Encoding two vectors in a function - c++

Question

Hey I'm writing a function that takes two std::vector<std::string> and returns a third std::vector<std::string>.
The function is going to encode the two vectors together and create the 3rd vector.
I'm currently debugging this to find out why it's not working, and I keep getting: vector subscript out of range. As far as I can tell it's crashing at this line:
if (file2[i].size() < file1[i].size())
Can I use size() to get the size of the element at i?
std::vector<std::string> Encode(std::vector<std::string> &file1,
std::vector<std::string> &file2)
{
std::vector<std::string> file3;
std::string temp;
for (unsigned int i = 0; i < file1.size(); i++) {
for (unsigned int x = 0; x < file1[i].size(); x++) {
if (file2[i].size() < file1[i].size()) {
for (unsigned int t = 0; t < file2[i].size(); t++) {
file3[i][x] = (int)file1[i][x] + (int)file2[i][t];
}
} else if (file2[i].size() > file1[i].size()) {
file3[i][x] = (int)file1[i][x] + (int)file2[i][x];
}
if (file3[i][x] > 126) {
file3[i][x] = file3[i][x] % 127;
} else {
file3[i][x] = file3[i][x] + 32;
}
}
}
return file3;
}
Any idea what's going on here?

Answer 1

I'd be very much inclined to simplify by factoring. At the lowest layer is a combine function to combine two chars into one:
char combine(char a, char b)
{
char result = a+b;
if (result > 126)
return result % 127;
return result+32;
}
The next level up would be to iterate through each of the letters in two strings of possibly different sizes. The algorithm works for differing length strings by "recycling" through the shorter string.
std::string mix(const std::string &first, const std::string &second)
{
unsigned len1 = first.length();
unsigned len2 = second.length();
if (len1 < len2)
return mix(second, first);
std::string result;
// if the strings are of different lengths, first is now the longer
unsigned j=0;
for (unsigned i=0; i < len1; ++i, ++j) {
if (j >= len2)
j = 0;
result.push_back(combine(first[i], second[j]));
}
return result;
}
Finally, the combination of the vector of string is much simpler:
std::vector<std::string> Encode(const std::vector<std::string> &file1,
const std::vector<std::string> &file2)
{
std::vector<std::string> file3;
assert(file1.size() == file2.size());
for (unsigned int i = 0; i < file1.size(); i++) {
file3.push_back(mix(file1[i], file2[i]));
}
return file3;
}
Note that the code currently uses an assert to assure that the two vectors are the same length, but this is probably an artificial constraint. Real code should either assure that they are the same length or do something else to handle that case. Since it's not clear what your function is intended to do, I've left it to you to decide how to handle it, but with the assert as a placeholder to remind you that it does need to be addressed.
Finally, some driver code using C++11:
int main()
{
std::vector<std::string> english{"one", "two", "three", "four"};
std::vector<std::string> spanish{"uno", "dos", "tres", "cuatro"};
auto result = Encode(english, spanish);
std::copy(result.begin(), result.end(),
std::ostream_iterator<std::string>(std::cout, " "));
}
Note, too that I've used push_back to append to the end of the strings and const declarations for the passed strings.

Answer 2

Try these three sets of inputs:
1. file1 is bigger than file2
2. file2 is bigger than file1
3. file1 is equal to file2 in size.
Let us know the cases when the error was reproduced and when it was not.
I think by this stage you will solve the problem by yourself.
If not,
write contents of the (smallest possible) file1 and file2 that reproduced the error.

Answer 3

Some problems enumerated:
-You are assuming that file1 and file2 have the same size or at least that file1 has size <= file2 (in the other case would cause invalid memory access in line if (file2[i].size() < file1[i].size()) {) and are not checking in the function for that. At least add an assert statement or a checking.
-You are initializing file3 empty, and are indexing later in the function.
-The other problem is what happen when file1[i] and file2[i] have the same length, this option is not cover in the if-else, the cover options are < and > but not ==.
-You are accessing invalid memory with statements like file3[i][x], because the strings in file3[i] are initialized empty, for that, don't contain any character.
This is the more close that i can get without known the exact steps of the encode algorithm
#include <iostream>
#include <vector>
#include <boost/lexical_cast.hpp>
using namespace std;
std::vector<std::string> Encode(std::vector<std::string> &file1,
std::vector<std::string> &file2) {
assert(file1.size() <= file2.size());
std::vector<std::string> file3(file1.size());
std::string temp;
for (unsigned int i = 0; i < file1.size(); i++) {
for (unsigned int x = 0; x < file1[i].size(); x++) {
int enc = 0;
if (file2[i].size() <= file1[i].size()) {
for (unsigned int t = 0; t < file2[i].size(); t++) {
enc = (int)file1[i][x] + (int)file2[i][t];
}
}
else if (file2[i].size() > file1[i].size()) {
enc = (int)file1[i][x] + (int)file2[i][x];
}
if (enc > 126) {
file3[i] += (enc % 127);
}
else {
file3[i] += (enc + 32);
}
}
}
return file3;
}
int main(int argc, char *argv[]) {
std::vector<std::string> a{ "1", "2", "3" };
std::vector<std::string> b{ "6", "7", "8" };
for (const auto& s : a)
std::cout << s << std::endl;
for (const auto& s : b)
std::cout << s << std::endl;
auto c = Encode(a, b);
for (const auto& s : c)
std::cout << s << std::endl;
return 0;
}