trick_bag 0.33.0 → 0.34.0

data/RELEASE_NOTES.md

@@ -1,3 +1,8 @@
+## v0.34.0
+
+* Added FileLineReader, StartMax.  Added strategies to dos2unix.
+
+
 ## v0.33.0
 
 * Added dos2unix and dos2unix! to Formatters.

data/lib/trick_bag/enumerables/file_line_reader.rb

@@ -0,0 +1,91 @@
+require 'trick_bag/numeric/start_and_max'
+
+module TrickBag
+module Enumerables
+
+# Reads a file containing strings, one on each line, and feeds them
+# in its each() method.  Strips all lines leading and trailing whitespace,
+# and ignores all empty lines and lines beginning with the comment character '#'.
+#
+# If calling each without a block to get an enumerator, call enumerator.close when done
+# if not all values have been exhausted, so that the input file will be closed.
+class FileLineReader
+
+  include Enumerable
+
+  attr_reader :filespec, :start_and_max
+
+  # @param filespec - the file from which to read domain names; blank/empty lines and lines
+  #                   with the first nonblank character == '#' will be ignored.
+  # @param start_pos - the record number, zero offset, at which to begin each() processing
+  # @param max_count - the maximum number of records to be served by each()
+  def initialize(filespec, start_pos = :first, max_count = :infinite)
+    @filespec = filespec
+    @start_and_max = TrickBag::Numeric::StartAndMax.new(start_pos, max_count)
+  end
+
+
+  def line_valid?(line)
+    ! (line.empty? || /^#/ === line)
+  end
+
+
+  def close_file(file)
+    if file && (! file.closed?)
+      file.close
+    end
+  end
+
+
+  # Any enum returned should have a close method to close the file from which lines are read.
+  # Get the enumerator from the superclass, and add a close method to it.
+  def to_enum(file)
+    enumerator = super()
+    enumerator.instance_variable_set(:@file, file)
+
+    def enumerator.close
+      if @file && (! @file.closed?)
+        @file.close
+        @file = nil
+      end
+    end
+
+    enumerator
+  end
+
+
+  # Returns an Enumerator if called without a block; in that case, remember
+  # to close the file explicitly by calling the enumerator's close method
+  # if you finish using it before all values have been exhausted.
+  def each
+
+    file = File.open(filespec, 'r')
+    return to_enum(file) unless block_given?
+
+    valid_record_num = 0
+    yield_count = 0
+
+    begin
+
+      file.each do |line|
+        line.strip!
+        if line_valid?(line)
+          if start_and_max.start_position_reached?(valid_record_num)
+            yield(line)
+            yield_count += 1
+            if start_and_max.max_count_reached?(yield_count)
+              return
+            end
+          end
+
+          valid_record_num += 1
+        end
+      end
+
+    ensure
+      close_file(file)
+    end
+  end
+end
+end
+end

data/lib/trick_bag/formatters/formatters.rb

@@ -65,18 +65,45 @@ module Formatters
   end
 
 
-  # Like the Unix dos2unix command, but on strings rather than files, strips CR characters.
+  # Like the Unix dos2unix command, but on strings rather than files,
+  # strips CR ("\r", 13, 0xD) characters.
+  #
+  # WARNING:
+  #
+  # Currently, two strategies are supported, but they do not account for
+  # character sets that might include characters that have "\r"'s numeric value,
+  # 13, or 0xd, as part of their legitimate values, so we may need to
+  # add strategies to accommodate this.
+  #
+  # An example of a more complex implementation is at:
+  # http://dos2unix.sourcearchive.com/documentation/5.0-1/dos2unix_8c-source.html.
+  #
   # Note: The 'os' gem can be used to determine os.
-  def dos2unix(string)
-    string ? string.gsub("\r", '') : string
+  #
+  # @param string the string to convert
+  # @param strategy the strategy to use for the conversion (note: the default
+  #        may change over time, so if you're sure you want to use the current
+  #        default even if the default changes, don't rely on the default; specify it)
+  def dos2unix(string, strategy = :remove_all_cr)
+    strategies = {
+        remove_all_cr:     -> { string.gsub("\r", '') },
+        remove_cr_in_crlf: -> { string.gsub("\r\n", "\n") }
+    }
+
+    unless strategies.keys.include?(strategy)
+      raise "Unsupported strategy: #{strategy}. Must be one of [#{strategies.keys.sort.join(', ')}]."
+    end
+
+    strategies[strategy].()
   end
 
 
   # Like the Unix dos2unix command, but on strings rather than files, strips CR characters.
   # Modifies the original string.
+  # See warning in dos2unix header.
   # Note: The 'os' gem can be used to determine os.
-  def dos2unix!(string)
-    string ? string.gsub!("\r", '') : string
+  def dos2unix!(string, strategy = :remove_all_cr)
+    string.replace(dos2unix(string, strategy))
   end
 
 end

data/lib/trick_bag/numeric/start_and_max.rb

@@ -0,0 +1,35 @@
+module TrickBag
+module Numeric
+
+  # Like a Range, but includes useful functions with understandable names
+  # that account for the absence of limits.
+  class StartAndMax
+
+  attr_reader :start_pos, :max_count
+
+  # @param start_pos - the record number, zero offset, at which to begin each() processing; or :first
+  # @param max_count - the maximum number of records to be served by each(); or :infinite
+  def initialize(start_pos = :first, max_count = :infinite)
+    @start_pos = ([nil, :first].include?(start_pos) || start_pos <= 0) ? :first : start_pos
+    @max_count = ([nil, :infinite].include?(max_count)     || max_count <= 0) ? :infinite : max_count
+  end
+
+
+  def start_position_reached?(num)
+    @start_pos == :first || num >= @start_pos
+  end
+
+
+  def max_count_reached?(count)
+    @max_count != :infinite && (count >= @max_count)
+  end
+
+
+  def to_s
+    "#{self.class}: start position=#{start_pos}, max count=#{max_count}"
+  end
+
+end
+end
+end
+

data/lib/trick_bag/version.rb

@@ -1,3 +1,3 @@
 module TrickBag
-  VERSION = "0.33.0"
+  VERSION = "0.34.0"
 end

data/spec/trick_bag/enumerables/file_line_reader_spec.rb

@@ -0,0 +1,120 @@
+require 'os'
+require 'tempfile'
+
+require_relative '../../spec_helper'
+require 'trick_bag/enumerables/file_line_reader'
+
+module TrickBag
+module Enumerables
+
+describe FileLineReader do
+
+  TEST_DOMAINS = %w(abc.com cbs.com nbc.com)
+
+  DOMAIN_FILE_CONTENT = \
+"# Comment text followed by a blank line
+
+abc.com
+cbs.com
+
+    # Another comment, but this time with whitespace before it
+nbc.com
+"
+
+    before(:each) do
+      @tempfile = Tempfile.new('file_domain_reader_spec')
+      @tempfile.write DOMAIN_FILE_CONTENT
+      @tempfile.close
+    end
+
+    after(:each) do
+      @tempfile.unlink
+    end
+
+    let(:can_run_lsof?) { OS.posix? && system('which lsof > /dev/null') }
+
+    subject { FileLineReader.new(@tempfile.path) }
+
+
+    it "should create a temporary file" do
+      expect(File.exist?(@tempfile.path)).to be_true
+    end
+
+    it "should produce an array" do
+      expect(subject.to_a).to be_a(Array)
+    end
+
+    it "should produce a nonempty array" do
+      expect(subject.to_a.empty?).to be_false
+    end
+
+    it "should produce a nonempty array containing 'abc.com'" do
+      expect(subject.to_a.include?('abc.com')).to be_true
+    end
+
+    it "should produce an array without blank lines" do
+      expect(subject.to_a.include?('')).to be_false
+    end
+
+    it "should produce an array without beginning comment characters" do
+      expect(subject.to_a.detect { |s| /^#/ === s }).to be_nil
+    end
+
+    it "should produce an array containing TEST_DOMAINS" do
+      expect(subject.to_a).to eq(TEST_DOMAINS)
+    end
+
+    it "should have a working 'each' function" do
+      array = []
+      subject.each { |name| array << name}
+      expect(array).to eq(TEST_DOMAINS)
+    end
+
+    it "should respect a start_pos" do
+      p = FileLineReader.new(@tempfile.path, 1)
+      expect(p.to_a).to eq(TEST_DOMAINS[1..-1])
+    end
+
+    it "should respect a max_count" do
+      p = FileLineReader.new(@tempfile.path, 0, 2)
+      expect(p.to_a).to eq(TEST_DOMAINS[0..1])
+    end
+
+    it 'should be usable as an Enumerable' do
+      e = subject.each
+      expect(e).to be_an(Enumerable)
+      expect(->{ 3.times { e.next} }).not_to raise_error
+      expect(->{ e.next }).to raise_error(StopIteration)
+      e.close
+    end
+
+    # If this test fails, it's probably the test before it that
+    # failed to close its file.
+    it 'closes the file even when its end is not reached' do
+
+      unless can_run_lsof?
+        pending "This test can only be run on a Posix-based OS having the 'lsof' command."
+      end
+
+      get_lsof_lines = -> do
+        output = `lsof -p #{Process.pid} | grep file_domain_reader_spec`
+        output.split("\n")
+      end
+
+      reader = FileLineReader.new(@tempfile.path, 0, 1)
+      reader.each do |line|
+        expect(get_lsof_lines.().size).to eq(1)
+      end
+      expect(get_lsof_lines.().size).to eq(0)
+    end
+
+
+    specify "calling the enumerator's close method multiple times will not raise an error" do
+      enumerator = subject.each
+      expect(-> { 3.times { enumerator.close } }).not_to raise_error
+    end
+  end
+end
+end
+
+

data/spec/trick_bag/formatters/formatters_spec.rb

@@ -86,20 +86,41 @@ describe Formatters do
     specify "CR characters are stripped" do
       expect(Formatters.dos2unix("foo\r\nbar\r\n")).to eq("foo\nbar\n")
     end
+
+    specify "a bad strategy will result in a raised error" do
+      expect(->() { Formatters.dos2unix('', :not_a_strategy) }).to raise_error
+    end
   end
 
   context ".dos2unix!" do
 
-    specify "strings not containing line endings remain unchanged" do
-      expect(Formatters.dos2unix('')).to    eq('')
-      expect(Formatters.dos2unix(' ')).to   eq(' ')
-      expect(Formatters.dos2unix('abc')).to eq('abc')
+    context ':remove_all_cr' do
+      specify "strings not containing line endings remain unchanged" do
+        expect(Formatters.dos2unix('', :remove_all_cr)).to    eq('')
+        expect(Formatters.dos2unix(' ', :remove_all_cr)).to   eq(' ')
+        expect(Formatters.dos2unix('abc', :remove_all_cr)).to eq('abc')
+      end
+
+      specify "CR characters are stripped" do
+        s = "foo\r\nbar\r\n"
+        Formatters.dos2unix!(s, :remove_all_cr)
+        expect(s).to eq("foo\nbar\n")
+      end
     end
 
-    specify "CR characters are stripped" do
-      s = "foo\r\nbar\r\n"
-      Formatters.dos2unix!(s)
-      expect(s).to eq("foo\nbar\n")
+    context 'remove_cr_in_crlf' do
+
+      specify "strings not containing line endings remain unchanged" do
+        expect(Formatters.dos2unix('', :remove_cr_in_crlf)).to    eq('')
+        expect(Formatters.dos2unix(' ', :remove_cr_in_crlf)).to   eq(' ')
+        expect(Formatters.dos2unix('abc', :remove_cr_in_crlf)).to eq('abc')
+      end
+
+      specify "cr is replaced only in crlf" do
+        s = "\rfoo\r\nbar\r\n\r"
+        expect(Formatters.dos2unix(s, :remove_cr_in_crlf)).to eq("\rfoo\nbar\n\r")
+
+      end
     end
   end
 

data/spec/trick_bag/numeric/start_and_max_spec.rb

@@ -0,0 +1,50 @@
+require 'trick_bag/numeric/start_and_max'
+
+module TrickBag
+module Numeric
+
+
+  describe StartAndMax do
+
+    let(:n) { 111 } # arbitrary number
+
+    context '#start_position_reached?' do
+      it 'considers first element to be included when no start position and max are specified' do
+        expect(StartAndMax.new.start_position_reached?(0)).to be_true
+      end
+
+      it 'considers first element to be included when start position == n' do
+        expect(StartAndMax.new(n).start_position_reached?(n)).to be_true
+      end
+
+      it 'considers first element to be included when start position < n' do
+        expect(StartAndMax.new(n).start_position_reached?(n + 1)).to be_true
+      end
+
+      it 'considers first element NOT to be included when start position is specified but n < start' do
+        expect(StartAndMax.new(n).start_position_reached?(n - 1)).to be_false
+      end
+    end
+
+    context '#max_count_reached?' do
+      it 'returns false for a very large number when max is not specified' do
+        expect(StartAndMax.new.max_count_reached?(10 ** 100)).to be_false
+      end
+
+      it 'returns true when when n == max' do
+        expect(StartAndMax.new(0, n).max_count_reached?(n)).to be_true
+      end
+
+      it 'returns false when when n < max' do
+        expect(StartAndMax.new(0, n).max_count_reached?(n - 1)).to be_false
+      end
+
+      it 'returns true when when n > max' do
+        expect(StartAndMax.new(0, n).max_count_reached?(n + 1)).to be_true
+      end
+    end
+  end
+
+
+end
+end

data/trick_bag.gemspec

@@ -18,9 +18,8 @@ Gem::Specification.new do |spec|
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
   spec.require_paths = ["lib"]
 
-  spec.add_dependency 'vrsn-ie-dnsruby'
-
   spec.add_development_dependency "bundler", "~> 1.3"
+  spec.add_development_dependency "os"
   spec.add_development_dependency "rake"
   spec.add_development_dependency "rspec"
   spec.add_development_dependency "guard"

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: trick_bag
 version: !ruby/object:Gem::Version
-  version: 0.33.0
+  version: 0.34.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,40 +9,40 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-03-17 00:00:00.000000000 Z
+date: 2014-03-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
-  name: vrsn-ie-dnsruby
+  name: bundler
   requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
-    - - ! '>='
+    - - ~>
       - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
+        version: '1.3'
+  type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     none: false
     requirements:
-    - - ! '>='
+    - - ~>
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '1.3'
 - !ruby/object:Gem::Dependency
-  name: bundler
+  name: os
   requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
-    - - ~>
+    - - ! '>='
       - !ruby/object:Gem::Version
-        version: '1.3'
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     none: false
     requirements:
-    - - ~>
+    - - ! '>='
       - !ruby/object:Gem::Version
-        version: '1.3'
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
@@ -126,12 +126,14 @@ files:
 - lib/trick_bag/enumerables/buffered_enumerable.rb
 - lib/trick_bag/enumerables/compound_enumerable.rb
 - lib/trick_bag/enumerables/endless_last_enumerable.rb
+- lib/trick_bag/enumerables/file_line_reader.rb
 - lib/trick_bag/enumerables/filtered_enumerable.rb
 - lib/trick_bag/formatters/formatters.rb
 - lib/trick_bag/io/temp_files.rb
 - lib/trick_bag/io/text_mode_status_updater.rb
 - lib/trick_bag/meta/classes.rb
 - lib/trick_bag/numeric/multi_counter.rb
+- lib/trick_bag/numeric/start_and_max.rb
 - lib/trick_bag/numeric/totals.rb
 - lib/trick_bag/operators/operators.rb
 - lib/trick_bag/timing/timing.rb
@@ -143,12 +145,14 @@ files:
 - spec/trick_bag/enumerables/buffered_enumerable_spec.rb
 - spec/trick_bag/enumerables/compound_enumerable_spec.rb
 - spec/trick_bag/enumerables/endless_last_enumerable_spec.rb
+- spec/trick_bag/enumerables/file_line_reader_spec.rb
 - spec/trick_bag/enumerables/filtered_enumerable_spec.rb
 - spec/trick_bag/formatters/formatters_spec.rb
 - spec/trick_bag/io/temp_files_spec.rb
 - spec/trick_bag/io/text_mode_status_updater_spec.rb
 - spec/trick_bag/meta/classes_spec.rb
 - spec/trick_bag/numeric/multi_counter_spec.rb
+- spec/trick_bag/numeric/start_and_max_spec.rb
 - spec/trick_bag/numeric/totals_spec.rb
 - spec/trick_bag/operators/operators_spec.rb
 - spec/trick_bag/timing/timing_spec.rb
@@ -186,12 +190,14 @@ test_files:
 - spec/trick_bag/enumerables/buffered_enumerable_spec.rb
 - spec/trick_bag/enumerables/compound_enumerable_spec.rb
 - spec/trick_bag/enumerables/endless_last_enumerable_spec.rb
+- spec/trick_bag/enumerables/file_line_reader_spec.rb
 - spec/trick_bag/enumerables/filtered_enumerable_spec.rb
 - spec/trick_bag/formatters/formatters_spec.rb
 - spec/trick_bag/io/temp_files_spec.rb
 - spec/trick_bag/io/text_mode_status_updater_spec.rb
 - spec/trick_bag/meta/classes_spec.rb
 - spec/trick_bag/numeric/multi_counter_spec.rb
+- spec/trick_bag/numeric/start_and_max_spec.rb
 - spec/trick_bag/numeric/totals_spec.rb
 - spec/trick_bag/operators/operators_spec.rb
 - spec/trick_bag/timing/timing_spec.rb