Fight CSV!

It's 2011, and parsing CSV with Ruby still sucks? Enter FightCSV! It will take the cumbersome out of your CSV parsing, while keeping the awesome! Want some taste of that juicy fresh? Check out this example:

Consider you have a csv file called log_entries.csv which looks like this:

Date,Person,Client/Project,Minutes,Tags,Billable
2011-08-15,John Doe,handsomelabs,60,blogpost,no
2011-08-15,Max Powers,beerbrewing,60,meeting,yes
2011-08-15,Tyler Durden,babysitting,180,"concepting, research",yes
2011-08-15,Hulk Hero,gardening,60,"meeting, research",no
2011-08-15,John Doe,handsomelabs,60,coding,yes
2011-08-08,John Doe,handsomelabs,60,"blabla, meeting",yes

Now you can define a class representing a row of the file. You only need to include FightCSV::Record.

class LogEntry
  include FightCSV::Record
end

But of course you want the values from each row to behave like proper Ruby objects. This can be easily achieved by defining a schema in the LogEntry class:

class LogEntry
  include FightCSV::Record
  schema do
    field "Name"
    field "Client/Project", {
      identifier: :project
    }
  end
end

Now the LogEntry objects will have a name method corresponding to the column called "Name" and a project method corresponding to the column called "Client/Project".

But sometimes you don't only want to adjust the field names, but also the values. In this case FightCSV offers converters. The "Billable" column seems to represent boolean values, so let's tackle that:

class LogEntry
  include FightCSV::Record
  schema do
    field "Name"
    field "Client/Project", {
      identifier: :project
    }

    field "Billable", {
      converter: ->(string) { string == "yes" ? true : false }
    }
  end
end

Often when converting something, we assume that it has a certain format. The "Date" column for example should always be of the format /\d{2}\.\d{2}\.\d{4}/. A validation can easily be added to a column with FightCSV:

class LogEntry
  include FightCSV::Record
  schema do
    field "Name"
    field "Client/Project", {
      identifier: :project
    }

    field "Billable", {
      converter: ->(string) { string == "yes" ? true : false }
    }

    field "Date", {
      validate: /\d{2}\.\d{2}\.\d{4}/,
      converter: ->(string) { Date.parse(string) }
    }
  end
end

The complete schema:

class LogEntry
  include FightCSV::Record
  schema do
    field "Name"
    field "Client/Project", {
      identifier: :project
    }

    field "Billable", {
      converter: ->(string) { string == "yes" ? true : false }
    }

    field "Date", {
      validate: /\d{2}\.\d{2}\.\d{4}/,
      converter: ->(string) { Date.parse(string) }
    }

    field "Tags", {
      converter: ->(string) { string.split(",") }
    }

    field "Minutes", {
      validate: /\d+/,
      converter: ->(string) { string.to_i }
    }
  end
end

With the schema definition you're finally able to parse some CSV. There are two possible ways of doing this:

1. LogEntry.records will return an array with all rows mapped to instances of LogEntry.

2. LogEntry.import will return an enumerator which will pass the same LogEntry instance with the row changed for every iteration.

   LogEntry.import(csv).map(&:minutes).reduce(:+)
   #=> 780

   Doing so you can avoid memory leaks on big csv documents.

Sometimes you may want to parse csv without a header. Instead of defining how the column is called you can specify the number of the column counting from left as an argument to field.

Consider the following CSV:

Ruby,object oriented
Scheme,functional

Now you can define a ProgrammingLanguage class like this:

class ProgrammingLanguage
  include FightCSV::Record


  schema do
    csv_options = { header: false }
    field 1, identifier: :name
    field 2, identifier: :main_paradigm
  end
end

See the examples section for executable versions of these examples.

Use the col_sep, row_sep and quote_char csv_options to costumize the csv format. Consider the following csv document:

Germany EUR/`United States` USD

You can costumize the csv format like so:

class Country
  schema do
     csv_options = { col_sep: " ", row_sep: "/", quote_char: "`" }
     field 1, identifier: :name
     field 2, identifier: :currency
  end
end

• Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet
• Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it
• Fork the project
• Commit and push until you are happy with your contribution
• Make sure to add tests for it. This is important so I don't break it in a future version unintentionally.
• Please try not to mess with the Rakefile, version, or history. If you want to have your own version, or is otherwise necessary, that is fine, but please isolate to its own commit so I can cherry-pick around it.

Only Ruby 1.9 Compatible

Manuel Korfmann

Copyright (c) 2011 Railslove. See LICENSE.txt for further details.