ElasticSearch - substring search

gistfile1.txt

You have two options:

    Use ngrams through the ngram token filter: http://www.elasticsearch.org/guide/reference/index-modules/analysis/ngram-tokenfilter.html. See Wikipedia for details: http://en.wikipedia.org/wiki/N-gram
    Use the compound token filter http://www.elasticsearch.org/guide/reference/index-modules/analysis/compound-word-tokenfilter.html (which I have contributed to Lucene btw.)

The ngrams are useful if you cannot provide a dictionary because you don't know what type of documents you have to index. The downside is that they use a lot of space in your indices. So you need to be careful here.