elastics-models - Result Extenders

The elastics-models gem adds a few extender to the standard extended Elastics::Result object (see Result Extenders). They are aware of your models, so you can do, for example:

# search the index
result = MyClass.my_search :my_tag => 'the tag value'

# retrieve the document from the index as usual
collection = result.collection
#=> [ ... the hits array ... ]

# you can load one record at a time
document = result['hits']['hits'][0]
document.load
#=> #<Project id: 167017 ...> # ActiveRecord object loaded from the DB

# or easier
collection.first.load
#=> #<Project id: 167017 ...> # same thing

# or retrieve all the records from your DB in one go
loaded_collection = result.loaded_collection
#=> [ #<Project id: 167017 ...>, #<Comment id: 2342 ...>, ... ]

# notice that record can belong to different models

Elastics Model Result Extenders

• Elastics::Result::SearchLoader
  It extends the results coming from a search query. It adds the following method:

  • result.loaded_collection
    This method takes the index, types and ids from the collection, map them back to your original models, and query the DB with as less as possible queries, returning a collection of records (instead of a collection of elasticsearch documents). The records might be composed of mixed classes, depending on what you retrieved.

    This is an operation that requires an extra query per each type returned. For example if your collection contains 20 documents of a total of 3 different types, it generates 3 extra DB queries to retrieve the original models. For this reason it should be used only when the performance is not a problem, besides, when you use it, you can avoid to load the document source, by using Elastics::Template::SlimSearch templates, which are just regular Search Templates, that don’t retrieve the source that in this case is not needed.

• Elastics::Result::DocumentLoader
  Applies to documents that contain (at least, but not limited to) _index, _type, _id. It adds the following method:

  • document.load
    Maps the document back to the original model and class and retrieves it from the DB.

  • document.load!
    Like document.load but raises a Elastics::DocumentMappingError if it cannot find the model class for the document.

• Elastics::Result::ActiveModel
  It extends the results when the context class (your class) includes Elastics::ActiveModel. It adds the following methods:

  • result.get_docs
    This method is used internally, in order to build the objects of your Elastics::ActiveModel extended class, out of the raw elasticsearch result.

    You will seldom need to use this method, unless you pass :raw_result => true.