modified 0001

effyhuihui · effyhuihui · commit cdca0affab30 · 2014-12-04T14:31:23.000-08:00
diff --git a/effy/0001/0001.py b/effy/0001/0001.py
@@ -25,11 +25,32 @@ def digit(raw):
 
 def codeGen(n):
 	codes_pool = []
-	for i in range(n-1):
+	for i in range(n):
 		code = ""
 		for i in range(10):
 			code += digit(codeSeedA)
 		codes_pool.append(code)
 	return codes_pool
 
-codes = codeGen(200)
+'''
+Standard uuid
+'''
+import uuid
+def uuidGen(n):
+	codes_pool =[]
+	for i in range(n):
+		codes_pool.append(uuid.uuid4())
+	return codes_pool
+
+#codes_udf = codeGen(10000)
+codes_uuid = uuidGen(10000)
+
+
+'''
+In the case of 200 activation code, it is almost equally fast.
+
+As the size of code pool grows (e.g. to 10000), uuidGen() is twice slower than codeGen(),
+since uuid is checking whether duplications. Plus, the confidence level of codeGen() to not
+produce duplication lowers down as pool size increases.
+
+'''
diff --git a/effy/0002/ActivationCodeGen.py b/effy/0002/ActivationCodeGen.py
@@ -1,24 +1,7 @@
 #!/usr/bin/env
-'''
-Self-defined activation code
-
-1. has length of 10
-2. mixture of letters and numbers
-3. unique throughtout 200 combos
-
-Pros: 
-total of 3656158440062976 combos, randomly generation of 200 is almost 
-guaranteed to be unique; Confident level if high without look up opeations.
-And without look up, it's super fast :) The more codes needed, the effect 
-is more obvious.
-
-Cons: 
-Without back tracking, small chance would occur of duplication.
-'''
 import random
 
 codeSeedA = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"
-
 def digit(raw):
 	l = len(raw)
 	return raw[random.randrange(l)] 
@@ -31,5 +14,3 @@ def codeGen(n):
 			code += digit(codeSeedA)
 		codes_pool.append(code)
 	return codes_pool
-
-codes = codeGen(200)
