Class: Latin1Validator

Inherits:
ActiveModel::EachValidator
  • Object
show all
Defined in:
app/validators/latin1_validator.rb

Overview

NPG expressed concerns that switching the Sequencescape to Utf8 would result in an increase in characters which could not be processed by their pipelines. This validator checks that the characters can be represented in latin1, ensuring that we don't make the problem worse in the short term. This will likely be replaced by more specific validations as we become aware of exactly what is supported. It has been applied to the key fields that get added to headers and the like.

Instance Method Summary collapse

Instance Method Details

#validate_each(record, attribute, value) ⇒ Object 



11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
 
# File 'app/validators/latin1_validator.rb', line 11

has_metadata do
  # NOTE: The following attribute is not required for Microarray Genotyping.
  # I think this might be broken and suggests that there should be separate classes for project: one for
  # next-gen sequencing that includes this attribute in it's metadata, and one for microarray genotyping
  # that doesn't.
  include ProjectManager::Associations
  include BudgetDivision::Associations

  custom_attribute(:project_cost_code, required: true)
  custom_attribute(:funding_comments)
  custom_attribute(:collaborators)
  custom_attribute(:external_funding_source)
  custom_attribute(:sequencing_budget_cost_centre)
  custom_attribute(:project_funding_model, in: PROJECT_FUNDING_MODELS)
  custom_attribute(:gt_committee_tracking_id)

  before_validation do |record|
    record.project_cost_code = nil if record.project_cost_code.blank?
    record.project_funding_model = nil if record.project_funding_model.blank?
  end
end