问题
I'm not sure which one is better. I need to parse each character of an input string and get a replacing string for the character. For some objects all alphanumeric characters are allowed so using switch/case will cause a lot of code and reduce readability and maintainability but I can use a static method. Using a HashTable also requires a lot of code
Using the static method:
private static string EncodeChar(char c)
{
var symbols = string.Empty;
switch (c)
{
case '0':
symbols = "Test";
break;
case '1':
symbols = "Hello";
break;
[...]
}
symbols;
}
Using HashTable:
private static Hashtable table = CreateTable();
private static Hashtable CreateTable()
{
var table = new HashTable();
table.Add('0',"Test");
table.Add('1', "Hello");
[...]
return table;
}
private static string EncodeChar(char c)
{
return table.ContainsKey(c) ? table[c].ToString() : string.Empty;
}
The Encode method:
public void Encode()
{
string output = string.Empty;
for (int i = 1; i < Data.Length; i++)
{
output = string.Concat(output, EncodeChar(Data[i]));
}
EncodedData = output;
}
What are the advantages/disadvantages concerning performance and memory allocation?
回答1:
I'd use the HashTable
, because the code is more readable and maintainable: you could one day decide to load the subtitution strings from an XML file, so that you won't change the code to change the mappings.
回答2:
Hashing is faster as you can directly access the 'encoded string'
for example, If you assume that all characters are '9', so it will have to evaluate 8 if conditions
before executing the right statement
, each time you process a character.
That's just a worst-case example while using the switch()
回答3:
The switch code does the same thing as the alternative, but it's much more readable which should be the main concern here.
I would however use an abstract factory to get one of a number of different encoding implementations for different types of input.
来源:https://stackoverflow.com/questions/9277415/better-use-hashtable-or-switch-case