This chapter is from the book
This chapter is from the book
This chapter is from the book
2.2 Common String Manipulation Errors
Programming with C-style strings, in C or C++, is error prone. The four most common errors are unbounded string copies, off-by-one errors, null termination errors, and string truncation.
Unbounded String Copies
Unbounded string copies occur when data is copied from an unbounded source to a fixed length character array (for example, when reading from standard input into a fixed length buffer). In Figure 2–1, the program reads characters from standard input using the gets() function (on line 4) into a fixed-length character array until a newline character is read or an end-of-file (EOF) condition is encountered.
Reading data from unbounded sources creates an interesting problem for a programmer. Because it is not possible to know beforehand how many characters a user will supply, it is not possible to pre-allocate an array of sufficient length. A common solution is to statically allocate an array that is much larger than needed, as shown in Figure 2–1. In this example, the programmer is only expecting the user to enter 8 characters so it is reasonable to assume that the 80-character length will not be exceeded. With friendly users, this approach works well. But with malicious users, a fixed-length character array can be easily exceeded.
It is also easy to make errors when copying and concatenating strings because the standard strcpy() and strcat() functions perform unbounded copy operations. In Figure 2–2, the command-line argument in argv[1] is copied into the fixed-length static array name (line 3). The static string " = " is concatenated after argv[1] in name (line 4). A second command-line argument (argv[2]) is concatenated after the static text (line 5). Can you tell which of
1. void main(void) {
2. char Password[80];
3. puts("Enter 8 character password:");
4. gets(Password); ...
5. }
Figure 2–1. Reading unbounded stream from standard input
1. int main(int argc, char *argv[]) {
2. char name[2048];
3. strcpy(name, argv[1]);
4. strcat(name, " = ");
5. strcat(name, argv[2]); ...
6. }
Figure 2–2. Unbounded string copy and concatenation
these string copy and concatenation operations may write outside the bounds of the statically allocated character array? The answer, of course, is all of them.
A simple solution is to test the length of the input using strlen() and dynamically allocate the memory, as shown in Figure 2–3. The call to malloc() on line 2 ensures that sufficient space is allocated to hold the command line argument argv[1] and a trailing null byte. The strdup() function can also be used on Single UNIX Specification, Version 2 compliant systems. The strdup() function accepts a pointer to a string and returns a pointer to a duplicate string. The strdup() function allocates memory for the duplicate string. This memory can be reclaimed by passing the return pointer to free().
Unbounded string copies are not limited to the C programming language. For example, if a user inputs more than 11 characters into the C++ program shown in Figure 2–4, it will result in an out-of-bounds write.
The standard object cin is an instantiation of the istream class. The istream class provides member functions to assist in reading and interpreting
1. int main(int argc, char *argv[]) {
2. char *buff = (char *)malloc(strlen(argv[1])+1);
3. if (buff != NULL) {
4. strcpy(buff, argv[1]);
5. printf("argv[1] = %s.\n", buff);
6. }
7. else {
/* Couldn't get the memory - recover */
8. }
9. return 0;
10. }
Figure 2–3. Dynamic allocation
1. #include <iostream.h>
2. int main() {
3. char buf[12];
4. cin >> buf;
5. cout << "echo: " << buf << endl;
6. }
Figure 2–4. Extracting characters from cin into a character array
input from a stream buffer. All formatted input is performed using the extraction operator operator>>. C++ also defines external operator>> overloaded functions that are global functions and not members of istream, including:
istream& operator>> (istream& is, char* str);
This operator extracts characters and stores them in successive locations starting at the location pointed to by str. Extraction ends when the next element is either a valid white space or a null character, or if the EOF is reached. A null character is automatically appended after the extracted characters.
The extraction operation can be limited to a specified number of characters (thereby avoiding the possibility of out-of-bounds write) if the field width inherited member (ios_base::width) is set to a value greater than 0. In this case, the extraction ends one character before the count of characters extracted reaches the value of field width leaving space for the ending null character. After a call to this extraction operation the value of the field width is reset to 0.
Figure 2–5 contains a corrected version of the Figure 2–4 program that sets the field width member to the length of the character array.
1. #include <iostream.h>
2. int main() {
3. char buf[12];
4. cin.width(12);
5. cin >> buf;
6. cout << "echo: " << buf << endl;
7. }
Figure 2–5. Extracting characters using the field width member
1. int main(int argc, char* argv[]) {
2. char source[10];
3. strcpy(source, "0123456789");
4. char *dest = (char *)malloc(strlen(source));
5. for (int i=1; i <= 11; i++) {
6. dest[i] = source[i];
7. }
8. dest[i] = '\0';
9. printf("dest = %s", dest);
10. }
Figure 2–6. Common off-by-one defects
Off-by-One Errors
Another common problem with C-style strings are off-by-one errors. Off-by-one errors are similar to unbounded string copies in that they both involve writing outside the bounds of an array. The program shown in Figure 2–6 compiles and links cleanly under Microsoft Visual C++ 6.0 and runs without error on Windows 2000 but contains several off-by-one errors.3 Can you find all the off-by-one errors in this program?
Off-by-one errors in this simple ten-line program include the following:
- The source character array (declared on line 2) is 10 bytes long, but strcpy() (line 3) copies 11 bytes, including a one-byte terminating null character.
- The malloc() function (line 4) allocates memory on the heap of the length of the source string. However, the value returned by strlen() does not account for the null byte.
- The index value i in the for loop (line 5) starts at 1, but the first position in a C array is indexed by 0.
- The ending condition for the loop (line 5) is i <= 11 . This means the loop will iterate one more time than the programmer likely intended.
- The assignment on line 8 also causes an out-of-bounds write.
Many of these mistakes are rookie errors, but experienced programmers may make them as well. It is easy to develop and deploy programs similar to this one that compile and run without error on most systems.
Null-Termination Errors
Another common problem with C-style strings is a failure to properly null terminate. In Figure 2–7, the static declarations for the three character arrays (a[], b[], and c[]) fail to allocate storage for the null-termination character. As a result, the strcpy() to a (line 5) writes a null character beyond the end of the array. Depending on how the compiler allocates storage, this null byte may be overwritten by the strcpy() on line 6. If this occurs, a now points to an array of 20 characters, while b points to an array of 10 characters. The strcpy() to c (line 7) fills c, causing the strcat() on line 8 to write well beyond the bounds of the array (particularly because the terminating null byte for b is overwritten by the strcpy() to c on line 7).
Null-termination errors, like the other string errors described in this chapter, are difficult to detect and can lie dormant in deployed code until a particular set of inputs causes a failure. The code in Figure 2–7 is also highly dependent on how the compiler allocates memory. When compiled on Windows XP, using Microsoft Visual C++ 2005 Beta 1, this program crashes while executing line 8. Interestingly, the same program in the same environment runs without error in the debugger.
1. int main(int argc, char* argv[]) {
2. char a[16];
3. char b[16];
4. char c[32];
5. strcpy(a, "0123456789abcdef");
6. strcpy(b, "0123456789abcdef");
7. strcpy(c, a);
8. strcat(c, b);
9. printf("a = %s\n", a);
10. return 0;
11. }
Figure 2–7. Null-termination defect
String Truncation
String truncation occurs when a destination character array is not large enough to hold the contents of a string. String truncation may occur while reading user input or copying a string and is often the result of a programmer trying to prevent a buffer overflow. While not as bas as a buffer overflow, string truncation results in a loss of data and, in some cases, can lead to software vulnerabilities. The code in Figure 2–5, for example, will truncate user input exceeding 11 characters.
String Errors without Functions
There are many standard string handling functions that are highly susceptible to error, including strcpy(), strcat(), gets(), streadd(), strecpy(), and strtrns(). Microsoft Visual Studio 2005, for example, has deprecated many of these functions as a result.
However, because C-style strings are character arrays, it is possible to perform an insecure string operation even without invoking a function. Figure 2–8 shows a sample C program that contains a defect resulting from a string copy operation but does not call any string library functions.
The defective program accepts a string argument (line 1), copies it to the buff character array (lines 5–9), and prints the contents of the buffer (line 10). The variable buff is declared as a fixed array of 128 characters (line 3). If the
1. int main(int argc, char *argv[]) {
2. int i = 0;
3. char buff[128];
4. char *arg1 = argv[1];
5. while (arg1[i] != '\0' ) {
6. buff[i] = arg1[i];
7. i++;
8. }
9. buff[i] = '\0';
10. printf("buff = %s\n", buff);
11. }
Figure 2–8. Defective string manipulation code
first argument to the program equals or exceeds 128 characters (remember the trailing null character), the program writes outside the bounds of the fixed-size array. Clearly, eliminating the use of dangerous functions does not guarantee your program is free from security flaws. In the following sections, you will see how these security flaws can lead to exploitable vulnerabilities.