28

I am playing with Pex and one of the parameters it passes into my method is "\0".

What does that mean? My guess is an empty string ("") based on the content of my method. However, if it is the same then why not just use "" instead of "\0"?

Anyone know what it is?

Vaccano
  • 78,325
  • 149
  • 468
  • 850
  • Is it passing the string containing a backslash followed by a zero, or is it passing a string containing the null character? – Anon. Feb 18 '10 at 22:56
  • This is what the code says: `this.TestSaveString("\0");` – Vaccano Feb 18 '10 at 23:10
  • Thanks for all the great answers! I chose Randolpho's because it gave me the most detail about what was going on. – Vaccano Feb 19 '10 at 18:52

6 Answers6

43

'\0' is a "null character". It's used to terminate strings in C and some portions of C++. Pex is doing a test to see how your code handles the null character, likely looking for the Poison Null Byte security exploit.

Most C# code has nothing to fear; if you pass your string to unmanaged code, however, you may have problems.

Edit:

Just to be explicit... Pex is passing a string containing a null character. This is not a null reference.

Randolpho
  • 55,384
  • 17
  • 145
  • 179
  • 3
    Note that a NULL string IS NOT the same as an empty string. From the MSDN article: "By contrast, a null string does not refer to an instance of a System.String object and any attempt to call a method on a null string results in a NullReferenceException. However, you can use null strings in concatenation and comparison operations with other strings. The following examples illustrate some cases in which a reference to a null string does and does not cause an exception to be thrown" – jfawcett Feb 18 '10 at 23:00
6

It's a string with a null character. Older string libraries — like that used in C or older C++ libraries — used the '\0' character to indicate the end of the string.

Newer environments like .Net use a different system, but there is a lot of history around ending a string with '\0', such that it's a common point of error. Testing libraries like Pex will use it to make sure your program handles it correctly.

Randolpho
  • 55,384
  • 17
  • 145
  • 179
Joel Coehoorn
  • 399,467
  • 113
  • 570
  • 794
  • Eh.... I can't bring myself to -1 this one, but it looks like you're implying that C# strings are null terminated. They are not. – Randolpho Feb 18 '10 at 22:59
  • 1
    @Randolpho "Newer environments like .Net use a different system" -- seems to me I'm implying the exact opposite. – Joel Coehoorn Feb 18 '10 at 23:01
  • Yep. I'm not sure how anything like that is implied. – Andrew Feb 18 '10 at 23:05
  • I guess I'm getting hung up on your first sentence. The asker is clearly using .NET. Pex is *not* passing a string with an "extra null character", it is passing a string that contains a single null character. "extra null character" implies that there are two null characters, one in the string and one terminating the string. This is not the case. I'm not dinging on it because it's not explicit like @mobrule's was, but it's pretty heavily implied to me. – Randolpho Feb 18 '10 at 23:05
  • The more I think about it... the rest of the answer is fine. I'm thinking I'm just gonna edit that first sentence and +1 you. :) – Randolpho Feb 18 '10 at 23:11
5

It's a string containing the character '\0'. C# doesn't treat this in any particularly special way - it's just unicode character U+0000. If you write:

int firstCodePoint = text[0];

then you'll find firstCodePoint is 0.

Jon Skeet
  • 1,421,763
  • 867
  • 9,128
  • 9,194
3
Escape Sequence  
\0    
Character Name 
Null    
Unicode Encoding     
0x0000

See this link.

Uke
  • 169
  • 3
  • 21
2

A string of length 1, containing the character \u0000 (aka NUL). This character is not treated specially.

In C, which uses \0 to terminate string, you also allocate a string of length 1. In this case the standard string functions will report a length of 0, since the string contains \0 as well as being terminated with it. You could safely modify str[0], or strncat a single character into it.

Lachlan Roche
  • 25,678
  • 5
  • 79
  • 77
1

I just found a good example in which \0 is very important and necessary.

Assume we want to remove the last unwanted , in the following code.

for (int i = 0; i < 5; i++)
{
    foreach (var a in "Money")
        Console.Write($"{a},");
}

enter image description here

If we only add Console.Write("\b\n"); as follows,

for (int i = 0; i < 5; i++)
{
    foreach (var a in "Money")
        Console.Write($"{a},");
    Console.Write("\b\n");
}

The output will be still the same.

But if we add \0 as follows,

for (int i = 0; i < 5; i++)
{
    foreach (var a in "Money")
        Console.Write($"{a},");
    Console.Write("\b\0\n");
}

The unwanted trailing , vanishes.

enter image description here

Second Person Shooter
  • 14,188
  • 21
  • 90
  • 165