class PatientScopedUpdater

Public Class Methods

Source

# File app/lib/patient_scoped_updater.rb, line 4
def initialize(patient_scope: nil, patient: nil)
  @patient_scope = (patient ? Patient.where(id: patient) : patient_scope)
end

Public Instance Methods

is_id_only_scope? (scope)

Source

# File app/lib/patient_scoped_updater.rb, line 41
def is_id_only_scope?(scope)
  if scope.nil? || scope.joins_values.any? ||
       scope.left_outer_joins_values.any?
    return false
  end

  values = scope.where_values_hash
  values.key?("id") && values.size == 1
end

is_patient_scope_id_only? ()

Source

# File app/lib/patient_scoped_updater.rb, line 37
def is_patient_scope_id_only?
  @is_patient_scope_id_only ||= is_id_only_scope?(patient_scope)
end

merge_patient_scope (scope)

Source

# File app/lib/patient_scoped_updater.rb, line 27
def merge_patient_scope(scope)
  if is_patient_scope_id_only?
    scope.where(patient_id: patient_scope.select(:id))
  elsif patient_scope
    scope.joins(:patient).merge(patient_scope)
  else
    scope
  end
end

The following code is necessary to support an optimisation where we’re

updating a single patient by ID.

This method allow us to check if the scope where clause is only checking

the ID of the patient, meaning we can optimise out the `JOIN` and filter
the table directly.

For example, without this optimisation, a scope of ‘Patient.where(id: 1)`

would result in the following SQL query:

SELECT patient_id, team_id FROM patient_teams JOIN patients ON patients.id = patient_teams.patient_id WHERE patients.id = 1

Whereas we know that the following SQL query is all we need in this case:

SELECT patient_id, team_id FROM patient_teams WHERE patient_teams.patient_id = 1