Why can't I call my function recursively? - c++

I'm trying to compile this code in order to reverse a string:
void reverse(char *str, int n)
{
if (n==0 || n==1) {
return; //acts as quit
} else {
char i = str[0]; //1st position of string
char j = str[n-1]; //Last position of string
char temp = str[i];
str[i] = str[j]; //Swap
str[j] = temp;
reverse(str[i+1],n-1); // <-- this line
}
}
#include <iostream>
int main()
{
char *word = "hello";
int n = sizeof word;
reverse(word, n);
std::cout << word << std::endl;
return 0;
}
The compiler reports an error where I call reverse() recursively:
invalid conversion from char to char* at reverse(str[i+1], n-1).
Why?
Any advice on other issues in my code is also welcome.

str[i+1] is a character, not a pointer to a character; hence the error message.
When you enter the function, str points to the character you're going to swap with the n:th character away from str.
What you need to do in the recursion is to increment the pointer so it points to the next character.
You also need to decrease n by two, because it should be a distance from str + 1, not from str.
(This is easy to get wrong; see the edit history of this answer for an example.)
You're also using the characters in the strings as indexes into the strings when swapping.
(If you had the input "ab", you would do char temp = str['a']; str['a'] = str['b']; str['b'] = temp;. This is obviously not correct.)
str[0] is not the position of the first character, it is the first character.
Use std::swap if you're allowed to, otherwise see below.
More issues: you shouldn't use sizeof word, as that is either 4 or 8 depending your target architecture - it's equivalent to sizeof(char*).
You should use strlen to find out how long a string is.
Further, you should get a warning for
char *word = "hello";
as that particular conversion is dangerous - "hello" is a const array and modifying it is undefined.
(It would be safe if you never modified the array, but you are, so it isn't.)
Copy it into a non-const array instead:
char word[] = "hello";
and increase the warning level of your compiler.
Here's a fixed version:
void reverse(char *str, int n)
{
if(n <= 1) // Play it safe even with negative n
{
return;
}
else
{
// You could replace this with std::swap(str[0], str[n-1])
char temp = str[0]; //1st character in the string
str[0] = str[n-1]; //Swap
str[n-1] = temp;
// n - 2 is one step closer to str + 1 than n is to str.
reverse(str + 1, n - 2);
}
}
int main()
{
char word[] = "hello";
// sizeof would actually work here, but it's fragile so I prefer strlen.
reverse(word, strlen(word));
std::cout << word << std::endl;
}

I'm going to dissect your code, as if you'd posted over on Code Review. You did ask for other observations, after all...
Firstly,
char *word = "hello";
Your compiler should warn you that pointing a char* at a literal string is undefined behaviour (if not, make sure that you have actually enabled a good set of warnings. Many compilers emit very few warnings by default, for historical reasons). You need to ensure that you have a writable string; for that you can use a char[]:
char word[] = "hello";
The next line
int n = sizeof word;
has now changed meaning, but is still wrong. In your original code, it was the size of a pointer to char, which is unlikely to be the same as the length of the word "hello". With the change to char[], it's now the size of an array of 6 characters, i.e. 6. The sixth character is the NUL that ends the string literal. Instead of the sizeof operator, you probably want to use the strlen() function.
Moving on to reverse():
You read characters from positions in the string, and then use those characters to index it. That's not what you want, and GCC warns against indexing using plain char as it may be signed or unsigned. You just want to index in one place, and your i and j are unnecessary.
Finally, the question you asked. str[i+1] is the character at position i+1, but your function wants a pointer to character, which is simply str+i+1. Or, since we worked out we don't want i in there, just str+1.
Note also that you'll need to subtract 2 from n, not 1, as it will be used as a count of characters from str+1. If you only subtract 1, you'll always be swapping with the last character, and you'll achieve a 'roll' rather than a 'reverse'.
Here's a working version:
void reverse(char *str, int n)
{
if (n < 2)
// end of recursion
return; //acts as quit
char temp = str[0];
str[0] = str[n-1]; //Swap
str[n-1] = temp;
reverse(str+1,n-2);
}
#include <iostream>
#include <cstring>
int main()
{
char word[] = "hello";
int n = std::strlen(word);
reverse(word, n);
std::cout << word << std::endl;
}
We can make further changes. For example, we could use std::swap to express the switching more clearly. And we could pass a pair of pointers instead of a pointer and a length:
#include <utility> // assuming C++11 - else <algorithm>
void reverse(char *str, char *end)
{
if (end <= str)
// end of recursion
return;
std::swap(*str, *end);
reverse(str+1, end-1);
}
and invoke it with reverse(word, word+n-1).
Finally (as I'm not going to mention std::reverse()), here's the idiomatic iterative version:
void reverse(char *str, char *end)
{
while (str < end)
std::swap(*str++, *end--);
}

use like this :
reverse(&str[i+1],n-1);
pass address of the (i+1)th position not value.

Related

How do I reverse a c string without the use of strlen?

I'm trying to implement a void function that takes a c string as its only parameter and reverses it and prints it. Below is my attempt at a solution however I'm not sure how to go about this problem.
void printBackwards(char forward[]) {
int i = 0;
char backwards[];
while (forward[i] != '\0') {
backwards[i] = forward[-i - 1];
i++;
}
cout << backwards;
}
Under such a condition, I guess you are expected to use recursion.
void printBackwards(char forward[]) {
if (!forward[0])
return;
printBackwards(forward + 1);
cout << forward[0];
}
Not being able to use strlen, we'll calculate it ourselves using a simple for loop. Then dynamically allocate a suitable buffer (add one character for the null terminating char, and I "cheated" by using calloc to zero the memory so I don't have to remember to set the null terminator. Then anoher simple loop to copy the original into the result in reverse.
#include <stdlib.h>
#include <stdio.h>
char *rev(char *s) {
size_t i;
char *s2 = s; // A pointer to the beginning as our first loop modifies s
for (i = 0; *s; s++, i++);
char *result = calloc(0, i + 1);
if (!result) return NULL; // In case calloc didn't allocate the requested memory.
for (size_t j = 0; j < i; j++)
result[j] = s2[i - j - 1];
return result;
}
Assuming you want to reverse the string rather than just printing it in reverse order, you first need to find the last character location (actually the position of the null terminator). Pseudo-code below (since this is an educational assignment):
define null_addr(pointer):
while character at pointer is not null terminator:
increment pointer
return pointer
Then you can use that inside a loop where you swap the two characters and move the pointers toward the center of the string. As soon as the pointers become equal or pass each other the string is reversed:
define reverse(left_pointer):
set right_pointer to null_addr(left_pointer)
while right_pointer > left_pointer plus one:
decrement right_pointer
swap character at left_pointer with character at right_pointer
increment left_pointer
Alternatively (and this appears to be the case since your attempt doesn't actually reverse the original string), if you need to print the string in reverse order without modifying it, you still find the last character. Then you run backwards through the string printing each character until you reach the first. That can be done with something like:
define print_reverse(pointer):
set right_pointer to null_addr(pointer)
while right_pointer > pointer:
decrement right_pointer
print character at right_pointer
That's probably better than creating a new string to hold the reverse of the original, and then printing that reverse.
One thing you should keep in mind. This very much appears to be a C-centric question, not a C++ one (it's using C strings rather than C++ strings, and uses C header files). If that's the case, you should probably avoid things like cout.
By using abstractions, like , your code will be much better at communication WHAT it is doing instead of HOW it is doing it.
#include <iostream>
#include <string>
#include <ranges>
int main()
{
std::string hello{ "!dlrow olleH" };
for (const char c : hello | std::views::reverse)
{
std::cout << c;
}
return 0;
}
Use a template
#include <iostream>
template<int N, int I=2>
void printBackwards(char (&forward)[N]) {
std::cout << forward[N-I];
if constexpr (I<N) printBackwards<N, I+1>(forward);
}
int main() {
char test[] = "elephant";
printBackwards(test);
}
While there seems to be several working answers, I thought I'd throw my hat in the stack (pun intended) since none of them take advantage of a FILO data structure (except #273K's answer, which uses a stack implicitly instead of explicitly).
What I would do is simply push everything onto a stack and then print the stack:
#include <stack>
#include <iostream>
void printBackwards(char forward[]) {
// Create a stack to hold our reversed string
std::stack<char> stk;
// Iterate through the string until we hit the null terminator
int i = 0;
while (forward[i] != '\0'){
stk.push(forward[i]);
++i;
}
// Iterate through the stack and print each character as we pop() it
while (stk.size() > 0){
std::cout << stk.top();
stk.pop();
}
// Don't forget the newline (assuming output lines should be separated)
std::cout << '\n';
}
int main(int argc, char* argv[]){
char s[] = "This is a string";
printBackwards(s);
return 0;
}
Hi guys as promised I have come back to add my own answer. This is my own way using array subscripts and using what I currently know.
#include <iostream>
using namespace std;
void printBackwards(char[]);
int main()
{
char word[] = "apples";
printBackwards(word);
return 0;
}
void printBackwards(char word[]) {
char* temp = word;
int count = 0;
while (*temp++ != '\0') {
count++;
}
for (int i = count - 1; i >= 0; i--) {
cout << word[i];
}
}
You can make a fixed-size buffer and create new ones if needed. Fill it reverse by moving the string offset back with every inserted character. Chars exceeding the buffer are returned to be processed later, so you can make a list of such buffers:
template<int SIZE>
struct ReversedCStr
{
static_assert(SIZE > 10); // just some minimal size treshold
// constexpr
ReversedCStr(char const* c_str, char const** tail = nullptr) noexcept
{
for(buffer[offset] = '\0'; *c_str != '\0';)
{
buffer[--offset] = *c_str++;
if(offset == 0) break;
}
if(tail) *tail = c_str;
}
//constexpr
char const* c_str() const noexcept { return buffer.data()+offset;};
private:
size_t offset = SIZE -1;
std::array<char,SIZE> buffer;
};
The tag is 'C++' so I assume you use C++ not C. The following code is C++11 so it should fit in every modern project. I posted the working example on godbolt.org.
It doesn't allocate memory, and is completely exception-free. The maximum memory wasted is {buffer_size + sizeof(char*)*number_of_chunks}, and can be easily turned into a list of reversed chunks like this:
char const* tail;
std::vector<ReversedCStr<11>> vec;
for(vec.emplace_back(str,&tail); *tail != '\0';)
vec.emplace_back(tail,&tail);

string to character conversion error?

After trying for about 1 hour, my code didn't work because of this:
void s_s(string const& s, char data[10])
{
for (int i = 0; i < 10; i++)
data[i] = s[i];
}
int main()
{
string ss = "1234567890";
char data[10];
s_s("1234567890", data);
cout << data << endl;//why junk
}
I simply don't understand why the cout displays junk after the char array. Can someone please explain why and how to solve it?
You need to null terminate your char array.
std::cout.operator<<(char*) uses \0 to know where to stop.
Your char[] decays to char* by the way.
Look here.
As already mentioned you want to NUL terminate your array, but here's something else to consider:
If s is your source string, then you want to loop to s.size(), so that you don't loop past the size of your source string.
void s_s(std::string const& s, char data[20])
{
for (unsigned int i = 0; i < s.size(); i++)
data[i] = s[i];
data[s.size()] = '\0';
}
Alternatively, you can try this:
std::copy(ss.begin(), ss.begin()+ss.size(),
data);
data[ss.size()] = '\0';
std::cout << data << std::endl;
You have ONLY allocated 10 bytes for data
The string is actually 11 bytes since there is an implied '\0' at the end
At a minimum you should increase the size of data to 11, and change your loop to copy the '\0' as well
The function std::ostream::operator<< that you are trying to use in the last line of the main will take your char array as a pointer and will print every char until the null sentinel character is found (the character is \0).
This sentinel character is generally generated for you in statements where a C-string literal is defined:
char s[] = "123";
In the above example sizeof(s) is 4 because the actual characters stored are:
'1', '2', '3', '\0'
The last character is fundamental in tasks that require to loop on every char of a const char* string, because the condition for the loop to terminate, is that the \0 must be read.
In your example the "junk" that you see are the bytes following the 0 char byte in the memory (interpreted as char). This behavior is clearly undefined and can potentially lead the program to crash.
One solution is to obviously add the \0 char at the end of the char array (of course fixing the size).
The best solution, though, is to never use const char* for strings at all. You are correctly using std::string in your example, which will prevent this kind of problems and many others.
If you ever need a const char* (for C APIs for example) you can always use std::string::c_str and retrieve the C string version of the std::string.
Your example could be rewritten to:
int main(int, char*[]) {
std::string ss = "1234567890";
const char* data = ss.c_str();
std::cout << data << std::endl;
}
(in this particular instance, a version of std::ostream::operator<< that takes a std::string is already defined, so you don't even need data at all)

Grabbing a portion of a string like substr

So I'm making a function that is similar to SubStr. This is an assignment so I cannot use the actual function to do this. So far I have created a function to take a string and then get the desired substring. My problem is returning the substring. In the function when I do Substring[b] = AString[b]; the substring is empty, but if I cout from inside the function I get the desired substring. So what is wrong with my code?
Here is a working demo: http://ideone.com/4f5IpA
#include <iostream>
using namespace std;
void subsec(char AString[], char Substring[], int start, int length);
int main() {
char someString[] = "abcdefg";
char someSubString[] = "";
subsec(someString, someSubString, 1, 3);
cout << someSubString << endl;
return 0;
}
void subsec(char AString[], char Substring[], int start, int length) {
for (int b = start; b <= length; b++) {
Substring[b] = AString[b];
}
}
Maybe this does what you're looking for? It's hard to say as your initial implementation used the length parameter as more of an end position.
#include <iostream>
using namespace std;
void subsec(char AString[], char Substring[], int start, int length)
{
const int end = start + length;
int pos = 0;
for(int b = start; b < end; ++b)
{
Substring[pos++] = AString[b];
}
Substring[pos] = 0;
}
int main()
{
char someString[50] = "abcdefghijklmnopqrstuvwxyz";
char someSubString[50];
subsec(someString, someSubString, 13, 10);
cout << someSubString << endl;
return 0;
}
There are several problems with the code:
1) The char arraysomeSubString has size 1 which cannot hold the substring.
2) The subsec is not correctly implemented, you should copy to the Substring from index 0.
Also remember to add \0 at the end of the substring.
void subsec(char AString[], char *Substring, int start, int length) {
int ii = 0;
for (int jj = start; jj <= length; jj++, ii++) {
Substring[ii] = AString[jj];
}
Substring[ii] = '\0';
}
You need to allocate more than 1 byte for someSubString i.e.
char someSubString[] = "xxxxxxxxxxxxxxxxxx";
or just
char someSubString[100];
if you know the max size you'll ever need.
Either would allocate enough space for the string you're copying to it. Then, you're not doing anything about the terminating 0 either. At the end of a C-style string there needs to be a terminating null to signify end of string. Otherwise cout will print something like;
abcdefgxxxxxxx
if you initialized with x's as I indicated.
There are a few problems with your code as it stands. Firstly, as your compiler is no doubt warning you, in C++ a string literal has type const char[], not just char[].
Secondly, you need to have enough space to store your substring. A good way to do this is for your function to allocate the space it needs, and then pass back a pointer to this memory. This is the way things are typically done in C code. The only thing is that you have to remember to delete the allocated array when you're done with it. (There are other, better ways to do this in C++, with things like smart pointers and wrapper objects, but those come later :-) ).
Thirdly, you'll have a problem if you request a length which is actually longer than the passed-in string -- you'll run off the end and start copying random memory (or just crash), which is definitely not what you want. C strings are terminated with a "nul byte" -- so you need to check whether you've come across this.
Speaking of the nul, you need to make sure that your substring ends with one.
Lastly, it's not really a problem but there's no need for the start parameter, you can just pass a pointer to the middle of the array if you want to.
char* substring(const char* str, int length)
{
// Allocate memory for substring;
char* subs = new char[length+1];
// Copy characters from given string
int i = 0;
while (i < length && str[i] != '\0') {
subs[i] = str[i];
i++;
}
// Append the nul byte
subs[i] = '\0';
return subs;
}
int main()
{
const char someString[] = "foobarbaz"; // Note -- must be const in C++
char* subs = substring(someString + 3, 3);
assert(strcmp(subs, "bar") == 0);
delete subs;
}

Convert non-null-terminated char* to int

I am working on some code that reads in a data file. The file frequently contains numeric values of various lengths encoded in ASCII that I need to convert to integers. The problem is that they are not null-terminated, which of course causes problems with atoi. The solution I have been using is to manually append a null to the character sequence, and then convert it.
This is the code that I have been using; it works fine, but it seems very kludgy.
char *append_null(const char *chars, const int size)
{
char *tmp = new char[size + 2];
memcpy(tmp, chars, size);
tmp[size + 1] = '\0';
return tmp;
}
int atoi2(const char *chars, const int size)
{
char *tmp = append_null(chars, size);
int result = atoi(tmp);
delete[] tmp;
return result;
}
int main()
{
char *test = new char[20];
test[0] = '1';
test[1] = '2';
test[2] = '3';
test[3] = '4';
cout << atoi2(test, 4) << endl;
}
I am wondering if there is a better way to approach this problem.
Fixed-format integer conversion is still well within handroll range where the library won't do:
size_t mem_tozd_rjzf(const char *buf, size_t len) // digits only
{
int n=0;
while (len--)
n = n*10 + *buf++ - '0';
return n;
}
long mem_told(const char *buf, size_t len) // spaces, sign, digits
{
long n=0, sign=1;
while ( len && isspace(*buf) )
--len, ++buf;
if ( len ) switch(*buf) {
case '-': sign=-1; \
case '+': --len, ++buf;
}
while ( len-- && isdigit(*buf) )
n = n*10 + *buf++ -'0';
return n*sign;
}
In C++11, you can say std::stoi(std::string(chars, size)), all from <string>.
int i = atoi(std::string(chars, size).c_str());
Your method will work, although you should only need size+1 for appending the null and the null will go at position size. Currently, your test code doesn't actually make the function call, but I'll assume that you have a way to determine when the null-terminated characters end. If possibly, I'd recommend making the null termination there so that you don't have to worry about catching cases where you hit an exception before you can deallocate the memory (memory which, honestly, may or may not have been allocated if you start catching exceptions).
std::string str = "1234";
boost::lexical_cast<int>(str); // 1234
The problem as formulated requires to construct a string given an array of known size, then converting its text into a numeric value.
To convert text into values, C++ has a unified mechanism: streams.
In your case, you can do the following:
int i = 0;
std::stringstream(std::string(yourbuffer, yoursize)) >> i;
This will completely avoid any plain old C reference.
But, since -as you say- all values come from a file... why just don't read the file itself as a stream via std::fstream ?
The question says (emph mine):
The file frequently contains numeric values of various lengths encoded
in ASCII that I need to convert to integers. The problem is that they
are not null-terminated, which of course causes problems with atoi.
This does not really pose a problem, as, if we look at the docs for atoi or strtol, they clearly state:
Function discards any whitespace characters until first non-whitespace
character is found. Then it takes as many characters as possible to
form a valid integer number representation and converts them to
integer value.
That means, it doesn't matter at all that the numbers aren't null terminated, as long as they are delimited by something that stops conversion.
And if they are not delimited, then you have to know the size, and when you know the size, I would also recommend a hand-coded solution like in the other answer.
I know this answer is not answering OP's question, but it helps if your source of char* is a char array with known size.
Live demo
#include <fmt/core.h>
#include <type_traits>
#include <iostream>
// SFINAE fallback
template<typename T, typename =
std::enable_if< std::is_pointer<T>::value >
>
int charArrayToInt(const T arr){ // Fall back for user friendly compiler errors
static_assert(false == std::is_pointer<T>::value, "`charArrayToInt()` dosen't allow conversion from pointer!");
return -1;
}
// Valid for both null or non-null-terminated char array
template<size_t sz>
int charArrayToInt(const char(&arr)[sz]){
// It doesn't matter whether it's null terminated or not
std::string str(arr, sz);
return std::stof(str);
}
int main() {
char number[2] = {'4','2'};
int ret = charArrayToInt(number);
fmt::print("The answer is {}. ", ret);
return 0;
}

Reversing a string, weird output c++

Okay, so I'm trying to reverse a C style string in C++ , and I'm coming upon some weird output. Perhaps someone can shed some light?
Here is my code:
int main(){
char str[] = "string";
int strSize = sizeof(str)/sizeof(char);
char str2[strSize];
int n = strSize-1;
int i =0;
while (&str+n >= &str){
str2[i] = *(str+n);
n--;
i++;
}
int str2size = sizeof(str)/sizeof(char);
int x;
for(x=0;x<str2size;x++){
cout << str2[x];
}
}
The basic idea here is just making a pointer point to the end of the string, and then reading it in backwards into a new array using pointer arithmetic.
In this particular case, I get an output of: " gnirts"
There is an annoying space at the beginning of any output which I'm assuming is the null character? But when I try to get rid of it by decrementing the strSize variable to exclude it, I end up with some other character on the opposite end of the string probably from another memory block.
Any ideas on how to avoid this? PS: (would you guys consider this a good idea of reversing a string?)
A valid string should be terminated by a null character. So you need to keep the null character in its original position (at the end of the string) and only reverse the non-null characters. So you would have something like this:
str2[strSize - 1] = str[strSize - 1]; // Copy the null at the end of the string
int n = strSize - 2; // Start from the penultimate character
There is an algorithm in the Standard Library to reverse a sequence. Why reinvent the wheel?
#include <algorithm>
#include <cstring>
#include <iostream>
int main()
{
char str[] = "string";
std::reverse(str, str + strlen(str)); // use the Standard Library
std::cout << str << '\n';
}
#ildjarn and #Blastfurnace have already given good ideas, but I think I'd take it a step further and use the iterators to construct the reversed string:
std::string input("string");
std::string reversed(input.rbegin(), input.rend());
std::cout << reversed;
I would let the C++ standard library do more of the work...
#include <cstddef>
#include <algorithm>
#include <iterator>
#include <iostream>
int main()
{
typedef std::reverse_iterator<char const*> riter_t;
char const str[] = "string";
std::size_t const strSize = sizeof(str);
char str2[strSize] = { };
std::copy(riter_t(str + strSize - 1), riter_t(str), str2);
std::cout << str2 << '\n';
}
while (&str+n >= &str){
This is nonsense, you want simply
while (n >= 0) {
and
str2[i] = *(str+n);
should be the much more readable
str2[i] = str[n];
Your while loop condition (&str+n >= &str) is equivalent to (n >= 0).
Your *(str+n) is equivalent to str[n] and I prefer the latter.
As HappyPixel said, your should start n at strSize-2, so the first character copied will be the last actual character of str, not the null termination character of str.
Then after you have copied all the regular characters in the loop, you need to add a null termination character at the end of the str2 using str2[strSize-1] = 0;.
Here is fixed, working code that outputs "gnirts":
#include <iostream>
using namespace std;
int main(int argc, char **argv){
char str[] = "string";
int strSize = sizeof(str)/sizeof(char);
char str2[strSize];
int n = strSize-2; // Start at last non-null character
int i = 0;
while (n >= 0){
str2[i] = str[n];
n--;
i++;
}
str2[strSize-1] = 0; // Add the null terminator.
int str2size = sizeof(str)/sizeof(char);
int x;
cout << str2;
}