refactoring, fixes, validation

Author: unknown

bayes.py

@@ -28,7 +28,7 @@ def register_mode(mode_class):
 
 		mode = modes[mode_name]()
 		mode.validate(args)
-		mode.execute()
+		mode.output(mode.execute())
 
 	except Exception as ex:
 		print ex

classify.py

@@ -8,15 +8,24 @@ class Classify(Mode):
 	RARE_WORD_PROB = 0.5
 	EXCLUSIVE_WORD_PROB = 0.99
 
-	def validate_file_name(self, file_name):
+	def set_text(self, text):
+		words = text_to_list(text)
+
+		if not len(words):
+			raise ValueError('Text did not contain any valid words')
+
+		self.words = words
+		return self
+
+	def set_file_name(self, file_name):
 		try:
 			file_contents = open(file_name, 'r').read()
-			self.words = text_to_list(file_contents)
+			return self.set_text(file_contents)
 
 		except Exception as e:
 			raise ValueError('Unable to read specified file "%s", the error message was: %s' % (file_name, e))
 
-	def validate_doctypes(self, doctype1, doctype2):
+	def set_doctypes(self, doctype1, doctype2):
 		if doctype1 == doctype2:
 			raise ValueError('Please enter two different doctypes')
 
@@ -34,8 +43,8 @@ def validate(self, args):
 		if len(args) != 5:
 			raise ValueError('Usage: %s classify <file> <doctype> <doctype>' % args[0])
 
-		self.validate_file_name(args[2])
-		self.validate_doctypes(args[3], args[4])
+		self.set_file_name(args[2])
+		self.set_doctypes(args[3], args[4])
 
 	def p_for_word(self, db, word):
 		total_word_count = self.doctype1_word_count + self.doctype2_word_count
@@ -64,7 +73,7 @@ def p_from_list(self, l):
 
 		return p_product / (p_product + p_inverse_product)
 
-	def calculate(self):
+	def execute(self):
 		pl = []
 		db = Db()
 
@@ -83,9 +92,5 @@ def calculate(self):
 
 		return result
 
-	def execute(self):
-		result = self.calculate()
-
+	def output(self, result):
 		print 'Probability that document is %s rather than %s is %1.2f' % (self.doctype1, self.doctype2, result)
-
-		return result

learn.py

@@ -36,4 +36,7 @@ def execute(self):
 		d = list_to_dict(l)
 		db.update_word_counts(d, self.doc_type)
 		db.update_doctype_count(self.count, self.doc_type)
+		return self.count
 
+	def output(self, _):
+		print "Processed %s documents of type '%s'" % (self.count, self.doc_type)

mode.py

@@ -5,5 +5,5 @@ def validate(self):
 	def execute(self):
 		raise NotImplementedError()
 
-	def name(self):
+	def output(self):
 		raise NotImplementedError()

reset.py

@@ -3,14 +3,13 @@
 from db import Db
 
 class Reset(Mode):
-	def name(self):
-		return 'reset'
-
 	def validate(self, args):
 		if len(args) != 2:
 			raise ValueError('Usage: %s reset' % args[0])
 
 	def execute(self):
 		Db().reset()
-		print 'Reset Complete'
 		Status().execute()
+
+	def output(self, _):
+		print 'Reset Complete'

status.py

@@ -2,17 +2,23 @@
 from mode import Mode
 
 class Status(Mode):
-	def name(self):
-		return 'status'
 
 	def validate(self, args):
 		if len(args) != 2:
 			raise ValueError('Usage: %s status' % args[0])
 
 	def execute(self):
 		db = Db()
+		return db.get_doctype_counts().items()
+
+	def output(self, results):
 		bar = '=' * 40
-		print 'Status:\n'
-		for doctype, count in db.get_doctype_counts().items():
-			print doctype, ': ', count
+		print '%s\nStatus:\n%s\n' % (bar, bar)
+
+		if results:
+			for doctype, count in results:
+				print '%s: %s' % (doctype, count)
+		else:
+			print 'No data'
 
+		print '\n%s' % bar

testharness.py

@@ -10,23 +10,31 @@ def is_doctype_valid(doctype):
 	return Db().get_words_count(doctype) > 0
 
 def check_file(f):
-	classifier.validate_file_name(f)
-	result = classifier.calculate()
-	print result
-	return [result]
+	results = []
+	for line in open(f, 'r').readlines():
+		try:
+			classifier.set_text(line)
+			results += [classifier.execute()]
+		except ValueError:
+			pass
+	
+	return results
 
 def check_dir(d):
 	results = []
 	for f in os.listdir(d):
-		if f.endswith(".txt"):
+		if f.endswith(".js"):
 			results += check_file(os.path.join(d,f))
 
 	return results
 
 def show_results(results):
 	result_count = len(results)
-	print 'Tested with %s document%s' % (result_count, '' if result_count == 1 else 's')
-	print 'Result was %1.2f (1 is good, 0 is bad)' % (sum(results) / result_count,)
+	if result_count:
+		print 'Tested with %s document%s' % (result_count, '' if result_count == 1 else 's')
+		print 'Result was %1.2f (0 = %s, 1 = %s)' % (sum(results) / result_count, doctype_other, doctype_expected)
+	else :
+		print 'No documents found'
 
 if __name__ == '__main__':
 	usage = 'Usage: %s <file> <expected doctype> <other doctype>' % sys.argv[0]
@@ -38,7 +46,7 @@ def show_results(results):
 	doctype_expected = sys.argv[2]
 	doctype_other = sys.argv[3]
 
-	classifier.validate_doctypes(doctype_expected, doctype_other)
+	classifier.set_doctypes(doctype_expected, doctype_other)
 
 	results = None
 	if os.path.isfile(input_file):