folder_data_sampler 0.2.0

data/lib/folder_data_sampler.rb

@@ -0,0 +1,142 @@
+#/usr/bin/env ruby
+
+require 'find'
+
+# This library samples leave folders within a given folder tree.
+# 
+# It takes as input a root path in the filesystem where to look for data.
+# All the leaves in the folder tree under that root are candidates for selection.
+# 
+# The candidates are then filtered by categories. A category is nothing more
+# than a folder name. Candidates whose path does not contain a folder named 
+# after the filter are discarded. 
+# 
+# For example candidate "/A/B/C/file.x" will be discarded if filter = "AB" but 
+# will be kept if filter = "A" or "B" or "C".
+# Note: filtering is case and white space insensitive.
+# 
+# From the final pool of candidates that passed all filters a random candidate
+# is picked. 
+#
+# For example given this folder tree:
+# 
+# data/photos/cars/renault/megane
+#                 /volkswagen/golf
+#            /drivers/f1/senna
+#                    /rally/sainz
+#                   
+# When calling DataSampler.sample with "data" as root folder and "photos" as
+# filter one of these leaves will be returned:
+# 
+# data/photos/cars/renault/megane
+# data/photos/cars/volkswagen/golf
+# data/photos/drivers/f1/senna
+# data/photos/drivers/rally/sainz
+# 
+# If the filter would be "drivers" then one of these leaves will be returned:
+# 
+# data/photos/drivers/f1/senna
+# data/photos/drivers/rally/sainz
+#                  
+# Options:
+# 
+# pre_filters specify an ordered list of categories (folders) where to search
+#             within the root folder. This basically makes the search faster
+#             when dealing with large folder trees. 
+#             
+#             Example: 
+#             DataSampler.sample "data" pre_filters=["photos", "cars"]
+#             
+#             returns one of:
+#             
+#             data/photos/cars/renault/megane
+#             data/photos/cars/volkswagen/golf
+#             
+#             It's equivalent to:
+#             
+#             DataSampler.sample "data/photos/car"
+#                      
+# Author::    Manuel Pais (mailto:manuelpais@gmail.com)
+# License::   Distributes under the same terms as Ruby       
+
+class DataSampler
+
+  def self.sample(data_source, filters, options = {})
+    sample = pick_sample(filter_data(data_source, filters, options).keys)
+    if sample
+      return "#{data_source}/#{sample}"
+    else
+      return ""
+    end
+  end
+
+  def self.filter_data(data_source, filters, options = {})
+    candidates = find_data(data_source, options)
+    candidates = filter_candidates(filters, candidates)
+  end
+
+  def self.find_data(data_source, options = {})
+    # declare accepted options here, not in middle of function
+    filter_by = options[:pre_filters]
+    return_folder = options[:return_folder]
+    discard_folder = options[:discard_folder]
+    
+    found_data = {}
+    search_within = "." 
+    
+    search_within = filter_by.join("/") if filter_by
+    
+    Dir.chdir(data_source) do
+      Find.find(search_within) do |path|
+        Find.prune if path =~ /\.hg/ # skip mercurial control files
+
+        if File.file?(path)
+          data_item = path.gsub("./", "").rstrip
+          data_item = File.dirname(data_item) if return_folder 
+          found_data[data_item] = categorize(data_item)          
+        end
+      end
+    end
+
+    return found_data
+  end
+  
+  def self.categorize(path)
+    categories = path.split("/") if path
+    return categories
+  end
+  
+  def self.filter_candidates(filters, candidates)
+    unless candidates.nil? or candidates.empty?
+      filters.each do |filter_name, filter_value| 
+        candidates.delete_if {|data_item, categories| not matches(filter_value, categories)}    
+      end
+    end
+    return candidates
+  end
+  
+  def self.matches(filter, categories)
+    categories.each do |category|
+      if spaceless(category).eql?(spaceless(filter))
+        return true
+      end
+    end
+    return false
+  end
+
+  def self.spaceless(sentence)
+    sentence.delete(" ").downcase
+  end
+  
+  def self.pick_sample(candidate_values)
+    unless candidate_values.nil? or candidate_values.empty?
+      if RUBY_VERSION < "1.9"
+        random_candidate = rand(candidate_values.length)
+        return candidate_values[random_candidate]
+      else
+        # The Array#sample function in ruby >= 1.9 does the same as above code
+        return candidate_values.sample         
+      end
+    end
+  end
+end

metadata

@@ -0,0 +1,65 @@
+--- !ruby/object:Gem::Specification 
+name: folder_data_sampler
+version: !ruby/object:Gem::Version 
+  hash: 23
+  prerelease: 
+  segments: 
+  - 0
+  - 2
+  - 0
+  version: 0.2.0
+platform: ruby
+authors: 
+- Manuel Pais
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2012-09-18 00:00:00 Z
+dependencies: []
+
+description: Samples data in a folder structure given a hash with criteria for filtering
+email: manuel.pais@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: []
+
+files: 
+- lib/folder_data_sampler.rb
+homepage: http://rubygems.org/gems/folder_data_sampler
+licenses: []
+
+post_install_message: 
+rdoc_options: []
+
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.7.2
+signing_key: 
+specification_version: 3
+summary: Folder Data Sampler
+test_files: []
+