Question

I have a question related to string comparison vs. character comparison.

Characters > and 0 (zero) have following decimal values 62 and 48 accordingly.

When I compare two characters in the following code, I get value True (which is correct)

Console.WriteLine('>' > '0');

When I compare two one-character strings in the following code, I get value -1 which indicates that “>” is less than “0” (default culture is English)

Console.WriteLine(string.Compare(">", "0"));

Whereas comparison of “3” and “1” (51 and 49 code values) in the following code returns 1 (as expected)

Console.WriteLine(string.Compare("3", "1"));

Also, string.Compare(string str1, string str2) documentation says:

The comparison uses the current culture to obtain culture-specific
information such as casing rules and the alphabetic order of
individual characters

Would you be able to explain (or provide reference to some documentation) how string comparison is implemented e.g. how alphabetic order of individual characters is calculated etc?

Answer 1

When you compare the characters '>' and '0', you are comparing their ordinal values.

To get the same behaviour from a string comparison, supply the ordinal string comparison type:

  Console.WriteLine(string.Compare(">", "0", StringComparison.Ordinal));
  Console.WriteLine(string.Compare(">", "0", StringComparison.InvariantCulture));
  Console.WriteLine(string.Compare(">", "0", StringComparison.CurrentCulture));

The current culture is used by default, which has a sorting order intended to sort strings ‘alphabetically’ rather in strictly lexical order, for some definition of alphabetically.