mrkplt/csv_find

CSV Find v1.0.0

Let's just get this out of the way: this gem is 60% an academic exercise in quickly navigating through a CSV without actually using the Ruby CSV library for much of it. How do we do that? We shell out to the system and use some tools to chomp the file down to the pieces we are looking for. It's actually pretty fast at finding arbitrary records in even very large CSV files without being crazy memory intensive. That's probably why I started writing this in the first place.

I have used this code in production, you can too, most of the time you probably want a solution backed by the CSV library. I don't think I have done any benchmarking against it. You can see how to do some of this in a more normal way over here:

https://stackoverflow.com/questions/9599568/how-to-find-a-specific-row-in-csv/33849136

Usage

Install

gem install csv_find

Add it to a class.

class People
  include CsvFind
  csv_file('path/to/people.csv', {optional: 'csv_configurations'})
end

Now you have a setters and getters based on the headers of the csv file.

Methods

Yourclass.new
Yourclass.find(line_number)
Yourclass.where(header1: 'value', header2: 'value')
Yourclass.first
Yourclass.last
Yourclass.each { |a| a.method  }

Plus, the Enumerable module is implemented!

License

CSV Find is Copyright © 2020 Mark Platt, Inc. It is free software, and may be redistributed under the terms specified in the MIT-LICENSE file.

csv_find's Issues

to_hash

Fix gemspec description and readme.

Generate caching/ review disk access.

Update #each to work off #all or a passed enumerable.

each is slow, we could make it a lot faster at the cost of memory. Maybe it should have a flag to not load into memory.

Private #new and #initialize

Make attr_accessor read only if they aren't

attribute_names, attributes

add those methods.

has_many relationships