DocumentHydrator takes a document, represented as a Ruby Hash, and efficiently updates it so that embedded references to other documents are replaced with their corresponding subdocuments.

Along with the document, DocumentHydrator takes a path (or array of paths) specifying the location of the references to be expanded and a Proc--known as the hydration proc--that is capable of providing expanded subdocuments for those references. The hydration proc is guaranteed to be called at most once during any given invocation of DocumentHydrator, ensuring efficient hydration of multiple subdocuments.


Install the gem:

gem install document_hydrator

Require the library:

require 'document_hydrator'

If you want to use the optional MongoDB collection hydrator, make sure you require the MongoDB driver first:

require 'mongo'
require 'document_hydrator'

Hydration Procs

Hydration procs are responsible for transforming an array of document references into a hash that maps those references to their corresponding subdocuments. The subdocuments must themselves be hashes.

DocumentHydrator provides a hydration proc factory that makes it simple to hydrate a document when pulling the subdocuments from a MongoDB collection. Use of the factory is described in the "Hydrating Documents with MongoDB Collections" section found later in the document.

Most of the following examples illustrate DocumentHydrator functionality that is independent of the choice of hydration proc. In order to keep them stand-alone, a simple "identity" hydration proc will be used:

identity_hydrator = do |ids|
  ids.inject({}) do |hash, id|
    hash[id] = { 'id' => id }

It simply maps IDs to hashes containing the ID under the key 'id'. For example:[1, 2, 3])
# => {1=>{"id"=>1}, 2=>{"id"=>2}, 3=>{"id"=>3}}

A Simple Example

Armed with identity_hydrator, the simplest example is:

doc = { 'thing' => 1, 'gizmo' => 3 }
DocumentHydrator.hydrate_document(doc, 'thing', identity_hydrator)
# => {"thing"=>{"id"=>1},"gizmo"=>3}

In this case DocumentHydrator is asked to hydrate a single document, doc, by replacing the reference found at doc['thing'] to its corresponding subdocument, as provided by identity_hydrator.


In the example above, the path was the name of a top level key in the document to be hydrated. DocumentHydrator also supports paths to arrays, nested paths, and nested paths that contain intermediate arrays.

For example, consider the document:

status_update = {
  'user' => 19,
  'text' => 'I am loving MongoDB!',
  'likers' => [37, 42, 99],
  'comments' => [
    { 'user' => 88, 'text' => 'Me too!' },
    { 'user' => 99, 'text' => 'Drinking the Kool-Aid, eh?' },
    { 'user' => 88, 'text' => "Don't be a hater. :)" }

The following are all valid hydration paths referencing user IDs in status_update:

  • 'user' -- single ID
  • 'likers' -- array of IDs
  • 'comments.user' -- single ID contained within an array of objects

Multi-path Hydration

DocumentHydrator will accept an array of paths to all be hydrated concurrently:

  ['user', 'likers', 'comments.user'],
pp status_update
# {"user"=>{"id"=>19},
#  "text"=>"I am loving MongoDB!",
#  "likers"=>[{"id"=>37}, {"id"=>42}, {"id"=>99}],
#  "comments"=>
#   [{"user"=>{"id"=>88}, "text"=>"Me too!"},
#    {"user"=>{"id"=>99}, "text"=>"Drinking the KoolAid, eh?"},
#    {"user"=>{"id"=>88}, "text"=>"Don't be a hater. :)"}]}

Regardless of the number of paths, the hydration is accomplished with a single call to the hydration proc.

Multi-document Hydration

Multiple documents may be hydrated at once using DocumentHydrator.hydrate_documents:

doc1 = { 'thing' => 1, 'gizmo' => 3 }
doc2 = { 'thing' => 2, 'gizmo' => 3 }
DocumentHydrator.hydrate_documents([doc1, doc2], 'thing', identity_hydrator)
# => [{"thing"=>{"id"=>1}, "gizmo"=>3}, {"thing"=>{"id"=>2}, "gizmo"=>3}]

The only difference between hydrate_document and hydrate_documents is that the latter takes an array of documents. The other parameters are the same.

_id Suffix Stripping

DocumentHydrator automatically strips any 'id' or 'ids' suffixes from keys that are the last step in a hydration path:

doc = { 
  'user_id' => 33,
  'follower_ids' => [11, 23]
  ['user_id', 'follower_ids'], identity_hydrator)
# => {"user"=>{"id"=>33}, "followers"=>[{"id"=>11}, {"id"=>23}]}

Notice that the document now has the keys 'user' and 'followers'.

Hydrating Documents with MongoDB Collections

DocumentHydrator provides a hydration proc factory that makes it simple to hydrate documents with subdocuments that are fecthed from a MongoDB collection.

The following examples require a bit of setup:

require 'mongo'
db ='document_hydrator_example')
users_collection = db['users']
users_collection.insert('_id' => 1, 'name' => 'Fred', 'age' => 33)
users_collection.insert('_id' => 2, 'name' => 'Wilma', 'age' => 30)
users_collection.insert('_id' => 3, 'name' => 'Barney', 'age' => 29)
users_collection.insert('_id' => 4, 'name' => 'Betty', 'age' => 28)

Now create a hydration proc that fetches documents from the users collection:

user_hydrator = DocumentHydrator::HydrationProc::Mongo.collection(users_collection)

Note that DocumentHydrator::HydrationProc::Mongo is automatically loaded by require 'document_dehydrator' if the MongoDB Ruby Driver has already been loaded.

Here is the hydration proc in action:

doc = { 'user_ids' => [1, 3] }
Documenthydrator.hydrate_document(doc, 'user_ids', user_hydrator)
# => {"users"=>[{"_id"=>1, "name"=>"Fred", "age"=>33}, {"_id"=>3, "name"=>"Barney", "age"=>29}]}    

Limiting Fields in Subdocuments

By default a Mongo collection hydrator will return all of the fields that are present for the subdocument in the database. This can be changed, however, by passing an optional :fields argument to the factory. This option takes the same form as it does for Mongo::Collection#find.

For example:

user_hydrator = DocumentHydrator::HydrationProc::Mongo.collection(users_collection,
  :fields => { '_id' => 0, 'name' => 1 })
DocumentHydrator.hydrate_document(doc, 'user_ids', user_hydrator)
# => {"users"=>[{"name"=>"Fred"}, {"name"=>"Barney"}]}

Supported Rubies

DocumentHydrator has been tested with:

  • Ruby 1.8.7 (p334)
  • Ruby 1.9.2 (p180)
  • JRuby 1.6.2


Copyright (c) 2011 Greg Spurrier. See LICENSE.txt for further details.