Added test harness

Author: unknown

classify.py

@@ -8,19 +8,15 @@ class Classify(Mode):
 	RARE_WORD_PROB = 0.5
 	EXCLUSIVE_WORD_PROB = 0.99
 
-	def validate(self, args):
-		if len(args) != 5:
-			raise ValueError('Usage: %s classify <file> <doctype> <doctype>' % args[0])
-
-		file_contents = None
+	def validate_file_name(self, file_name):
 		try:
-			file_contents = open(args[2], 'r').read()
+			file_contents = open(file_name, 'r').read()
+			self.words = text_to_list(file_contents)
+
 		except Exception as e:
-			raise ValueError('Unable to read specified file "%s", the error message was: %s' % (args[2], e))
-		
-		doctype1 = args[3]
-		doctype2 = args[4]
+			raise ValueError('Unable to read specified file "%s", the error message was: %s' % (file_name, e))
 
+	def validate_doctypes(self, doctype1, doctype2):
 		if doctype1 == doctype2:
 			raise ValueError('Please enter two different doctypes')
 
@@ -33,7 +29,13 @@ def validate(self, args):
 
 		self.doctype1 = doctype1
 		self.doctype2 = doctype2
-		self.words = text_to_list(file_contents)
+
+	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])
 
 	def p_for_word(self, db, word):
 		total_word_count = self.doctype1_word_count + self.doctype2_word_count
@@ -62,8 +64,7 @@ def p_from_list(self, l):
 
 		return p_product / (p_product + p_inverse_product)
 
-	
-	def execute(self):
+	def calculate(self):
 		pl = []
 		db = Db()
 
@@ -80,5 +81,11 @@ def execute(self):
 
 		result = self.p_from_list(pl)
 
+		return result
+
+	def execute(self):
+		result = self.calculate()
+
 		print 'Probability that document is %s rather than %s is %1.2f' % (self.doctype1, self.doctype2, result)
+
 		return result

testharness.py

@@ -0,0 +1,51 @@
+from __future__ import division
+import sys
+import os
+from classify import Classify
+from db import Db
+
+classifier = Classify()
+
+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]
+
+def check_dir(d):
+	results = []
+	for f in os.listdir(d):
+		if f.endswith(".txt"):
+			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 __name__ == '__main__':
+	usage = 'Usage: %s <file> <expected doctype> <other doctype>' % sys.argv[0]
+
+	if len(sys.argv) != 4:
+		raise ValueError(usage)
+
+	input_file = sys.argv[1]
+	doctype_expected = sys.argv[2]
+	doctype_other = sys.argv[3]
+
+	classifier.validate_doctypes(doctype_expected, doctype_other)
+
+	results = None
+	if os.path.isfile(input_file):
+		results = check_file(input_file)
+	elif os.path.isdir(input_file):	
+		results = check_dir(input_file)
+	else:
+		raise ValueError("Unable to find file/directory '%s'\n%s" % (input_file, usage))
+
+	show_results(results)