masticate 0.5.1 → 0.6.0

data/bin/masticate

@@ -2,7 +2,4 @@
 
 require_relative "../lib/masticate"
 
-masticator = Masticate::MyOptionParser.new
-command, options, filenames = masticator.parse
-
-masticator.execute(command, options, filenames)
+Masticate::Command.start

data/lib/masticate.rb

@@ -1,7 +1,7 @@
 require "open-uri"
 require "csv"
 
-%w{version base myoptparse sniffer mender csvify plucker datify gsubber max_rows concat relabel exclude transform include cook}.each do |f|
+%w{version base sniffer mender csvify plucker datify gsubber max_rows concat relabel exclude transform include cook command myoptparse}.each do |f|
   require_relative "masticate/#{f}"
 end
 

data/lib/masticate/command.rb

@@ -0,0 +1,153 @@
+require 'thor'
+require 'pp'
+
+class Masticate::Command < Thor
+
+  no_tasks do
+    def params
+      opts = options.dup
+      col_sep = opts[:delim]
+      col_sep = "TAB" if col_sep == "\t"
+      opts[:col_sep] = col_sep
+      opts
+    end
+
+    def execute(filename = nil)
+      method = caller[0][/`.*'/][1..-2]
+      results = Masticate.send(method, filename, params)
+      logmessage(method, params, results)
+    end
+
+    def logmessage(command, options, results)
+      $stderr.puts <<-EOT
+  * masticate #{command} (#{options.keys.join(', ')})
+    Lines in input: #{results[:input_count]}
+    Lines in output: #{results[:output_count]}
+    Headers: #{results[:headers].inspect}
+  EOT
+      if results[:field_counts]
+        $stderr.puts "  Field counts: #{results[:field_counts].inspect}"
+      end
+    end
+  end
+
+  desc "sniff", "determine structure of source file"
+  method_option :stats => :boolean, :desc => "Read entire file, report distribution of field counts"
+  def sniff
+    results = Masticate.sniff(filename, params)
+    col_sep = results[:col_sep]
+    col_sep = "TAB" if col_sep == "\t"
+    quote_char = results[:quote_char] || "NONE"
+    $stderr.puts <<-EOT
+Processing complete.
+  Input delimiter: #{col_sep}
+  Quote char: #{quote_char}
+  Field counts: #{results[:field_counts].inspect}
+  Headers: #{results[:headers].join(',')}
+EOT
+  end
+
+  desc "mend", "repair source file format"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :inlined => :boolean, :desc => "Source records repeat header fields on every line"
+  method_option :snip => :numeric, :desc => "Ignore first N fields of each row"
+  method_option :dejunk => :boolean, :desc => "Ignore short lines"
+  method_option :buried => :numeric, :desc => "Remove embedded delimiters in field N"
+  def mend(filename = nil)
+    results = Masticate.mend(filename, params)
+    logmessage(__method__, params, results)
+  end
+
+  desc "csvify", "convert input to standard CSV"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  def csvify(filename = nil)
+    execute(filename)
+  end
+
+  desc 'pluck', "ignore all but the specified columns"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :fields, :required => true, :desc => "field names to extract"
+  def pluck(filename = nil)
+    execute(filename)
+  end
+
+  desc "datify", "parsed named field as formatted time/datestamp"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :field, :required => true, :desc => "Fieldname to interpret as a date/time"
+  method_option :format, :required => true, :desc => "strptime format string"
+  def datify(filename = nil)
+    execute(filename)
+  end
+
+  desc "gsub", "applied substitution rule to named field"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :field, :required => true, :desc => "field to convert"
+  method_option :from, :required => true, :desc => "regexp to apply to original value"
+  method_option :to, :required => true, :desc => "string to replace capture with"
+  def gsub(filename = nil)
+    execute(filename)
+  end
+
+  desc "maxrows", "compute SUM(max-field) GROUP BY(by-field)"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :max, :required => true, :desc => "field to sum"
+  method_option :by, :required => true, :desc => "field to aggregate over"
+  def maxrows(filename = nil)
+    execute(filename)
+  end
+
+  desc "concat", "concatenate multiple input files, ignoring header lines on all but first file"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  def concat(*args)
+    Masticate.concat(args, params)
+  end
+
+  desc "relabel", "replace header line in output"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :fields, :required => true, :desc => "list of field names to use in output"
+  def relabel(filename = nil)
+    Masticate.relabel(filename, params)
+  end
+
+  desc "exclude", "ignore input lines that match criteria"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :field, :required => true, :desc => "field to check for exclusion"
+  method_option :value, :required => true, :desc => "value to compare with for exclusion"
+  def exclude(filename = nil)
+    execute(filename)
+  end
+
+  desc "include", "ignore all input lines *except* those that match criteria"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :field, :required => true, :desc => "field to check for inclusion"
+  method_option :value, :required => true, :desc => "value to compare with for inclusion"
+  def include(filename = nil)
+    execute(filename)
+  end
+
+  desc "transform", "apply transformation rule to named field"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :rule, :required => true, :desc => "valid values are {upcase, downcase}"
+  def transform(filename = nil)
+    execute(filename)
+  end
+
+  desc "cook", "apply conversion recipe to input records"
+  method_option :delim, :desc => "field delimiter (default is comma)"
+  method_option :quote, :desc => "delimiter escape string (default is double-quotes when delim is comma, otherwise none)"
+  method_option :recipe, :required => true, :desc => "filename containing recipe"
+  def cook(filename = nil)
+    execute(filename)
+  end
+end

data/lib/masticate/version.rb

@@ -1,3 +1,3 @@
 module Masticate
-  VERSION = "0.5.1"
+  VERSION = "0.6.0"
 end

data/spec/spec_helper.rb

@@ -1,35 +1,4 @@
-# require File.expand_path("../../config/environment", __FILE__)
-# require 'rspec/rails'
 require 'rspec/autorun'
-# require 'capybara/rspec'
 require "tempfile"
 
-# Requires supporting ruby files with custom matchers and macros, etc,
-# in spec/support/ and its subdirectories.
-# Dir[Rails.root.join("spec/support/**/*.rb")].each {|f| require f}
-
 require File.expand_path('../../lib/masticate', __FILE__)
-
-RSpec.configure do |config|
-  # ## Mock Framework
-  #
-  # If you prefer to use mocha, flexmock or RR, uncomment the appropriate line:
-  #
-  # config.mock_with :mocha
-  # config.mock_with :flexmock
-  # config.mock_with :rr
-
-  # Remove this line if you're not using ActiveRecord or ActiveRecord fixtures
-  # config.fixture_path = "#{::Rails.root}/spec/fixtures"
-
-  # If you're not using ActiveRecord, or you'd prefer not to run each of your
-  # examples within a transaction, remove the following line or assign false
-  # instead of true.
-  # config.use_transactional_fixtures = true
-  # config.use_instantiated_fixtures  = false
-
-  # If true, the base class of anonymous controllers will be inferred
-  # automatically. This will be the default behavior in future versions of
-  # rspec-rails.
-  # config.infer_base_class_for_anonymous_controllers = false
-end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: masticate
 version: !ruby/object:Gem::Version
-  version: 0.5.1
+  version: 0.6.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-10-03 00:00:00.000000000 Z
+date: 2012-11-14 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -93,6 +93,7 @@ files:
 - bin/masticate
 - lib/masticate.rb
 - lib/masticate/base.rb
+- lib/masticate/command.rb
 - lib/masticate/concat.rb
 - lib/masticate/cook.rb
 - lib/masticate/csvify.rb
@@ -172,7 +173,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 89441162680926451
+      hash: 2907326929962451974
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:
@@ -181,7 +182,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 89441162680926451
+      hash: 2907326929962451974
 requirements: []
 rubyforge_project: masticate
 rubygems_version: 1.8.24