csv_madness 0.0.3 → 0.0.4

data/CHANGELOG.markdown

@@ -0,0 +1,17 @@
+0.0.4
+=====
+
+* Fixed serious bug for default column names.  record.middle_name instead of record.middlename
+* Spreadsheet can be re-read from file by calling @sheet.reload_spreadsheet(opts)
+
+
+0.0.3
+=====
+
+* Feature:  can add and drop columns from spreadsheets
+
+
+0.0.2
+=====
+
+* Aw, hell.  I don't remember.

data/Gemfile

@@ -5,6 +5,9 @@ source "http://rubygems.org"
 
 # Add dependencies to develop your gem here.
 # Include everything needed to run rake, tests, features, etc.
+
+gem "fun_with_files"
+
 group :development do
   gem "shoulda", ">= 0"
   gem "rdoc", "~> 3.12"

data/README.rdoc

@@ -121,9 +121,9 @@ sheet.write_to_file( "~/data/people.clean.csv", force_quotes: true ) # save the
 You could do something similar to clean and standardize phone numbers, detect and delete/complete invalid emails, etc.
 
 
-=== Adding and removing columns
+=== Adding, removing, and renaming columns ===
 
-# Add 72 years to the date born.  
+# Add 72 years to the date born.
 sheet.set_column_type( :born, :date )  # replace date strings with Time objects
 
 sheet.add_column( :expected_death_date ) do |date, record|
@@ -135,8 +135,10 @@ puts sheet[0].expected_death_date     # should be in 2058
 # But that's just morbid, so we drop the column
 sheet.drop_column( :expected_death_date )
 
+# Or, if you think you need the information, but need to be a bit more euphemistic about it
+sheet.rename_column( :expected_death_date, :expiration_date )
 
-=== Using columns
+=== Using columns ===
 
 sheet.set_column_type( :id, :integer )
 

data/Rakefile

@@ -33,9 +33,11 @@ Jeweler::Tasks.new do |gem|
                 "./VERSION",
                 "./README.rdoc",
                 "./Rakefile",
+                "./CHANGELOG.markdown",
                 "./test/csv/simple.csv",
                 "./test/helper.rb",
-                "./test/test_csv_madness.rb" ]
+                "./test/test_csv_madness.rb",
+                "./test/test_sheet.rb" ]
 end
 
 Jeweler::RubygemsDotOrgTasks.new

data/VERSION

@@ -1 +1 @@
-0.0.3
+0.0.4

data/lib/csv_madness/record.rb

@@ -13,11 +13,21 @@ module CsvMadness
     end
     
     def [] key
-      @csv_data[key] 
+      case key
+      when String, Integer
+        @csv_data[key] 
+      when Symbol
+        @csv_data[key.to_s]
+      end
     end
     
     def []= key, val
-      @csv_data[key] = val
+      case key
+      when String, Integer
+        @csv_data[key] = val
+      when Symbol
+        @csv_data[key.to_s] = val
+      end
     end
     
     def columns
@@ -37,7 +47,7 @@ module CsvMadness
     end
     
     def blank?( col )
-      (self.send( col ).to_s || "").strip.length == 0
+      (self.send( col.to_sym ).to_s || "").strip.length == 0
     end
   end
 end

data/lib/csv_madness/sheet.rb

@@ -30,7 +30,8 @@ module CsvMadness
     # Used to make getter/setter names out of the original header strings.
     # " hello;: world! " => :hello_world
     def self.getter_name( name )
-      name = name.strip.gsub(/\s+/,"_").gsub(/(\W|_)+/, "" ).downcase
+      name = name.strip.gsub( /\s+/, "_" ).gsub( /(\W|_)+/, "_" ).downcase
+      name = name.gsub( /_+$/, "" )
       if name.match( /^\d/ )
         name = "_#{name}"
       end
@@ -43,9 +44,17 @@ module CsvMadness
     def self.add_search_path( path )
       @search_paths ||= []
       path = Pathname.new( path ).expand_path
+      unless path.directory?
+        raise "The given path does not exist"
+      end
+      
       @search_paths << path unless @search_paths.include?( path )
     end
     
+    def self.search_paths
+      @search_paths
+    end
+    
     def self.from( csv_file, opts = {} )
       if f = find_spreadsheet_in_filesystem( csv_file )
         Sheet.new( f, opts )
@@ -83,7 +92,7 @@ module CsvMadness
     end
     
     def self.write_to_file( spreadsheet, file, opts = {} )
-      file = Pathname.new(file).expand_path
+      file = file.fwf_filepath.expand_path
       File.open( file, "w" ) do |f|
         f << spreadsheet.to_csv( opts )
       end
@@ -99,7 +108,7 @@ module CsvMadness
       end
     end
     
-    attr_reader :columns, :records, :spreadsheet_file, :record_class
+    attr_reader :columns, :index_columns, :records, :spreadsheet_file, :record_class
     # opts: 
     #   index: ( [:id, :id2 ] )
     #       columns you want mapped for quick 
@@ -115,28 +124,37 @@ module CsvMadness
     #   header:   false       
     #       anything else, we assume the csv file has a header row
     def initialize( spreadsheet, opts = {} )
-      @spreadsheet_file = self.class.find_spreadsheet_in_filesystem( spreadsheet )
+      if spreadsheet.is_a?(Array)
+        @spreadsheet_file = nil
+        
+      else
+        @spreadsheet_file = self.class.find_spreadsheet_in_filesystem( spreadsheet )
+      end
       @opts = opts
       @opts[:header] = (@opts[:header] == false ? false : true)  # true unless already explicitly set to false
       
+      reload_spreadsheet
+    end
+    
+    def reload_spreadsheet( opts = @opts )
       load_csv
-      
-      set_initial_columns( @opts[:columns] )
-      
-
+      set_initial_columns( opts[:columns] )
       create_record_class
       package
       
-      @index_columns = case @opts[:index]
+      set_index_columns( opts[:index] )
+      reindex
+    end
+
+    def set_index_columns( index_columns )
+      @index_columns = case index_columns
       when NilClass
         []
       when Symbol
-        [ @opts[:index] ]
+        [ index_columns ]
       when Array
-        @opts[:index]
+        index_columns
       end
-        
-      reindex
     end
     
     def [] offset
@@ -196,8 +214,9 @@ module CsvMadness
       end
     end
 
-    # if column doesn't exist, silently fails.  Proper behavior?  Dunno.
     def alter_column( column, blank = :undefined, &block )
+      raise "Column does not exist: #{column}" unless @columns.include?( column )
+      
       if cindex = @columns.index( column )
         for record in @records
           if record.blank?(column) && blank != :undefined
@@ -209,8 +228,9 @@ module CsvMadness
       end
     end
     
+    # If no block given, adds an empty column
     def add_column( column, &block )
-      raise "Column already exists" if @columns.include?( column )
+      raise "Column already exists: #{column}" if @columns.include?( column )
       @columns << column
       
       # add empty column to each row
@@ -241,10 +261,45 @@ module CsvMadness
       update_data_accessor_module
     end
     
+    def rename_column( column, new_name )
+      @columns[@columns.index(column)] = new_name
+      rename_index_column( column, new_name ) if @index_columns.include?( column )
+      update_data_accessor_module
+    end
+    
     def set_column_type( column, type, blank = :undefined )
       alter_column( column, blank, &COLUMN_TYPES[type] )
     end
 
+    
+    # If :reverse_merge is true, then the dest column is only overwritten for records where :dest is blank
+    def merge_columns( source, dest, opts = {} )
+      opts = { :drop_source => true, :reverse_merge => false, :default => "" }.merge( opts )
+      column_must_exist( source, dest )
+      
+      self.records.each do |record|
+        if opts[:reverse_merge] == false || record.blank?( dest )
+          record[dest] = record.blank?(source) ? opts[:default] : record[source]
+        end
+      end
+      
+      self.drop_column( source ) if opts[:drop_source]
+    end
+    
+    # By default, the
+    def concat_columns( col1, col2, opts = {} )
+      opts =  {:separator => '', :out => col1}.merge( opts )
+      
+      column_must_exist( col1, col2 )
+      self.add_column( opts[:out] ) unless self.columns.include?( opts[:out] )
+      
+      for record in self.records
+        record[ opts[:out] ] = "#{record[col1]}#{opts[:separator]}#{record[col2]}"
+      end
+    end
+    
+    alias :concatenate :concat_columns
+    
     # Note: If a block is given, the mod arg will be ignored.
     def add_record_methods( mod = nil, &block )
       if block_given?
@@ -285,6 +340,13 @@ module CsvMadness
       end
     end
     
+    # shouldn't require reindex
+    def rename_index_column( column, new_name )
+      @index_columns[ @index_columns.index( column ) ] = new_name
+      @indexes[new_name] = @indexes[column]
+      @indexes.delete(column)
+    end
+    
     # Each spreadsheet has its own anonymous record class, and each CSV row instantiates
     # a record of this class.  This is where the getters and setters come from.
     def create_record_class
@@ -374,5 +436,11 @@ module CsvMadness
     def update_data_accessor_module
       @module.remap_accessors( columns_to_mapping )
     end
+  
+    def column_must_exist( *cols )
+      for col in cols
+        raise ArgumentError.new( "#{caller[0]}: column :#{col} does not exist.") unless self.columns.include?(col)
+      end
+    end
   end
 end

data/lib/csv_madness.rb

@@ -1,5 +1,5 @@
 require 'csv'
-require 'pathname'
+require 'fun_with_files'
 require 'time'        # to use Time.parse to parse cells to get the date
 require 'debugger'
 
@@ -7,6 +7,9 @@ require_relative 'csv_madness/data_accessor_module'
 require_relative 'csv_madness/sheet'
 require_relative 'csv_madness/record'
 
+
+FunWith::Files::RootPath.rootify( CsvMadness, __FILE__.fwf_filepath.dirname.up )
+
 CsvMadness.class_eval do
   def self.load( csv, opts = {} )
     CsvMadness::Sheet.from( csv, opts )

data/test/helper.rb

@@ -19,3 +19,85 @@ require 'csv_madness'
 
 class Test::Unit::TestCase
 end
+
+class MadTestCase < Test::Unit::TestCase
+  MARY_ID = "1"
+  BILL_ID = "2"
+  DARWIN_ID = "3"
+  CHUCK_ID = "4"
+  
+  def load_mary
+    id = @simple.index_columns.first
+    @mary = @simple.fetch( id, MARY_ID )
+  end
+  
+  def load_bill
+    id = @simple.index_columns.first
+    @bill = @simple.fetch( id, BILL_ID )
+  end
+  
+  def load_darwin
+    id = @simple.index_columns.first
+    @darwin = @simple.fetch( id, DARWIN_ID )
+  end
+  
+  def load_chuck
+    id = @simple.index_columns.first
+    @chuck = @simple.fetch( id, CHUCK_ID )
+  end
+     
+  def set_person_records
+    load_mary
+    load_darwin
+    load_bill
+    load_chuck
+  end
+  
+  def muck_up_spreadsheet
+    @simple.add_column(:scrambled_name) do |val, record|
+      record.fname.chars.map(&:to_s).zip( record.lname.chars.map(&:to_s) ).flatten.compact.join
+    end
+    
+    @simple.alter_column(:id) do |val|
+      (val.to_i << 8) % 27
+    end
+    
+    @simple.set_column_type(:id, :float)
+    @simple.set_column_type(:born, :date)
+    
+    @simple.alter_column(:born) do |val|
+      if val.is_a?(String)
+        -1.0
+      else
+        (Time.now - val).to_f
+      end
+    end
+
+    @simple.drop_column(:fname)
+    @simple.drop_column(:lname)
+  end
+   
+  def set_spreadsheet_paths
+    @csv_search_path = Pathname.new( __FILE__ ).dirname.join("csv")
+    @csv_output_path = @csv_search_path.join("out")
+    CsvMadness::Sheet.add_search_path( @csv_search_path )
+    CsvMadness::Sheet.add_search_path( @csv_output_path )
+  end
+  
+  def load_simple_spreadsheet
+    @simple = CsvMadness.load( "simple.csv", index: [:id] )
+  end
+  
+  def unload_simple_spreadsheet
+    @simple = nil
+  end
+  
+  def empty_output_folder
+    if defined?(FileUtils)
+      FileUtils.rm_rf( Dir.glob( @csv_output_path.join("**","*") ) )
+    else
+      puts "fileutils not defined"
+      `rm -rf #{@csv_output_path.join('*')}`
+    end
+  end
+end

data/test/test_csv_madness.rb

@@ -1,26 +1,21 @@
 require 'helper'
 
-class TestCsvMadness < Test::Unit::TestCase
+class TestCsvMadness < MadTestCase
   context "all:" do
     setup do
-      @csv_search_path = Pathname.new( __FILE__ ).dirname.join("csv")
-      @csv_output_path = @csv_search_path.join("out")
-      CsvMadness::Sheet.add_search_path( @csv_search_path )
+      set_spreadsheet_paths
+      load_simple_spreadsheet
     end
 
     teardown do
-      if defined?(FileUtils)
-        FileUtils.rm_rf( Dir.glob( @csv_output_path.join("**","*") ) )
-      else
-        puts "fileutils not defined"
-        `rm -rf #{@csv_output_path.join('*')}`
-      end
+      empty_output_folder
     end
     
     context "testing sheet basics" do
       should "not accept duplicate search paths" do
+        @path_count = CsvMadness::Sheet.search_paths.length
         CsvMadness::Sheet.add_search_path( Pathname.new( __FILE__ ).dirname.join("csv") )
-        assert_equal 1, CsvMadness::Sheet.instance_variable_get("@search_paths").length
+        assert_equal @path_count, CsvMadness::Sheet.search_paths.length
       end
     
       should "load a simple spreadsheet" do
@@ -49,11 +44,11 @@ class TestCsvMadness < Test::Unit::TestCase
     context "testing transformations" do
       context "with a simple spreadsheet loaded" do
         setup do
-          @simple = CsvMadness.load( "simple.csv", index: [:id] )
+          load_simple_spreadsheet
         end
       
         teardown do
-          @simple = nil
+          unload_simple_spreadsheet
         end
       
         should "transform every cell" do
@@ -154,6 +149,49 @@ class TestCsvMadness < Test::Unit::TestCase
           assert_equal records.length, records.compact.length
           assert_equal 2, records.length
           assert_equal "1", records.first.id
+        end 
+        
+        should "rename columns" do
+          load_mary
+          
+          assert_equal "Mary", @mary.fname
+          assert_equal "Moore", @mary.lname
+          assert_equal "Mary", @mary[1]
+          assert_equal "Moore", @mary[2]
+          
+          @simple.rename_column( :fname, :first_name )
+          @simple.rename_column( :lname, :last_name )
+          
+          assert_equal "Mary", @mary.first_name
+          assert_equal "Moore", @mary.last_name
+          assert_equal "Mary", @mary[1]
+          assert_equal "Moore", @mary[2]
+        end
+        
+        should "rename an index column" do
+          @simple.rename_column( :id, :identifier )
+          @mary = @simple.fetch( :identifier, "1" )
+          assert_equal "1", @mary.identifier
+        end
+        
+        should "rename an index column and ensure that the outputted spreadsheet has the new column name" do
+          @simple.rename_column( :id, :identifier )
+          @outfile = @csv_output_path.join("output.csv")
+          @simple.write_to_file( @outfile, force_quotes: true )
+
+          @simple = CsvMadness.load( @outfile, index: [:identifier] )
+          
+          load_mary
+          assert_equal "Mary", @mary.fname
+        end
+        
+        should "filter! records" do
+          @simple.filter! do |record|
+            record.id == BILL_ID || record.id == CHUCK_ID
+          end
+          
+          assert_equal 2, @simple.records.length
+          
         end
       end
     end
@@ -194,7 +232,7 @@ class TestCsvMadness < Test::Unit::TestCase
     context "testing add/remove column transformations" do
       context "with simple spreadsheet loaded" do
         setup do
-          load_simple
+          load_simple_spreadsheet
         end
         
         should "add column" do
@@ -221,12 +259,4 @@ class TestCsvMadness < Test::Unit::TestCase
       end
     end
   end
-  
-  def load_simple
-    @simple = CsvMadness.load( "simple.csv", index: [:id] )
-  end
-  
-  def load_mary
-    @mary = @simple.fetch( :id, "1" )
-  end
 end

data/test/test_sheet.rb

@@ -0,0 +1,24 @@
+require 'helper'
+
+class TestCsvMadness < MadTestCase
+  context "testing getter_name()" do
+    should "return proper function names" do
+      assert_equal :hello_world, CsvMadness::Sheet.getter_name( "  heLLo __ world " )
+      assert_equal :_0_hello_world, CsvMadness::Sheet.getter_name( "0  heLLo __ worlD!!! " )
+    end
+  end
+  
+  context "testing default spreadsheet paths" do
+    should "only load existing paths" do
+      assert_raise(RuntimeError) do
+        CsvMadness::Sheet.add_search_path( CsvMadness.root.join("rocaganthor") )
+      end
+    end
+    
+    should "check a search path for files to load" do
+      CsvMadness::Sheet.add_search_path( CsvMadness.root.join("test", "csv") )
+      sheet = CsvMadness.load( "with_nils.csv" )
+      assert sheet.is_a?(CsvMadness::Sheet)
+    end
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: csv_madness
 version: !ruby/object:Gem::Version
-  version: 0.0.3
+  version: 0.0.4
   prerelease: 
 platform: ruby
 authors:
@@ -9,8 +9,24 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-06-03 00:00:00.000000000 Z
+date: 2013-11-07 00:00:00.000000000 Z
 dependencies:
+- !ruby/object:Gem::Dependency
+  name: fun_with_files
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: shoulda
   requirement: !ruby/object:Gem::Requirement
@@ -100,6 +116,7 @@ extra_rdoc_files:
 - LICENSE.txt
 - README.rdoc
 files:
+- ./CHANGELOG.markdown
 - ./Gemfile
 - ./README.rdoc
 - ./Rakefile
@@ -111,6 +128,7 @@ files:
 - ./test/csv/simple.csv
 - ./test/helper.rb
 - ./test/test_csv_madness.rb
+- ./test/test_sheet.rb
 - LICENSE.txt
 - README.rdoc
 homepage: http://github.com/darthschmoo/csv_madness
@@ -128,7 +146,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 2889880782640461400
+      hash: 1185350757253474800
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements: