getting body of HTTP request

How can a CGI program read the entire body of an HTTP request and output it? I’ve tried a number of things and this is what I’ve currently got:

Code:

---

    char temp[7];
    char read[1000];
    strcpy(temp ,getenv("CONTENT_LENGTH"));
    int inputLength = atoi(temp);
    cin.get(temp, sizeof(temp));//divide by size of character
    fread( read, inputLength, 1, stdin );
cout << "Request body: " << read << "</p>\n";

---

and some funny stuff gets printed out near end

Fread() does not add a null terminator and there is not necessarily one in the body.

At the end, you need to do something like:

Code:

---

read[inputLength] = '\0';

---

This is C, not C++ (except for cout), and also a ticking time bomb. Make up your mind as to whether you want to use C or C++.

Well to expand on the "time bomb" bit, this is also unsafe C: strcpy is vulnerable to buffer overflow. Call strncat on an empty string, or call strncpy and '\0' terminate the result yourself.

Not to mention copying into fixed-size buffers without checking to see if there is actually enough room before doing the copying!

Quote:

---

Originally Posted by Elysia

Not to mention copying into fixed-size buffers without checking to see if there is actually enough room before doing the copying!

---

Actually, if you call the functions I mentioned then you shouldn't have to check and if you do check then calls like strcpy can be safe. Since it's fixed size arrays we're using, if some of the data ends up truncated all you can do is increase the fixed size.