String escape sequences are transformed to the corresponding character at compile time. Ordinary strings that happen to contain backwards slashes are not transformed.
For example, the strings notEscaped
and notEscaped2
below are not transformed to a newline character, but will stay as two different characters ('\'
and 'n'
).
string escaped = "\n";
string notEscaped = "\\" + "n";
string notEscaped2 = "\\n";
Console.WriteLine(escaped.Length); // 1
Console.WriteLine(notEscaped.Length); // 2
Console.WriteLine(notEscaped2.Length); // 2
string sqrt = "\u221A"; // β
string emoji = "\U0001F601"; // π
string text = "\u0022Hello World\u0022"; // "Hello World"
string variableWidth = "\x22Hello World\x22"; // "Hello World"
Apostrophes
char apostrophe = '\'';
Backslash
char oneBackslash = '\\';
Backslash
// The filename will be c:\myfile.txt in both cases
string filename = "c:\\myfile.txt";
string filename = @"c:\myfile.txt";
The second example uses a verbatim string literal, which doesn't treat the backslash as an escape character.
Quotes
string text = "\"Hello World!\", said the quick brown fox.";
string verbatimText = @"""Hello World!"", said the quick brown fox.";
Both variables will contain the same text.
"Hello World!", said the quick brown fox.
Newlines
Verbatim string literals can contain newlines:
string text = "Hello\r\nWorld!";
string verbatimText = @"Hello
World!";
Both variables will contain the same text.
The following examples will not compile:
string s = "\c";
char c = '\c';
Instead, they will produce the error Unrecognized escape sequence
at compile time.
Escape sequences are not restricted to string
and char
literals.
Suppose you need to override a third-party method:
protected abstract IEnumerable<Texte> ObtenirΕuvres();
and suppose the character Ε
is not available in the character encoding you use for your C# source files. You are lucky, it is permitted to use escapes of the type \u####
or \U########
in identifiers in the code. So it is legal to write:
protected override IEnumerable<Texte> Obtenir\u0152uvres()
{
// ...
}
and the C# compiler will know Ε
and \u0152
are the same character.
(However, it might be a good idea to switch to UTF-8 or a similar encoding that can handle all characters.)