Class: Kiba::Extend::Transforms::Reshape::FieldsToFieldGroupWithConstant

Inherits:
Object
  • Object
show all
Defined in:
lib/kiba/extend/transforms/reshape/fields_to_field_group_with_constant.rb

Overview

Convenience transform to rename one or more fields and add a field with a constant value, by default ensuring explicit empty field values and field evenness expected in field groups. Runs the following other transforms, which you will want to understand before using this one:

Examples

Source data:

[
  {note: 'foo', date: '2022'},
  {note: nil, date: '2022'},
  {note: 'foo', date: nil},
  {note: '', date: nil},
  {note: 'foo|bar|baz', date: '2022|2021'},
  {note: 'foo|bar', date: nil},
  {note: 'foo|bar|baz', date: '2022||2021'},
  {note: '|bar|baz', date: '2022|2021'},
  {note: 'foo|bar|', date: '2022|2021'},
]

Value of Kiba::Extend.nullvalue = '%NULLVALUE%'. Value of Kiba::Extend.delim = '|'.

Default behavior

Used in job as:

transform Reshape::FieldsToFieldGroupWithConstant,
  fieldmap: {note: :a_note, date: :a_date},
  constant_target: :a_type,
  constant_value: 'a thing'

Results in:

[
  {a_type: 'a thing', a_note: 'foo', a_date: '2022'},
  {a_type: 'a thing', a_note: nil, a_date: '2022'},
  {a_type: 'a thing', a_note: 'foo', a_date: nil},
  {a_type: nil, a_note: nil, a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|2021|%NULLVALUE%'},
  {a_type: 'a thing|a thing', a_note: 'foo|bar', a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|%NULLVALUE%|2021'},
  {a_type: 'a thing|a thing|a thing', a_note: '%NULLVALUE%|bar|baz', a_date: '2022|2021|%NULLVALUE%'},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|%NULLVALUE%', a_date: '2022|2021|%NULLVALUE%'}
]

With custom treat_as_null string

Used in job as:

transform Reshape::FieldsToFieldGroupWithConstant,
  fieldmap: {note: :a_note, date: :a_date},
  constant_target: :a_type,
  constant_value: 'a thing',
  treat_as_null: '%BLANK%'

Results in:

[
  {a_type: 'a thing', a_note: 'foo', a_date: '2022'},
  {a_type: 'a thing', a_note: nil, a_date: '2022'},
  {a_type: 'a thing', a_note: 'foo', a_date: nil},
  {a_type: nil, a_note: nil, a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|2021|%BLANK%'},
  {a_type: 'a thing|a thing', a_note: 'foo|bar', a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|%BLANK%|2021'},
  {a_type: 'a thing|a thing|a thing', a_note: '%BLANK%|bar|baz', a_date: '2022|2021|%BLANK%'},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|%BLANK%', a_date: '2022|2021|%BLANK%'}
]

With evener: :value (repeat final value of field to even out field values)

Used in job as:

transform Reshape::FieldsToFieldGroupWithConstant,
  fieldmap: {note: :a_note, date: :a_date},
  constant_target: :a_type,
  constant_value: 'a thing',
  evener: :value

Results in:

[
  {a_type: 'a thing', a_note: 'foo', a_date: '2022'},
  {a_type: 'a thing', a_note: nil, a_date: '2022'},
  {a_type: 'a thing', a_note: 'foo', a_date: nil},
  {a_type: nil, a_note: nil, a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|2021|2021'},
  {a_type: 'a thing|a thing', a_note: 'foo|bar', a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|%NULLVALUE%|2021'},
  {a_type: 'a thing|a thing|a thing', a_note: '%NULLVALUE%|bar|baz', a_date: '2022|2021|2021'},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|%NULLVALUE%', a_date: '2022|2021|2021'}
]

With enforce_evenness: false

Used in job as:

transform Reshape::FieldsToFieldGroupWithConstant,
  fieldmap: {note: :a_note, date: :a_date},
  constant_target: :a_type,
  constant_value: 'a thing',
  enforce_evenness: false

Results in:

[
  {a_type: 'a thing', a_note: 'foo', a_date: '2022'},
  {a_type: 'a thing', a_note: nil, a_date: '2022'},
  {a_type: 'a thing', a_note: 'foo', a_date: nil},
  {a_type: nil, a_note: nil, a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|2021'},
  {a_type: 'a thing|a thing', a_note: 'foo|bar', a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|%NULLVALUE%|2021'},
  {a_type: 'a thing|a thing|a thing', a_note: '%NULLVALUE%|bar|baz', a_date: '2022|2021'},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|%NULLVALUE%', a_date: '2022|2021'}
]

With treat_as_null: nil and evener: '%NULL%

Used in job as:

transform Reshape::FieldsToFieldGroupWithConstant,
  fieldmap: {note: :a_note, date: :a_date},
  constant_target: :a_type,
  constant_value: 'a thing',
  treat_as_null: nil,
  evener: '%NULL%'

Results in:

[
  {a_type: 'a thing', a_note: 'foo', a_date: '2022'},
  {a_type: 'a thing', a_note: nil, a_date: '2022'},
  {a_type: 'a thing', a_note: 'foo', a_date: nil},
  {a_type: nil, a_note: nil, a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022|2021|%NULL%'},
  {a_type: 'a thing|a thing', a_note: 'foo|bar', a_date: nil},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|baz', a_date: '2022||2021'},
  {a_type: 'a thing|a thing|a thing', a_note: '|bar|baz', a_date: '2022|2021|%NULL%'},
  {a_type: 'a thing|a thing|a thing', a_note: 'foo|bar|', a_date: '2022|2021|%NULL%'}
]

Since:

  • 2.9.0

Instance Method Summary collapse

Constructor Details

#initialize(fieldmap:, constant_target:, constant_value:, delim: Kiba::Extend.delim, replace_empty: true, treat_as_null: Kiba::Extend.nullvalue, enforce_evenness: true, evener: nil, uneven_warning: true, remove_empty_groups: true) ⇒ FieldsToFieldGroupWithConstant

Returns a new instance of FieldsToFieldGroupWithConstant.

Parameters:

  • fieldmap (Hash{Symbol => Symbol}) —

    map for renaming existing fields. Keys: existing field names. Values: new field names. Forwarded to Kiba::Extend::Transforms::Rename::Fields

  • constant_target (Symbol) —

    new field that will be added to contain constant value

  • constant_value (String) —

    value used to populate constant field

  • replace_empty (Boolean) (defaults to: true) —

    whether to run Kiba::Extend::Transforms::Replace::EmptyFieldValues

  • treat_as_null (nil, String, Array(String)) (defaults to: Kiba::Extend.nullvalue) —

    value(s) to be treated as empty when replacing empty field values and deleting empty field groups. If nil, ‘’ will be used. NOTE: If you need different treat_as_null values for replacing empty field values and deleting empty field groups, use Kiba::Extend::Transforms::Replace::EmptyFieldValues separately in your job before you use this transform.

  • enforce_evenness (Boolean) (defaults to: true) —

    whether to pad field values to ensure all resulting fields have the same number of values

  • evener (String, :value, nil) (defaults to: nil) —

    value used to even out uneven field values. If given a String, that string will be appended to even out fields. If nil, the value of treat_as_null will be used. If :value, the final value in the field will be repeated to even out the field.

  • uneven_warning (Boolean) (defaults to: true) —

    whether to print warnings about uneven field groups to STDOUT

  • remove_empty_groups (Boolean) (defaults to: true) —

    whether to run Delete::EmptyFieldGroups before finalizing row

  • delim (String) (defaults to: Kiba::Extend.delim) —

    used to split/join multiple values in a field

Since:

  • 2.9.0



198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
 
# File 'lib/kiba/extend/transforms/reshape/fields_to_field_group_with_constant.rb', line 198

def initialize(fieldmap:,
  constant_target:,
  constant_value:,
  delim: Kiba::Extend.delim,
  replace_empty: true,
  treat_as_null: Kiba::Extend.nullvalue,
  enforce_evenness: true,
  evener: nil,
  uneven_warning: true,
  remove_empty_groups: true)
  @renamer = Rename::Fields.new(fieldmap: fieldmap)
  @renamed = fieldmap.values
  @target = constant_target
  @value = constant_value
  @delim = delim
  @replace_empty = replace_empty
  @treat_as_null = treat_as_null.nil? ? "" : treat_as_null
  @enforce_evenness = enforce_evenness
  @evener = evener.nil? ? treat_as_null : evener
  @uneven_warning = uneven_warning
  @remove_empty_groups = remove_empty_groups

  @empty_replacer = Replace::EmptyFieldValues.new(
    fields: @renamed,
    value: treat_as_null,
    delim: delim,
    treat_as_null: treat_as_null
  )
  @even_xform = Clean::EvenFieldValues.new(
    fields: @renamed,
    evener: @evener,
    delim: delim,
    warn: uneven_warning
  )
  @group_cleaner = Delete::EmptyFieldGroups.new(
    groups: [[renamed, target].flatten],
    delim: delim,
    treat_as_null: treat_as_null
  )
  @value_getter = Helpers::FieldValueGetter.new(
    fields: renamed,
    delim: delim,
    treat_as_null: treat_as_null
  )
end

Instance Method Details

#process(row) ⇒ Object

Parameters:

  • row (Hash{ Symbol => String, nil })

Since:

  • 2.9.0



245
246
247
248
249
250
251
252
253
254
255
256
257
258
 
# File 'lib/kiba/extend/transforms/reshape/fields_to_field_group_with_constant.rb', line 245

def process(row)
  renamer.process(row)
  vals = value_getter.call(row)
  if vals.empty?
    row[target] = nil
  else
    empty_replacer.process(row) if replace_empty
    max_vals = find_max_vals(row)
    add_constant(row, max_vals)
    even_xform.process(row) if enforce_evenness
    group_cleaner.process(row) if remove_empty_groups
  end
  row
end