Merge pull request JohannesBuchner#54 from mynameisvinn/master

disallow negative hash size arguments

.gitignore

@@ -6,3 +6,4 @@ build
 dist
 ImageHash.egg-info/
 .eggs
+.DS_Store

README.rst

@@ -10,6 +10,12 @@ A image hashing library written in Python. ImageHash supports:
 
 |Travis|_ |Coveralls|_
 
+Rationale
+---------
+Why can’t we use md5, sha-1, etc.?
+
+Unfortunately, we cannot use cryptographic hashing algorithms in our implementation. Due to the nature of cryptographic hashing algorithms, very tiny changes in the input file will result in a substantially different hash. In the case of image fingerprinting, we actually want our similar inputs to have similar output hashes as well.
+
 Requirements
 -------------
 Based on PIL/Pillow Image, numpy and scipy.fftpack (for pHash)

imagehash/__init__.py

@@ -121,6 +121,8 @@ def average_hash(image, hash_size=8):
 
 	@image must be a PIL instance.
 	"""
+	if hash_size < 0:
+		raise ValueError("Hash size must be positive")
 
 	# reduce size and complexity, then covert to grayscale
 	image = image.convert("L").resize((hash_size, hash_size), Image.ANTIALIAS)
@@ -143,6 +145,9 @@ def phash(image, hash_size=8, highfreq_factor=4):
 
 	@image must be a PIL instance.
 	"""
+	if hash_size < 0:
+		raise ValueError("Hash size must be positive")
+
 	import scipy.fftpack
 	img_size = hash_size * highfreq_factor
 	image = image.convert("L").resize((img_size, img_size), Image.ANTIALIAS)
@@ -184,6 +189,9 @@ def dhash(image, hash_size=8):
 	@image must be a PIL instance.
 	"""
 	# resize(w, h), but numpy.array((h, w))
+	if hash_size < 0:
+		raise ValueError("Hash size must be positive")
+
 	image = image.convert("L").resize((hash_size + 1, hash_size), Image.ANTIALIAS)
 	pixels = numpy.array(image.getdata(), dtype=numpy.float).reshape((hash_size, hash_size + 1))
 	# compute differences between columns

imagehash/tests/__init__.py

@@ -58,3 +58,7 @@ def check_hash_stored(self, func, image):
                 '- stringified hash {}'.format(distance, image_hash,
                                                other_hash))
         self.assertEqual(distance, 0, emsg)
+
+    def check_hash_size(self, func, image, size):
+        with self.assertRaises(ValueError):
+            func(image, -1)

imagehash/tests/test_average_hash.py

@@ -21,6 +21,10 @@ def test_average_hash_length(self):
     def test_average_hash_stored(self):
         self.check_hash_stored(self.func, self.image)
 
+    def test_average_hash_size(self):
+        self.check_hash_size(self.func, self.image, -1)
+
+
 
 if __name__ == '__main__':
     unittest.main()

imagehash/tests/test_dhash.py

@@ -21,6 +21,8 @@ def test_dhash_length(self):
     def test_dhash_stored(self):
         self.check_hash_stored(self.func, self.image)
 
+    def test_dhash_size(self):
+        self.check_hash_size(self.func, self.image, -1)
 
 if __name__ == '__main__':
     unittest.main()

imagehash/tests/test_phash.py

@@ -20,7 +20,9 @@ def test_phash_length(self):
 
     def test_phash_stored(self):
         self.check_hash_stored(self.func, self.image)
-
+
+    def test_phash_size(self):
+        self.check_hash_size(self.func, self.image, -1)
 
 if __name__ == '__main__':
     unittest.main()

imagehash/tests/test_whash.py

@@ -57,6 +57,5 @@ def test_image_scale_not_2power(self):
             with six.assertRaisesRegex(self, AssertionError, emsg):
                 imagehash.whash(self.image, image_scale=image_scale+1)
 
-
 if __name__ == '__main__':
     unittest.main()