Module: WorkerTools::CsvInput

Defined in:
lib/worker_tools/csv_input.rb

Defined Under Namespace

Classes: CsvInputForeach

Instance Method Summary collapse

Instance Method Details

#csv_input_columns ⇒ Object

If an array is provided, the names will be used as the row keys, the row values will be assign according to the columns order.

Ex: %w(tenant segment area) row => 

tenant: _value_at_first_column_,
segment: _value_at_second_column_,
area: _value_at_third_column_

If a hash if provided, the keys will turn into the row keys, the values will be used to find the corresponding columns (the order in the csv won’t affect the import)

Ex: { tenant: ‘Mandant’, segment: ‘Segment’, area: ‘Bereich’) row => 

tenant: _value_at_column_Mandant,
segment: _value_at_column_Segment,
area: _value_at_column_Bereich

The name of the column is filtered using the csv_input_header_normalized method, which takes care of extra spaces and looks for a case insentive match (so ‘Bereich’ matches ‘ Bereich’, ‘bereich’, etc.). You can override that method as well.

Besides matching the columns using strings, it is possible to use a regular expression or a proc: {

tenant: 'Mandant',
segment: /Segment/i,
area: ->(name) { name.downcase == 'area' }

}



38
39
40
 
# File 'lib/worker_tools/csv_input.rb', line 38

def csv_input_columns
  raise "csv_input_columns has to be defined in #{self}"
end

#csv_input_columns_array_check(csv_rows_enum) ⇒ Object

Raises:



61
62
63
64
65
66
67
68
69
70
 
# File 'lib/worker_tools/csv_input.rb', line 61

def csv_input_columns_array_check(csv_rows_enum)
  return if csv_rows_enum.first.nil?

  expected_columns_length = csv_input_columns.length
  actual_columns_length = csv_rows_enum.first.length
  return if expected_columns_length == actual_columns_length

  msg = "The number of columns (#{actual_columns_length}) is not the expected (#{expected_columns_length})"
  raise Errors::WrongNumberOfColumns, msg
end

#csv_input_columns_check(csv_rows_enum) ⇒ Object 



54
55
56
57
58
59
 
# File 'lib/worker_tools/csv_input.rb', line 54

def csv_input_columns_check(csv_rows_enum)
  # override and return true if you do not want this check to be performed
  return csv_input_columns_array_check(csv_rows_enum) if csv_input_columns.is_a?(Array)

  csv_input_columns_hash_check(csv_rows_enum)
end

#csv_input_columns_hash_check(csv_rows_enum) ⇒ Object

Raises:



72
73
74
75
76
77
78
79
 
# File 'lib/worker_tools/csv_input.rb', line 72

def csv_input_columns_hash_check(csv_rows_enum)
  raise Errors::MissingColumns, 'The headers are missing' if csv_rows_enum.first.nil?

  expected_names = csv_input_columns.values
  filtered_actual_names = csv_rows_enum.first.map { |n| csv_input_header_normalized(n) }
  csv_input_columns_hash_check_duplicates(filtered_actual_names)
  csv_input_columns_hash_check_missing(filtered_actual_names, expected_names)
end

#csv_input_columns_hash_check_duplicates(names) ⇒ Object

Raises:



81
82
83
84
85
86
 
# File 'lib/worker_tools/csv_input.rb', line 81

def csv_input_columns_hash_check_duplicates(names)
  dups = names.group_by(&:itself).select { |_, v| v.count > 1 }.keys
  return unless dups.present?

  raise Errors::DuplicatedColumns, "The file contains duplicated columns: #{dups}"
end

#csv_input_columns_hash_check_missing(actual_names, expected_names) ⇒ Object

Raises:



88
89
90
91
92
93
94
 
# File 'lib/worker_tools/csv_input.rb', line 88

def csv_input_columns_hash_check_missing(actual_names, expected_names)
  missing = expected_names.reject do |name|
    matchable = name.is_a?(String) ? csv_input_header_normalized(name) : name
    actual_names.any? { |n| case n when matchable then true end } # rubocop does not like ===
  end
  raise Errors::MissingColumns, "Some columns are missing: #{missing}" unless missing.empty?
end

#csv_input_csv_options ⇒ Object 



96
97
98
99
 
# File 'lib/worker_tools/csv_input.rb', line 96

def csv_input_csv_options
  # Ex: { col_sep: ';', encoding: Encoding::ISO_8859_1 }
  { col_sep: ';' }
end

#csv_input_file_path ⇒ Object 



140
141
142
 
# File 'lib/worker_tools/csv_input.rb', line 140

def csv_input_file_path
  model.attachment.path.to_s
end

#csv_input_file_presence_check ⇒ Object

Raises:



152
153
154
155
 
# File 'lib/worker_tools/csv_input.rb', line 152

def csv_input_file_presence_check
  raise Errors::EmptyFile, 'The file does not exist' unless File.exist?(csv_input_file_path)
  raise Errors::EmptyFile, 'The file is empty' if File.zero?(csv_input_file_path)
end

#csv_input_foreach ⇒ Object 



157
158
159
160
161
162
163
164
165
166
167
168
169
 
# File 'lib/worker_tools/csv_input.rb', line 157

def csv_input_foreach
  @csv_input_foreach ||= begin
    csv_input_file_presence_check
    csv_input_columns_check(csv_rows_enum)

    CsvInputForeach.new(
      rows_enum: csv_rows_enum,
      mapping_order: csv_input_mapping_order(csv_rows_enum.first),
      cleanup_method: method(:cvs_input_value_cleanup),
      headers_present: csv_input_headers_present
    )
  end
end

#csv_input_header_normalize? ⇒ Boolean

Returns:

  • (Boolean) 


105
106
107
 
# File 'lib/worker_tools/csv_input.rb', line 105

def csv_input_header_normalize?
  true
end

#csv_input_header_normalized(name) ⇒ Object 



42
43
44
45
46
 
# File 'lib/worker_tools/csv_input.rb', line 42

def csv_input_header_normalized(name)
  name = name.to_s.strip
  name = name.downcase if csv_input_header_normalize?
  name
end

#csv_input_headers_present ⇒ Object 



148
149
150
 
# File 'lib/worker_tools/csv_input.rb', line 148

def csv_input_headers_present
  true
end

#csv_input_include_other_columns ⇒ Object 



101
102
103
 
# File 'lib/worker_tools/csv_input.rb', line 101

def csv_input_include_other_columns
  false
end

#csv_input_mapping_order(header_names) ⇒ Object

Compares the first row (header names) with the csv_input_columns hash to find the corresponding positions.

Ex: csv_input_columns: ‘Mandant’, area: ‘Bereich’

headers: ['Bereich', 'Mandant']
=>  { tenant: 1, area: 0}



115
116
117
118
119
 
# File 'lib/worker_tools/csv_input.rb', line 115

def csv_input_mapping_order(header_names)
  return csv_input_columns.map.with_index { |n, i| [n, i] }.to_h if csv_input_columns.is_a?(Array)

  csv_input_mapping_order_for_hash(header_names)
end

#csv_input_mapping_order_for_hash(header_names) ⇒ Object 



121
122
123
124
125
126
127
128
129
130
 
# File 'lib/worker_tools/csv_input.rb', line 121

def csv_input_mapping_order_for_hash(header_names)
  filtered_column_names = header_names.map { |n| csv_input_header_normalized(n) }
  mapping = csv_input_columns.each_with_object({}) do |(k, v), h|
    matchable = v.is_a?(String) ? csv_input_header_normalized(v) : v
    h[k] = filtered_column_names.index { |n| case n when matchable then true end }
  end
  return mapping unless csv_input_include_other_columns

  csv_input_mapping_order_with_other_columns(mapping, filtered_column_names)
end

#csv_input_mapping_order_with_other_columns(mapping, filtered_column_names) ⇒ Object 



132
133
134
135
136
137
138
 
# File 'lib/worker_tools/csv_input.rb', line 132

def csv_input_mapping_order_with_other_columns(mapping, filtered_column_names)
  positions_taken = mapping.values
  filtered_column_names.each_with_index do |header, index|
    mapping[header.to_sym] = index unless positions_taken.include?(index)
  end
  mapping
end

#csv_rows_enum ⇒ Object 



144
145
146
 
# File 'lib/worker_tools/csv_input.rb', line 144

def csv_rows_enum
  @csv_rows_enum ||= CSV.foreach(csv_input_file_path, **csv_input_csv_options)
end

#cvs_input_value_cleanup(value) ⇒ Object

Allows for some basic cleanup of the values, such as applying strip to the strings.



50
51
52
 
# File 'lib/worker_tools/csv_input.rb', line 50

def cvs_input_value_cleanup(value)
  value.is_a?(String) ? value.strip : value
end