C-Strings

Why this matters

Python has a str type. JavaScript has strings as a first-class kind of value. C has neither – what you’ve seen called a “string” so far (%s, "hello") is really just an ordinary char array, plus one convention: somewhere in that array sits a special value marking “the text ends here.” Understanding that convention precisely – not just using it – is what this lesson is for, because getting it wrong is one of the most common sources of real bugs (and CVEs) in C code in the wild.

The null terminator: one character marks the end

A C-string is a char array containing your text, followed by the character '\0' (the “null character,” whose numeric value is 0 – unrelated to the digit '0', which has a different value). Every function that works with strings – printf’s %s, strlen, the ones below – finds the end of the text by scanning forward until it hits a '\0', not by being told the array’s length.

Writing char name[20] = "Ada"; does two things at once: it copies the three characters 'A', 'd', 'a' into the array, and the compiler automatically appends a '\0' right after them – so name actually holds four meaningful bytes ('A', 'd', 'a', '\0') inside its 20-byte capacity, with 16 bytes left unused. The worked example prints name[3] as a number specifically to make that terminator visible: it’s 0, not a printable character, sitting right after "Ada".

This is why an array declared without a '\0' isn’t a valid C-string at all, even if every character in it is meaningful text – %s or strlen run against it will keep reading past the array’s actual bounds, looking for a '\0' that isn’t there, straight into whatever memory happens to follow. This is undefined behavior, and it’s a real, common bug class – not a hypothetical one.

strlen walks the string; sizeof doesn’t

These are frequently confused, and they measure genuinely different things. strlen(name) returns 3 for "Ada" – the number of characters before the '\0', found by actually scanning the array at runtime, one byte at a time, until it hits one. sizeof(name), by contrast, would return 20 – the array’s total declared capacity, known at compile time, with no scanning involved and no relationship to what’s actually stored in it. Use strlen to ask “how much text is actually in here”; use sizeof to ask “how much room does this array have.”

Useful functions from <string.h>

• strlen(s) – length of the string, not counting the terminator (as above).
• strcpy(dest, src) – copies src’s characters, including its terminator, into dest.
• strcat(dest, src) – appends src onto the end of whatever’s already in dest (found via dest’s own terminator), then re-terminates the result. The worked example uses this to turn "Ada" into "Ada Lovelace".
• strcmp(a, b) – compares two strings and returns 0 if they’re equal, nonzero otherwise. This one trips people up constantly: if (strcmp(a, b)) means “if they’re different,” the opposite of what the English reads like. Write if (strcmp(a, b) == 0) for “if equal” – it’s more verbose, but it says what it means.

None of strcpy or strcat check that the destination array is actually big enough. If dest doesn’t have room for src’s characters plus a terminator, they’ll write past the end of the array anyway – exactly the kind of out-of-bounds write covered in Module 3’s discussion of memory-safety tools. The worked example’s name array is declared with room to spare (char name[20]) specifically so strcat-ing " Lovelace" onto it stays safely inside bounds; sizing a buffer correctly is your responsibility, not something these functions check for you.

Reading a whole word with scanf, and its limit

You’ve used scanf("%s", word) since Module 1 to read a single word into a char array – worth stating precisely now: %s stops at the first whitespace character (space, tab, or newline) and automatically null-terminates what it read. That’s exactly why it only ever reads one word, never a whole line containing spaces – a real limitation you’ll hit the moment you need to read, say, a full name typed on one line. (Reading a full line of text needs a different tool, fgets, which this course doesn’t need yet – worth knowing the limitation exists, not necessarily the fix.)