AppEngine FullTextSearch: Custom scoring

[Filipe Caldas]

Hi,

  We have a considerably big collection on Full Text Search (about 17M rows) and we are using FullTextSearch for searching documents.
  For the search we are basically using 3-Grams of the words and creating a set of potential matches this way, but we need a way to score documents using a custom similarity function (similar to levenshtein distance/Apache Lucene similarity operator), 

after searching a bit on StackOverflow I found a post where someone from the Search API say this was a "high priority" feature, but the post is from 2013. 

https://stackoverflow.com/questions/20659188/google-app-engine-search-api?lq=1 

  Is there a way to track the current status of this? This is a vital functionality in the software and therefore we would need to migrate to another solution if the Search API is not able to handle this need.

Best,

[George (Cloud Platform Support)]

Hello Filipe, 

Where did you find the "high priority" post from 2013? It would come handy to have a reference to it. 

You are right about Elasticsearch, you may use it on compute engine, on in conjunction with Bitnami, "Elasticsearch Certified by Bitnami". 

You may check the full text search "Solution: Algolia" if you use Firebase. 

[Filipe Caldas]

Hi George,

  The reference to this functionality is in the answer for this question on Stack Overflow, I might try deploying the ElasticSearch cluster and see how that goes.

Thanks for your support,