Typecaster

This gem was built for create text files based in fixed columns.

Instalation

gem install typecaster

Usage

The Typecaster can be used for two things. To create text files based on positions and read text files based on some descriptor.

Creating

Given you need generate a text with the following format:

column name | size  | starting | ending | type   |
name        | 10    | 0        | 9      | string |
price       | 8     | 10       | 17     | number |
code        | 6     | 18       | 23     | string |

Here's how to use it:

module StringTypecaster
  def self.call(value, options)
    value.to_s.ljust(options[:size], " ")
  end
end

module NumerTypecaster
  def self.call(value, options)
    value.to_s.rjust(options[:size], "0")
  end
end

class ProductFormatter
  include Typecaster

  attribute :code,  :size => 6,  :position => 3, :caster => StringTypecaster
  attribute :name,  :size => 10, :position => 1, :caster => StringTypecaster
  attribute :price, :size => 8,  :position => 2, :caster => NumberTypecaster
end

product = ProductFormatter.new(:name => 'Coca', :price => '25.0', :code => '6312')
puts product.to_s # => 'Coca      000025.06312  '

And you also can group the attributes with common options using with_options method passing a block

class ProductFormatter
  include Typecaster

  with_options :caster => StringTypecaster do
    attribute :code, :size => 6,  :position => 2
    attribute :name, :size => 10, :position => 1
  end

  attribute :price, :size => 8, :position => 3, :caster => NumberTypecaster
end

You can parse a collection with values. This collection can be any object that responds to []. (Hash and ActiveRecord objects are more common). See:

module StringTypecaster
  def self.call(value, options)
    value.to_s.ljust(options[:size], " ")
  end
end

module NumerTypecaster
  def self.call(value, options)
    value.to_s.rjust(options[:size], "0")
  end
end

class ProductFormatter
  include Typecaster

  output_separator ";"

  attribute :code,  :size => 6,  :position => 3, :caster => StringTypecaster
  attribute :name,  :size => 10, :position => 1, :caster => StringTypecaster
  attribute :price, :size => 8,  :position => 2, :caster => NumberTypecaster
end

product = ProductFormatter.new([{:name => 'Coca', :price => '25.0', :code => '6312'}, {:name => 'Coca', :price => '25.0', :code => '6312'}])
puts product.to_s # => 'Coca      ;000025.0;6312  \nCoca      ;000025.0;6312  '

You can change the output separator

module StringTypecaster
  def self.call(value, options)
    value.to_s.ljust(options[:size], " ")
  end
end

module NumerTypecaster
  def self.call(value, options)
    value.to_s.rjust(options[:size], "0")
  end
end

class ProductFormatter
  include Typecaster

  output_separator ";"

  attribute :code,  :size => 6,  :position => 3, :caster => StringTypecaster
  attribute :name,  :size => 10, :position => 1, :caster => StringTypecaster
  attribute :price, :size => 8,  :position => 2, :caster => NumberTypecaster
end

product = ProductFormatter.new(:name => 'Coca', :price => '25.0', :code => '6312')
puts product.to_s # => 'Coca      ;000025.0;6312  '

Reading a uniform file

Given a file with products like that:

Coca      000025.06312  
Pepsi     000023.06313

module StringTypecaster
  def self.call(value, options)
    value.to_s.ljust(options[:size], " ")
  end
end

module NumerTypecaster
  def self.call(value, options)
    value.to_s.rjust(options[:size], "0")
  end
end

class ProductFormatter
  include Typecaster

  attribute :code,  :size => 6,  :position => 3, :caster => StringTypecaster
  attribute :name,  :size => 10, :position => 1, :caster => StringTypecaster
  attribute :price, :size => 8,  :position => 2, :caster => NumberTypecaster
end

products = ProductFormatter.parse_file(File.open('your_file_path', 'r'))
puts products.inspect # => [{:name => 'Coca', :price => '25.0', :code => '6312'}, {:name => 'Pepsi', :price => '23.0', :code => '6313'}]

Reading a file containing different formats

Given a file like that:

0SOME IMPORTANT TEXT 000001
119999FOO BAR        000002
110000XPTO BAR       000003
109901JOAO BAR       000004
939900               000005

Each row in that file was identified by a number 0 for header, 1 for the records inside and 9 for footer, each row has 27 chars, to read then you must implement the parsers for each row type.

module StringTypecaster
  def self.parse(text)
    text.strip
  end
end

module IntegerTypecaster
  def self.parse(text)
    text.to_i
  end
end

class MyFileHeader
  include Typecaster

  with_options :caster => StringTypecaster do
    attribute :identifier, :size => 1,  :position => 1
    attribute :text,       :size => 20, :position => 2
  end

  attribute :sequential, :size => 5, :position => 3, :caster => IntegerTypecaster
end

class MyFileRow
  include Typecaster

  with_options :caster => StringTypecaster do
    attribute :identifier, :size => 1,  :position => 1
    attribute :name,       :size => 15, :position => 3
  end

  with_options :caster => IntegerTypecaster, :size => 5 do
    attribute :amount,     :position => 2
    attribute :sequential, :position => 4
  end
end

class MyFileFooter
  include Typecaster

  with_options :caster => StringTypecaster do
    attribute :identifier, :size => 1,  :position => 1
    attribute :blanks,     :size => 15, :position => 3
  end

  with_options :caster => IntegerTypecaster do
    attribute :total,      :size => 5, :position => 2
    attribute :sequential, :size => 6, :position => 4
  end
end

class MyFileParser
  include Typecaster::Parse

  parser :header, :with => MyFileHeader, :identifier => '0'
  parser :rows,   :with => MyFileRow,    :identifier => '1', :array => true
  parser :footer, :with => MyFileFooter, :identifier => '9'
end

MyFileParser.parse(File.new('my_file.txt'))

Its results in a hash, with keys the following keys: header, rows and footer. The key 'rows' will be an array.

Contributing

Fork the project.
Make your feature addition or bug fix.
Add tests for it.
Commit, do not mess with rakefile, version, or history.
Send me a pull request.