I'm learning C and am currently experimenting with storing strings in variables. I put together the following to try different stuff.
#include <stdio.h>
int main() {
char *name = "Tristan";
char today[] = "January 1st, 2016";
char newyear[] = {'H','a','p','p','y',' ','N','e','w',' ','Y','e','a','r','!','\n'};
printf("Hello world!\n");
printf("My name is %s.\n", name);
printf("Today is: %s.\n", today);
printf(newyear);
return 0;
}
After compiling this code and running it, I get the following results:
Hello world!
My name is Tristan.
Today is: January 1st, 2016.
Happy New Year!
January 1st, 2016
Now this is pretty much what I would expect, by why would "January 1st, 2016" get printed out again at the end of the program's output?
If I take the "\n" out of the "newyear" array, it will not do this.
Would someone please explain why this is?
newyear misses a trailing null byte, so printfing it is undefined behavior.
Only string literals implicitly append a null byte. You explicitly initialize every character, so no null byte is appended.
Undefined behavior means that something the standard does not define in this occasion will happen. That includes nothing happening, you bursting into tears, or, yes, printing some string twice.
Just add an additional character, i.e., a null byte to the array to resolve the problem:
char newyear[] = {'H','a','p','p','y',' ','N','e','w',' ','Y','e','a','r','!','\n', '\0'};
Note that no sane person initializes an automatic char array with a string like that. Just stick to string literals! (I think you did it just for learning purposes, though.)
Remember that strings in C are terminated by the special '\0' character.
Not having this terminator at the end of data that is treated as a string will lead to undefined behavior as the string functions pass the end of the data searching for the terminator.
This because you are defining newyear directly as a char array and not through the string literal "" syntax. This prevents the compiler from adding a trailing \0 character which is required to mark the end of a string.
Since both newyear and today reside on stack, in this case they have contiguous storage there so printf keeps after the \n of newyear and prints contents of memory until a \0 is found.
newyear should finish with a '\0' instead of the newline, to be a C string. You can then put the newline in the printf statement, like the others:
char newyear[] = {'H','a','p','p','y',' ','N','e','w',' ','Y','e','a','r','!','\0'};
//...
printf("%s.\n", newyear);
Or, you can add the string terminator to the array, and use the printf as you did:
char newyear[] = {'H','a','p','p','y',' ','N','e','w',' ','Y','e','a','r','!','\n','\0'};
//...
printf(newyear);
In your first two examples, a string defined as "my string" automatically has the '\0' appended, by the compiler.
Related
I have been working with strings in C. While working with ways to declare them and initialize them, I found some weird behavior I don't understand.
#include<stdio.h>
#include<string.h>
int main()
{
char str[5] = "World";
char str1[] = "hello";
char str2[] = {'N','a','m','a','s','t','e'};
char* str3 = "Hi";
printf("%s %zu\n"
"%s %zu\n"
"%s %zu\n"
"%s %zu\n",
str, strlen(str),
str1, strlen(str1),
str2, strlen(str2),
str3, strlen(str3));
return 0;
}
Sample output:
Worldhello 10
hello 5
Namaste 7
Hi 2
In some cases, the above code makes str contain Worldhello, and the rest are as they were intialized. In some other cases, the above code makes str2 contain Namastehello. It happens with different variables I never concatenated. So, how are they are getting combined?
To work with strings, you must allow space for a null character at the end of each string. Where you have char str[5]="World";, you allow only five characters, and the compiler fills them with “World”, but there is no space for a null character after them. Although the string literal "World" includes an automatic null character at its end, you did not provide space for it in the array, so it is not copied.
Where you have char str1[]="hello";, the compiler determines the array size by counting the characters, including the null character at the end of the string literal.
Where you have char str2[]={'N','a','m','a','s','t','e'};, there is no string literal, just a list of individual characters. The compiler determines the array size by counting those. Since there is no null character, it does not provide space for it.
One potential consequence of failing to terminate a string with a null character is that printf will continue reading memory beyond the string and printing characters from the values it finds. When the compiler has placed other character arrays after such an array you are printing, characters from those arrays may appear in the output.
If you allow space for a null character in str and provide a zero value in str2, your program will print strings in an orderly way:
#include <stdio.h>
#include <string.h>
int main(void)
{
char str[6] = "World"; // 5 letters plus a null character.
char str1[] = "hello";
char str2[] = {'N', 'a', 'm', 'a', 's', 't', 'e', 0}; // Include a null.
char *str3 = "Hi";
printf("%s %zu\n%s %zu\n%s %zu\n%s %zu\n",
str, strlen(str),
str1, strlen(str1),
str2, strlen(str2),
str3, strlen(str3));
return 0;
}
Undefined behavior in non-null-terminated, adjacently-stored C-strings
Why do you get this part:
Worldhello 10
hello 5
...instead of this?
World 5
hello 5
The answer is that printf() prints chars until it hits a null character, which is a binary zero, frequently written as the '\0' char. And, the compiler happens to have placed the character array containing hello right after the character array containing World. Since you explicitly forced the size of str to be 5 via str[5], the compiler was unable to fit the automatic null character at the end of the string. So, with hello happening to be (not guaranteed to be) right after World, and printf() printing until it sees a binary zero, it printed World, saw no terminating null char, and continued right on into the hello string right after it. This resulted in it printing Worldhello, and then stopping only when it saw the terminating character after hello, which string is properly terminated.
This code relies on undefined behavior, which is a bug. It cannot be relied upon. But, that is the explanation for this case.
Run it with gcc on a 64-bit Linux machine online here: Online GDB: undefined behavior in NON null-terminated C strings
#Eric Postpischil has a great answer and provides more insight here.
From the C tag wiki:
This tag should be used with general questions concerning the C language, as defined in the ISO 9899 standard (the latest version, 9899:2018, unless otherwise specified — also tag version-specific requests with c89, c99, c11, etc).
You've asked a "how?" question about something that none of those documents defines, and so the answer is undefined in the context of C. You can only experience this phenomenon through undefined behaviour.
how are they are getting combined?
There is no such requirement that any of these variables are "combined" or are immediately located after each other; trying to observe that is undefined behaviour. It may appear to coincidentally work (whatever that means) for you at times on your machine, while failing at other times or using some other machine or compiler, etc. That's purely coincidental and not to be relied upon.
In some cases, the above code assigns str with Worldhello and the rest as they were intitated.
In the context of undefined behaviour, it makes no sense to make claims about how your code functions, as you've already noticed, the functionality is erratic.
I found some weird Behaviour with them.
If you want to prevent erratic behaviour, stop invoking undefined behaviour by accessing arrays out of bounds (i.e. causing strlen to run off the end of an array).
Only one of those variables is safe to pass to strlen; you need to ensure the array contains a null terminator.
Given for example a char *p that points to the first character in "there is so \0ma\0ny \0 \\0 in t\0his stri\0ng !\0\0\0\0",
how would Strrchr() find the last occurrence of null-character?
the following questions arises:
=>What conditions would it depend on to stop the loop!
=>I think in all cases it'll try to access the next memory area to check for its condition?at some point bypassing the string boundaries, UB! so is it safe !
please if i'am wrong feel free to correct me!
It's very simple, as explained in the comments.
The first \0 is the last and the only one in a C string.
So if you write
char *str = "there is so \0ma\0ny \0 \\0 in t\0his stri\0ng !\0\0\0\0";
char *p = strrchr(str, 's');
printf("%s\n", p);
it will print
so
because strchr will find the 's' in "so", which is the last 's' in the string you gave it. And (to answer your specific question) if you write
p = strrchr(str, '\0');
printf("%d %s\n", (int)(p - str), p+1);
it will print
12 ma
proving that strchr found the first \0.
It's obvious to you that str is a long string with some embedded \0's in it. But, in C, there is no such thing as a "string with embedded \0's in it". It is impossible, by definition, for a C string to contain an embedded \0. The first \0, by definition, ends the string.
One more point. You had mentioned that if you were to "access the next memory area", that you would "at some point bypassing the string boundaries, UB!" And you're right. In my answer, I skirted with danger when I said
p = strrchr(str, '\0');
printf("%d %s\n", (int)(p - str), p+1);
Here, p points to what strrchr thinks is the end of the string, so when I compute p+1 and try to print it using %s, if we don't know better it looks like I've indeed strayed into Undefined Behavior. In this case it's safe, of course, because we know exactly what's beyond the first \0. But if I were to write
char *str2 = "hello";
p = strrchr(str2, '\0');
printf("%s\n", p+1); /* WRONG */
then I'd definitely be well over the edge.
There is a difference between "a string", "an array of characters" and "a char* pointer".
A C String is a number of characters terminated by a null character.
An array of characters is a defined number of characters.
A char* pointer is technically a pointer to a single character, but often used to mark a point in a C style string.
You say you have a pointer to a character (char*p) and the value of *p is 't', but you believe that *p is the first character of a C style string
"there is so \0ma\0ny \0 \\0 in t\0his stri\0ng !\0\0\0\0".
As others have said, because you said this is a C style string and you don't know the length of it then the first null after p will mark the end of the string.
If this was a character array char str[40] then you could find the last null by looping from the end of the array towards the start for (i=39; i>=0; i--) BUT you don't know then length, so that won't work.
Hope that helps, and please excuse me if I have strayed into C++, its 25 years since I did C :)
In the case you present, you can never know if the null character you've found is the last one since you have no guarantee for the end of the string. As it is a c-string, it is guaranteed that the string ends with a '\0', but if you decide to go beyond that, you can't know if the memory you're accessing is yours. Accessing memory out of an array has undefined behaviour as you can either be accessing just the next object that is in memory that is yours or you could touch memory that is unallocated, but its block still belongs to your process, or you can try to touch a segment that is not yours at all. And only the third one will cause a SIGSEGV. You can see this question to check for segmentation fault without crashing your program, but your string could have ended way before you can catch it that way.
There is a reason for the strings to have an ending character. If you insist to have \0 in multiple places in your string, you can just terminate with another character, but note that all library functions will still consider the first \0 to be the end of the string.
It is considered a bad practice and a very bad thing to have multiple \0 in your strings so if you can, avoid it.
As we know a string terminates with '\0'.
It's because to know the compiler that string ended, or to secure from garbage values.
But how does an array terminate?
If '\0' is used it will take it as 0 a valid integer,
So how does the compiler knows the array ended?
C does not perform bounds checking on arrays. That's part of what makes it fast. However that also means it's up to you to ensure you don't read or write past the end of an array. So the language will allow you to do something like this:
int arr[5];
arr[10] = 4;
But if you do, you invoke undefined behavior. So you need to keep track of how large an array is yourself and ensure you don't go past the end.
Note that this also applies to character arrays, which can be treated as a string if it contains a sequence of characters terminated by a null byte. So this is a string:
char str[10] = "hello";
And so is this:
char str[5] = { 'h', 'i', 0, 0, 0 };
But this is not:
char str[5] = "hello"; // no space for the null terminator.
C doesn't provide any protections or guarantees to you about 'knowing the array is ended.' That's on you as the programmer to keep in mind in order to avoid accessing memory outside your array.
C language does not have native string type. In C, strings are actually one-dimensional array of characters terminated by a null character '\0'.
From C Standard#7.1.1p1 [emphasis mine]
A string is a contiguous sequence of characters terminated by and including the first null character. The term multibyte string is sometimes used instead to emphasize special processing given to multibyte characters contained in the string or to avoid confusion with a wide string. A pointer to a string is a pointer to its initial (lowest addressed) character. The length of a string is the number of bytes preceding the null character and the value of a string is the sequence of the values of the contained characters, in order.
String is a special case of character array which is terminated by a null character '\0'. All the standard library string related functions read the input string based on this rule i.e. read until first null character.
There is no significance of null character '\0' in array of any type apart from character array in C.
So, apart from string, for all other types of array, programmer is suppose to explicitly keep the track of number of elements in the array.
Also, note that, first null character ('\0') is the indication of string termination but it is not stopping you to read beyond it.
Consider this example:
#include <stdio.h>
int main(void) {
char str[5] = {'H', 'i', '\0', 'z'};
printf ("%s\n", str);
printf ("%c\n", str[3]);
return 0;
}
When you print the string
printf ("%s\n", str);
the output you will get is - Hi
because with %s format specifier, printf() writes every byte up to and not including the first null terminator [note the use of null character in the strings], but you can also print the 4th character of array as it is within the range of char array str though beyond first '\0' character
printf ("%c\n", str[3]);
the output you will get is - z
Additional:
Trying to access array beyond its size lead to undefined behavior which includes the program may execute incorrectly (either crashing or silently generating incorrect results), or it may fortuitously do exactly what the programmer intended.
It’s just a matter of convention. If you wanted to, you could totally write code that handled array termination (for arrays of any type) via some sentinel value. Here’s an example that does just that, arbitrarily using -1 as the sentinel:
int length(int arr[]) {
int i;
for (i = 0; arr[i] != -1; i++) {}
return i;
}
However, this is obviously utterly unpractical: You couldn’t use -1 in the array any longer.
By contrast, for C strings the sentinel value '\0' is less problematic because it’s expected that normal test won’t contain this character. This assumption is kind of valid. But even so there are obviously many strings which do contain '\0' as a valid character, and null-termination is therefore by no means universal.
One very common alternative is to store strings in a struct that looks something like this:
struct string {
unsigned int length;
char *buffer;
}
That is, we explicitly store a length alongside a buffer. This buffer isn’t null-terminated (although in practice it often has an additional terminal '\0' byte for compatibility with C functions).
Anyway, the answer boils down to: For C strings, null termination is a convenient convention. But it is only a convention, enforced by the C string functions (and by the C string literal syntax). You could use a similar convention for other array types but it would be prohibitively impractical. This is why other conventions developed for arrays. Notably, most functions that deal with arrays expect both an array and a length parameter. This length parameter determines where the array terminates.
My code is crashing because of a lack of the char '\0' at the end of some strings.
It's pretty clear to me why we have to use this termination char. My question is,
is there a problem adding a potential 2nd null character to a character array - to solve string problems?
I think it's cheaper just add a '\0' to every string than verify if it needs and then add it, but I don't know if it's a good thing to do.
is there a problem to have this char ('\0') twice at the end of a string?
This question lacks clarity as "string" means different things to people.
Let us use the C specification definition as this is a C post.
A string is a contiguous sequence of characters terminated by and including the first null character. C11 §7.1.1 1
So a string, cannot have 2 null characters as the string ends upon reaching its first one. #Michael Walz
Instead, re-parse to "is there a problem adding a potential 2nd null character to a character array - to solve string problems?"
A problem with attempting to add a null character to a string is confusion. The str...() functions work with C strings as defined above.
// If str1 was not a string, strcpy(str1, anything) would be undefined behavior.
strcpy(str1, "\0"); // no change to str1
char str2[] = "abc";
str2[strlen(str2)] = '\0'; // OK but only, re-assigns the \0 to a \0
// attempt to add another \0
str2[strlen(str2)+1] = '\0'; // Bad: assigning outside `str2[]` as the array is too small
char str3[10] = "abc";
str3[strlen(str3)+1] = '\0'; // OK, in this case
puts(str3); // Adding that \0 served no purpose
As many have commented, adding a spare '\0' is not directly attending the code's fundamental problem. #Haris #Malcolm McLean
That unposted code is the real issue that need solving #Yunnosch, and not by attempting to append a second '\0'.
I think it's cheaper just add a '\0' to every string than verify if it needs and then add it, but I don't know if it's a good thing to do.
Where would you add it? Let's assume we've done something like this:
char *p = malloc(32);
Now, if we know the allocated length, we could put a '\0' as the last character of the allocated area, as in p[31] = '\0'. But we don't how long the contents of the string are supposed to be. If there's supposed to be just foobar, then there'd still be 25 bytes of garbage, which might cause other issues if processed or printed.
Let alone the fact that if all you have is the pointer to the string, it's hard to know the length of the allocated area.
Probably better to fix the places where you build the strings to do it correctly.
Having '\0' is not a problem, unless you have not gone out of bounds of that char array.
You do have to understand that, having '\0' twice would mean, any string operation would not even know that there is a second '\0'. They will just read till the first '\0', and be with it. For them, the first '\0' is the Null terminating character and there should not be anything after that.
I am new to C and I am very much confused with the C strings. Following are my questions.
Finding last character from a string
How can I find out the last character from a string? I came with something like,
char *str = "hello";
printf("%c", str[strlen(str) - 1]);
return 0;
Is this the way to go? I somehow think that, this is not the correct way because strlen has to iterate over the characters to get the length. So this operation will have a O(n) complexity.
Converting char to char*
I have a string and need to append a char to it. How can i do that? strcat accepts only char*. I tried the following,
char delimiter = ',';
char text[6];
strcpy(text, "hello");
strcat(text, delimiter);
Using strcat with variables that has local scope
Please consider the following code,
void foo(char *output)
{
char *delimiter = ',';
strcpy(output, "hello");
strcat(output, delimiter);
}
In the above code,delimiter is a local variable which gets destroyed after foo returned. Is it OK to append it to variable output?
How strcat handles null terminating character?
If I am concatenating two null terminated strings, will strcat append two null terminating characters to the resultant string?
Is there a good beginner level article which explains how strings work in C and how can I perform the usual string manipulations?
Any help would be great!
Last character: your approach is correct. If you will need to do this a lot on large strings, your data structure containing strings should store lengths with them. If not, it doesn't matter that it's O(n).
Appending a character: you have several bugs. For one thing, your buffer is too small to hold another character. As for how to call strcat, you can either put the character in a string (an array with 2 entries, the second being 0), or you can just manually use the length to write the character to the end.
Your worry about 2 nul terminators is unfounded. While it occupies memory contiguous with the string and is necessary, the nul byte at the end is NOT "part of the string" in the sense of length, etc. It's purely a marker of the end. strcat will overwrite the old nul and put a new one at the very end, after the concatenated string. Again, you need to make sure your buffer is large enough before you call strcat!
O(n) is the best you can do, because of the way C strings work.
char delimiter[] = ",";. This makes delimiter a character array holding a comma and a NUL Also, text needs to have length 7. hello is 5, then you have the comma, and a NUL.
If you define delimiter correctly, that's fine (as is, you're assigning a character to a pointer, which is wrong). The contents of output won't depend on delimiter later on.
It will overwrite the first NUL.
You're on the right track. I highly recommend you read K&R C 2nd Edition. It will help you with strings, pointers, and more. And don't forget man pages and documentation. They will answer questions like the one on strcat quite clearly. Two good sites are The Open Group and cplusplus.com.
A "C string" is in reality a simple array of chars, with str[0] containing the first character, str[1] the second and so on. After the last character, the array contains one more element, which holds a zero. This zero by convention signifies the end of the string. For example, those two lines are equivalent:
char str[] = "foo"; //str is 4 bytes
char str[] = {'f', 'o', 'o', 0};
And now for your questions:
Finding last character from a string
Your way is the right one. There is no faster way to know where the string ends than scanning through it to find the final zero.
Converting char to char*
As said before, a "string" is simply an array of chars, with a zero terminator added to the end. So if you want a string of one character, you declare an array of two chars - your character and the final zero, like this:
char str[2];
str[0] = ',';
str[1] = 0;
Or simply:
char str[2] = {',', 0};
Using strcat with variables that has local scope
strcat() simply copies the contents of the source array to the destination array, at the offset of the null character in the destination array. So it is irrelevant what happens to the source after the operation. But you DO need to worry if the destination array is big enough to hold the data - otherwise strcat() will overwrite whatever data sits in memory right after the array! The needed size is strlen(str1) + strlen(str2) + 1.
How strcat handles null terminating character?
The final zero is expected to terminate both input strings, and is appended to the output string.
Finding last character from a string
I propose a thought experiment: if it were generally possible to find the last character
of a string in better than O(n) time, then could you not also implement strlen
in better than O(n) time?
Converting char to char*
You temporarily can store the char in an array-of-char, and that will decay into
a pointer-to-char:
char delimiterBuf[2] = "";
delimiterBuf[0] = delimiter;
...
strcat(text, delimiterBuf);
If you're just using character literals, though, you can simply use string literals instead.
Using strcat with variables that has local scope
The variable itself isn't referenced outside the scope. When the function returns,
that local variable has already been evaluated and its contents have already been
copied.
How strcat handles null terminating character?
"Strings" in a C are NUL-terminated sequences of characters. Both inputs to
strcat must be NUL-terminated, and the result will be NUL-terminated. It
wouldn't be useful for strcat to write an extra NUL-byte to the result if it
doesn't need to.
(And if you're wondering what if the input strings have multiple trailing
NUL bytes already, I propose another thought experiment: how would strcat know
how many trailing NUL-bytes there are in a string?)
BTW, since you tagged this with "best-practices", I'll also recommend that you take care not to write past the end of your destination buffers. Typically this means avoiding strcat and strcpy (unless you've already checked that the input strings won't overflow the destination) and using safer versions (e.g. strncat. Note that strncpy has its own pitfalls, so that's a poor substitute. There also are safer versions that are non-standard, such as strlcpy/strlcat and strcpy_s/strcat_s.)
Similarly, functions like your foo function always should take an additional argument specifying what the size of the destination buffer is (and documentation should make it explicitly clear whether that size accounts for a NUL terminator or not).
How can I find out the last character
from a string?
Your technique with str[strlen(str) - 1] is fine. As pointed out, you should avoid repeated, unnecessary calls to strlen and store the results.
I somehow think that, this is not the
correct way because strlen has to
iterate over the characters to get the
length. So this operation will have a
O(n) complexity.
Repeated calls to strlen can be a bane of C programs. However, you should avoid premature optimization. If a profiler actually demonstrates a hotspot where strlen is expensive, then you can do something like this for your literal string case:
const char test[] = "foo";
sizeof test // 4
Of course if you create 'test' on the stack, it incurs a little overhead (incrementing/decrementing stack pointer), but no linear time operation involved.
Literal strings are generally not going to be so gigantic. For other cases like reading a large string from a file, you can store the length of the string in advance as but one example to avoid recomputing the length of the string. This can also be helpful as it'll tell you in advance how much memory to allocate for your character buffer.
I have a string and need to append a
char to it. How can i do that? strcat
accepts only char*.
If you have a char and cannot make a string out of it (char* c = "a"), then I believe you can use strncat (need verification on this):
char ch = 'a';
strncat(str, &ch, 1);
In the above code,delimiter is a local
variable which gets destroyed after
foo returned. Is it OK to append it to
variable output?
Yes: functions like strcat and strcpy make deep copies of the source string. They don't leave shallow pointers behind, so it's fine for the local data to be destroyed after these operations are performed.
If I am concatenating two null
terminated strings, will strcat
append two null terminating characters
to the resultant string?
No, strcat will basically overwrite the null terminator on the dest string and write past it, then append a new null terminator when it's finished.
How can I find out the last character from a string?
Your approach is almost correct. The only way to find the end of a C string is to iterate throught the characters, looking for the nul.
There is a bug in your answer though (in the general case). If strlen(str) is zero, you access the character before the start of the string.
I have a string and need to append a char to it. How can i do that?
Your approach is wrong. A C string is just an array of C characters with the last one being '\0'. So in theory, you can append a character like this:
char delimiter = ',';
char text[7];
strcpy(text, "hello");
int textSize = strlen(text);
text[textSize] = delimiter;
text[textSize + 1] = '\0';
However, if I leave it like that I'll get zillions of down votes because there are three places where I have a potential buffer overflow (if I didn't know that my initial string was "hello"). Before doing the copy, you need to put in a check that text is big enough to contain all the characters from the string plus one for the delimiter plus one for the terminating nul.
... delimiter is a local variable which gets destroyed after foo returned. Is it OK to append it to variable output?
Yes that's fine. strcat copies characters. But your code sample does no checks that output is big enough for all the stuff you are putting into it.
If I am concatenating two null terminated strings, will strcat append two null terminating characters to the resultant string?
No.
I somehow think that, this is not the correct way because strlen has to iterate over the characters to get the length. So this operation will have a O(n) complexity.
You are right read Joel Spolsky on why C-strings suck. There are few ways around it. The ways include either not using C strings (for example use Pascal strings and create your own library to handle them), or not use C (use say C++ which has a string class - which is slow for different reasons, but you could also write your own to handle Pascal strings more easily than in C for example)
Regarding adding a char to a C string; a C string is simply a char array with a nul terminator, so long as you preserve the terminator it is a string, there's no magic.
char* straddch( char* str, char ch )
{
char* end = &str[strlen(str)] ;
*end = ch ;
end++ ;
*end = 0 ;
return str ;
}
Just like strcat(), you have to know that the array that str is created in is long enough to accommodate the longer string, the compiler will not help you. It is both inelegant and unsafe.
If I am concatenating two null
terminated strings, will strcat append
two null terminating characters to the
resultant string?
No, just one, but what ever follows that may just happen to be nul, or whatever happened to be in memory. Consider the following equivalent:
char* my_strcat( char* s1, const char* s2 )
{
strcpy( &str[strlen(str)], s2 ) ;
}
the first character of s2 overwrites the terminator in s1.
In the above code,delimiter is a local
variable which gets destroyed after
foo returned. Is it OK to append it to
variable output?
In your example delimiter is not a string, and initialising a pointer with a char makes no sense. However if it were a string, the code would be fine, strcat() copies the data from the second string, so the lifetime of the second argument is irrelevant. Of course you could in your example use a char (not a char*) and the straddch() function suggested above.