3 (default) letter lengths, so 'apple' gets indexed with 3 smaller strings 'app', 'ppl', 'ple'. The effect of this is that as long as your search matches X percentage (administerable in the admin settings) of the word the node will be pulled up in the results.
I think i will implement this module in one week on a server test because my actual boss wants some result even if there are some mispellings.
