vicqstream / Philter

Filter ruby arrays is a pleasure, now even more

It helps your life and without weighing too much on global warming

Sometimes it help you to filter some arrays

Filter any kind of array with the power of ruby, with ease and 😄

It's short, dynamic and readable. Its trace is a usefull tool for teachers.

The performance is inversely proportional to the comfort although efficiency is improved in the latest versions. See the section below.

require 'philter'

# You can use everything to filter
# ...single value
[1,2,3].philter 1
=> [1]

# ...array of values
[1,2,3].philter [2,3]
=> [2,3]

# ...operators
[1,2,3].philter '<= 2'
=> [1,2]

[1,2,3].philter '!= 2'
=> [1,3]

# ...range
[1,2,3,4,5].philter 2..4
=> [2, 3, 4]

%w[red green blue].philter 'red'
=> ["red"]

%w[red green blue].philter %w(red blue)
=> ["red", "blue"]

# You can pass a block
[1,2,3].philter([1,2]) { |e| e*2 }
=> [2, 4]

Things get more interesting with array of hashes or objects 😋

people = [{ id: 1, name: 'Mark',  email: '[e-mail]'  },
          { id: 2, name: 'Larry', email: '[e-mail]' },
          { id: 3, name: 'Bill',  email: '[e-mail]'   }
]

people.philter id: 1
=> [{:id=>1, :name=>"Mark", :email=>"[e-mail]"}]

people.philter id: [1,3]
=> [{:id=>1, :name=>"Mark", :email=>"[e-mail]"}, {:id=>3, :name=>"Bill", :email=>"[e-mail]"}]

people.philter id: '>2'
=> [{:id=>3, :name=>"Bill", :email=>"[e-mail]"}]

# Regular expression
people.philter email: /@gmail/
=> [{:id=>1, :name=>"Mark", :email=>"[e-mail]"}, {:id=>2, :name=>"Larry", :email=>"[e-mail]"}]

# Select attributes
people.philter({ email: /@gmail/ }, get: :name)
=> ["Mark", "Larry"]

# Philter with more attributes -and-
people.philter name: /M.+/, email: /@gmail/
=> [{:id=>1, :name=>"Mark", :email=>"[e-mail]"}]

# Philter with more attributes -or-
people.philter({ name: /M.+/, email: /@live/ }, or: true)
=> [{:id=>1, :name=>"Mark", :email=>"[e-mail]"}, {:id=>3, :name=>"Bill", :email=>"[e-mail]"}]

To me the power!

# Select and update attributes
regexp = /gmail/
people.philter(email: regexp) { |e| "#{e[:name]} use #{e[:email].match(regexp)}!"}
=> ["Mark use gmail!", "Larry use gmail!"]

# Add attributes
people.philter(email: /@gmail/) do |e|
  e[:filtered] = true
  e
end
=> :try_yourself

Get the trace with the option debug: true

[1,2,3].philter '<= 2', debug: true
--------------- Start debugging philter 1.1.0 -------------
 Search by String: <= 2 with operator
 item Fixnum 1
 item <= value | 1 <= 2 => x
 item Fixnum 2
 item <= value | 2 <= 2 => x
 item Fixnum 3
 item <= value | 3 <= 2
--------------- End debugging philter 1.1.0 ---------------
 2 item(s) found
=> [1, 2]

people.philter({ name: 'Mark', email: /\A.+gmail/ }, debug: true)
--------------- Start debugging philter 1.1.0 ---------------
 Search by Hash:

 item {:id=>1, :name=>"Mark", :email=>"[e-mail]"} (Hash)
  evaluating with hash
   search: name: Mark (String)
   .4  v1 item[name]  item == value | Mark == Mark  => x
   search: email: (?-mix:\A.+gmail) (Regexp)
   .1  v1 item[email]  item =~ value | [e-mail] =~ (?-mix:\A.+gmail)  => x

 - SELECTED - (And)

 item {:id=>2, :name=>"Bill", :email=>"[e-mail]"} (Hash)
  evaluating with hash
   search: name: Mark (String)
   .4  v1 item[name]  item == value | Bill == Mark
   search: email: (?-mix:\A.+gmail) (Regexp)
   .1  v1 item[email]  item =~ value | [e-mail] =~ (?-mix:\A.+gmail)

 item {:id=>3, :name=>"Larry", :email=>"[e-mail]"} (Hash)
  evaluating with hash
   search: name: Mark (String)
   .4  v1 item[name]  item == value | Larry == Mark
   search: email: (?-mix:\A.+gmail) (Regexp)
   .1  v1 item[email]  item =~ value | [e-mail] =~ (?-mix:\A.+gmail)  => x

--------------- End debugging philter 1.1.0 ---------------
 1 item(s) found
=> [{:id=>1, :name=>"Mark", :email=>"[e-mail]"}]

Rails

Rails return relation objects that must be turned to array

cities = City.all.to_a
  City Load (1.0ms)  SELECT "cities".* FROM "cities"
=> ... [cut]

cities.philter id: 1
=> [#<City id: 1, name: "Milano", code: "MI", region: "Lombardia", created_at: "2016-05-10 09:07:22", updated_at: "2016-05-10 09:07:22">]

cities.philter code: 'PA'
=> [#<City id: 4, name: "Palermo", code: "PA", region: "Sicilia", created_at: "2016-05-10 09:08:13", updated_at: "2016-05-10 09:08:13">]

# Pass a block to select, update or change the result
cities.philter(region: /\Alomb/i) { |city| "#{city.name}-#{city.code}" }
=> ["Milano-MI", "Lecco-LC", "Pavia-PV", "Piacenza-PC", ... [cut]

Performance

Since version 1.0.0 performance are greatly improved! Ruby 2.2.3p173 on windows 7 with i5 3570K Ivy Bridge @4200 Mhz Ram 16Gb 10-10-10-27 2T @686Mhz

require 'benchmark'
require 'philter'

ar_test = 100.times.map { |n| n }
Benchmark.bmbm do |x|
  x.report("philter: ") { 10_000.times { ar_test.philter 1 } }
  x.report("grep: ")    { 10_000.times { ar_test.grep    1 } }
end

#version 1.0.0
                user     system      total        real
philter:    0.031000   0.000000   0.031000 (  0.021759)
grep:       0.016000   0.000000   0.016000 (  0.007115)

#version 0.7.0
                user     system      total        real
philter:    9.204000   0.000000   9.204000 (  9.254443)
grep:       0.062000   0.000000   0.062000 (  0.054257)

range = 1..10
Benchmark.bmbm do |x|
  x.report("philter: ") { 10_000.times { ar_test.philter range  } }
  x.report("grep: ")    { 10_000.times { ar_test.grep    range  } }
end

#version 1.0.0
                user     system      total        real
philter:    0.015000   0.000000   0.015000 (  0.023891)
grep:       0.000000   0.000000   0.000000 (  0.009134)

#version 0.7.0
=> Range was not managed
                user     system      total        real
philter:   91.136000   0.000000  91.136000 ( 91.305855)
grep:       0.172000   0.000000   0.172000 (  0.182490)

ar_search = [1,3,5,7]
Benchmark.bmbm do |x|
  x.report("philter: ") { 10_000.times { ar_test.philter ar_search } }
  x.report("select: ")  { 10_000.times { ar_test.select { |item| ar_search.include?(item) } } }
end

#version 1.0.0
                user     system      total        real
philter:    0.062000   0.000000   0.062000 (  0.052933)
select:     0.031000   0.000000   0.031000 (  0.022178)

#version 0.7.0
                user     system      total        real
philter:   36.176000   0.000000  36.176000 ( 36.182101)
select:     0.078000   0.000000   0.078000 (  0.073341)

Benchmark.bmbm do |x|
  x.report("philter: ") { 1_000.times { ar_test.philter '< 50'            } }
  x.report("select: ")  { 1_000.times { ar_test.select { |item| item < 50 } } }
end

#version 1.0.0
                user     system      total        real
philter:    2.855000   0.000000   2.855000 (  2.851040)
select:     0.015000   0.000000   0.015000 (  0.004312)

#version 0.7.0
                user     system      total        real
philter:    3.744000   0.000000   3.744000 (  3.746851)
select:     0.016000   0.000000   0.016000 (  0.004338)

Strings

require 'benchmark'
require 'philter'
ar_test   = %w(black white grey red green blue yellow orange pink purple violet)
ar_search = %w(red green blue)
Benchmark.bmbm do |x|
  x.report("philter: ") { 10_000.times { ar_test.philter ar_search } }
  x.report("select: ")  { 10_000.times { ar_test.select { |item| ar_search.include? item } } }
end

#version 1.0.0
                user     system      total        real
philter:    0.063000   0.000000   0.063000 (  0.062890)
select:     0.015000   0.000000   0.015000 (  0.011407)

#version 0.7.0
                user     system      total        real
philter:    7.363000   0.000000   7.363000 (  7.359162)
select:     0.000000   0.000000   0.000000 (  0.011539)

ar_test = [ { id: 1, name: 'Mark',  email: '[e-mail]'  },
            { id: 2, name: 'Bill',  email: '[e-mail]'   },
            { id: 3, name: 'Larry', email: '[e-mail]' }]
regexp = /\A.+gmail/
Benchmark.bmbm do |x|
  x.report("philter: ") { 10_000.times { ar_test.philter email: regexp } }
  x.report("select: ")  { 10_000.times { ar_test.select { |item| item[:email] =~ regexp } } }
end

#version 1.0.0
                user     system      total        real
philter:    0.218000   0.000000   0.218000 (  0.221822)
select:     0.000000   0.000000   0.000000 (  0.003418)

#version 0.7.0
                user     system      total        real
philter:    0.468000   0.000000   0.468000 (  0.473782)
select:     0.000000   0.000000   0.000000 (  0.003429)

Compatibility

Ruby 1.9+

Install

gem install philter

To use it in a bundle, add to gem file gem 'philter' and run bundle install

To Do

• Add boolean operator to chain of conditions v1.0.0
• Improve performance keeping the operations's trace v1.0.0
• Add block v1.0.0
• Increase performance further

Contributing

1. Fork it
2. Create your feature branch (git checkout -b my-feature)
3. Commit your changes (git commit -am 'I made extensive use of all my creativity')
4. Push to the branch (git push origin my-feature)
5. Create new Pull Request

Testing

Wide coverage with 38 unit tests and 143 assertions

To test locally install the development requirements

bundle install

Then execute

bundle exec ruby test\unit_test.rb

Performance tests are calibrated to not exceed 1.2 seconds on my pc with a tolerance become 2 seconds:

bundle exec ruby test\performance_test.rb

Loaded suite test/performance_test
Started
........

Finished in 8.505 seconds.
--------------------------------------------------------------------------------

8 tests, 8 assertions, 0 failures, 0 errors, 0 pendings, 0 omissions, 0 notifications
100% passed
--------------------------------------------------------------------------------

0.94 tests/s, 0.94 assertions/s

If you have a very slow pc it could not pass. In this case you can pass a higher tolerance value as argument, for example 3 seconds:

bundle exec ruby test\performance_test.rb 3.0

Found a bug?

Please open an issue.

License

The GNU Lesser General Public License, version 3.0 (LGPL-3.0) See LICENSE file