parseinput 0.0.1

data/test/tc_reading.rb

@@ -0,0 +1,70 @@
+#!/usr/local/bin/ruby -w
+
+# tc_reading.rb
+#
+#  Created by James Edward Gray II on 2005-08-14.
+#  Copyright 2005 Gray Productions. All rights reserved.
+
+require "test/unit"
+
+require "parse/input"
+
+class TestReading < Test::Unit::TestCase
+	def test_line_by_line_read
+		path  = File.join(File.dirname(__FILE__), "five_lines.txt")
+		lines = File.readlines(path)
+		test  = self
+		
+		Parse::Input.new(path) do
+			read { |line| test.assert_equal(lines.shift, line) }
+		end
+	end
+	
+	def test_simple_interface
+		path  = File.join(File.dirname(__FILE__), "five_lines.txt")
+		lines = File.readlines(path)
+		test  = self
+
+		input(path) do
+			read { |line| test.assert_equal(lines.shift, line) }
+		end
+	end
+
+	def test_paragraph_read
+		path       = File.join(File.dirname(__FILE__), "five_paragraphs.txt")
+		paragraphs = File.readlines(path, "")
+		test       = self
+
+		input(path, "") do
+			read { |paragraph| test.assert_equal(paragraphs.shift, paragraph) }
+		end
+	end
+
+	def test_restricted_reading
+		path    = File.join(File.dirname(__FILE__), "five_lines.txt")
+		numbers = %w{two three}
+		test    = self
+
+		input(path) do
+			read(/ (t\w+)\./) do |number|
+				test.assert_equal("This is line #{numbers.first}.\n", @read)
+				test.assert_equal(numbers.shift, number)
+			end
+		end
+	end
+	
+	def test_data_saving_and_retrieving
+		path    = File.join(File.dirname(__FILE__), "five_lines.txt")
+
+		data = input path do
+			read(/ (o\w+)\./) { |number| @number = number }
+			read { |line| (@lines ||= Array.new) << line }
+		end
+		
+		assert_equal("one", data[:number])
+		assert_equal("one", data.number)
+		assert_equal(path, data.path)
+		assert_equal(5, data.lines.size)
+		assert_equal(File.readlines(path), data.lines)
+	end
+end

data/test/tc_state.rb

@@ -0,0 +1,142 @@
+#!/usr/local/bin/ruby -w
+
+# tc_state.rb
+#
+#  Created by James Edward Gray II on 2005-08-14.
+#  Copyright 2005 Gray Productions. All rights reserved.
+
+require "test/unit"
+
+require "parse/input"
+
+class TestState < Test::Unit::TestCase
+	def test_single_skips
+		path  = File.join(File.dirname(__FILE__), "five_lines.txt")
+		lines = File.readlines(path).values_at(0..2, 4)
+		test  = self
+
+		input(path) do
+			skip(?u)
+
+			read { |line| test.assert_equal(lines.shift, line) }
+		end
+	end
+
+	def test_skipping_range
+		path  = File.join(File.dirname(__FILE__), "five_lines.txt")
+		lines = File.readlines(path).values_at(0, 3..4)
+		test  = self
+
+		input(path) do
+			start_skipping_at("two")
+			stop_skipping_at("three")
+
+			read { |line| test.assert_equal(lines.shift, line) }
+		end
+	end
+	
+	def test_searching_skipped
+		path  = File.join(File.dirname(__FILE__), "five_lines.txt")
+
+		data = input(path) do
+			@state = :skip
+			stop_skipping_at("four")
+			find_in_skipped(/ou/) { |line| @found_in_skip = line }
+
+			read { |line| (@lines ||= Array.new) << line }
+		end
+		
+		assert_equal("This is line four.\n", data.found_in_skip)
+		assert_equal(["This is line five.\n"], data.lines)
+	end
+	
+	def test_stop
+		path  = File.join(File.dirname(__FILE__), "five_lines.txt")
+
+		data = input(path) do
+			stop_at("three")
+
+			read { |line| @last_line = line }
+		end
+		
+		assert_equal("This is line two.\n", data.last_line)
+
+		data = input(path) do
+			read do |line|
+				@last_line = line
+
+				@state = :stop if @read.index("three")
+			end
+		end
+
+		assert_equal("This is line three.\n", data.last_line)
+	end
+	
+	def test_pre_and_post
+		path  = File.join(File.dirname(__FILE__), "five_lines.txt")
+
+		data = input(path) do
+			@pre = @post = 0
+			
+			pre do 
+				@pre  += 1
+				@post =  0
+			end
+			read { |line| (@lines ||= Array.new) << line }
+			read(/\w+\./) { |number| (@numbers ||= Array.new) << number }
+			post { @post += 1 }
+		end
+
+		assert_equal(5, data.pre)
+		assert_equal(1, data.post)
+	end
+	
+	def test_complex
+		path  = File.join(File.dirname(__FILE__), "ross_report.txt")
+		test  = self
+
+		input(path) do
+			@state = :skip
+			start_skipping_at("\f")
+			stop_skipping_at(/\A-[- ]+-\Z/)
+			skip(/\A\s*\Z/)
+			skip(/--\Z/)
+			
+			find_in_skipped(/((?:Period|Week)\s+\d.+?)\s*\Z/) do |period|
+				test.assert_equal("Period  02/2002", period)
+			end
+
+			stop_at("*** Selection Criteria ***")
+
+			read do |line|
+				test.assert_match(/\A\s+(?:Sales|Cust|SA)|\A[-\w]+\s+/, line)
+			end
+		end
+
+		path  = File.join(File.dirname(__FILE__), "car_ads.txt")
+
+		data = input(path, "") do
+			@state = :skip
+			stop_skipping_at("Save Ad")
+			skip(/\A\s*\Z/)
+
+			pre { @price = @miles = nil }
+			read(/\$([\d,]+\d)/) { |price| @price = price.delete(",").to_i }
+			read(/([\d,]*\d)\s*m/) { |miles| @miles = miles.delete(",").to_i }
+
+			read do |ad|
+				if @price and @price < 20_000 and @miles and @miles < 40_000
+					(@ads ||= Array.new) << ad.strip
+				end
+			end
+		end
+		
+		assert_equal([<<END_AD.strip], data.ads)
+2003 Chrysler Town & Country LX
+     $16,990, green, 21,488 mi, air, pw, power locks, ps, power mirrors,
+dual air bags, keyless entry, intermittent wipers, rear defroster, alloy,
+pb, abs, cruise, am/fm stereo, CD, cassette, tinted glass
+VIN:2C4GP44363R153238, Stock No:C153238, CALL DAN PERKINS AT 1-800-432-6326
+END_AD
+	end
+end

data/test/ts_all.rb

@@ -0,0 +1,11 @@
+#!/usr/local/bin/ruby -w
+
+# ts_all.rb
+#
+#  Created by James Edward Gray II on 2005-08-14.
+#  Copyright 2005 Gray Productions. All rights reserved.
+
+require "test/unit"
+
+require "tc_reading"
+require "tc_state"

metadata

@@ -0,0 +1,47 @@
+--- !ruby/object:Gem::Specification 
+rubygems_version: 0.8.11
+specification_version: 1
+name: parseinput
+version: !ruby/object:Gem::Version 
+  version: 0.0.1
+date: 2005-08-18 00:00:00 -05:00
+summary: Parse Input is a chain-saw tool for data mining.
+require_paths: 
+  - lib
+email: james@grayproductions.net
+homepage: http://input.rubyforge.org
+rubyforge_project: input
+description: "Parse Input is a library that aids in parsing generic input with Ruby. This
+  isn't intended to be a full-blown parser, but instead a chain-saw tool for data
+  mining arbitrary inputs quickly and easily."
+autorequire: parse/input
+default_executable: 
+bindir: bin
+has_rdoc: false
+required_ruby_version: !ruby/object:Gem::Version::Requirement 
+  requirements: 
+    - 
+      - ">"
+      - !ruby/object:Gem::Version 
+        version: 0.0.0
+  version: 
+platform: ruby
+signing_key: 
+cert_chain: 
+authors: 
+  - James Edward Gray II
+files: 
+  - lib/parse/input.rb
+  - test/tc_reading.rb
+  - test/tc_state.rb
+  - test/ts_all.rb
+  - Rakefile
+  - setup.rb
+test_files: 
+  - test/ts_all.rb
+rdoc_options: []
+extra_rdoc_files: []
+executables: []
+extensions: []
+requirements: []
+dependencies: []