C++: English to Pig Latin - c++

Question

I am receiving an error from the following code when I try to dynamically allocate the array (seen after my attempt to incrementing through each letter in the users array using the bool function). This is the error:
main.cpp: In function ‘Word* splitSentene(std::string, int&)’:
main.cpp:81:32: error: cannot convert ‘std::string* {aka std::basic_string*}’ to ‘Word*’ in assignment
words = new string[i];
I am trying to count how many words the user inputs and dynamically allocate an array for the string of words. This is my code thus far:
#include <iostream>
#include <cctype>
#include <string>
using namespace std;
struct Word
{
string english; // English sentence
string piglatin; // Pig latin sentence
};
// PT 1. Function prototype
Word * splitSentence(const string words, int &size){};
int main()
{
string userSentence;
int size;
// Get the users sentence to convert to pig latin
cout << "Please enter a string to convert to pig latin:\n";
getline(cin, userSentence);
// Directs to Word * splitSentence function
Word* tempptr = splitSentence(userSentence, size);
delete [] tempptr;
return 0;
}
//PT 1. Analyze the sentence
Word * splitSentene(const string words, int &size)
{
bool flag = true;
int num = 0;
for (int i = 0; i < words.length() + 1; i++)
{
//test for white space, then when you hit the first alphabetical character after a space,
//increment up the size of the array
if (isspace(words[i]))
flag = true;
if (isalpha(words[i]));
{
if (flag == true)
{
flag = false;
cout << words[i++];
}
}
// Dynamically allocate the array for the words
Word *sentence = nullptr;
sentence = new string[i];
}
}
Here are the pt 1 instructions for further clarification:
PT. 1) Write a function that takes in an English sentence as one string. This function should first calculate how many “words” are in the sentence (words being substrings separated by whitespace). It should then allocate a dynamic array of size equal to the number of words. The array contains Word structures (i.e. array of type Word). The function would then store each word of that sentence to the english field of the corresponding structure. The function should then return this array to the calling function using the return statement, along with the array size using a reference parameter.
This function should also remove all capitalization and special characters other than letters. Implement the function with the following prototype:
Word * splitSentence(const string words, int &size);
This is my first post here, so I will appreciate any input on how to dynamically allocate the array and format it (if I have successfully coded how to count the words in the sentence the user inputs). If more information needs to be provided, let me know!

Answer 1

The compiler error is because you are trying to assign a string[] array to a Word* pointer. You need to allocate a Word[] array instead.
You also have other errors in your code:
You have an erroneous {} at the end of the declaration of splitSentence().
You misspelled splitSentene in the defintion of splitSentence().
You have an erroneous ; on if (isalpha(words[i]));
You are not return'ing the array that you allocate.
In fact, you are not even following the instructions properly at all. You are not "calculating how many words are in the sentence" BEFORE allocating the array (you tried, but you are doing the allocation in the wrong place), and you are not filling the array at all, let alone "removing all capitalization and special characters other than letters".
Try something more like this:
#include <iostream>
#include <cctype>
#include <string>
using namespace std;
struct Word
{
string english; // English sentence
string piglatin; // Pig latin sentence
};
// PT 1. Function prototype
Word* splitSentence(const string words, int &size);
int main()
{
string userSentence;
int size;
// Get the users sentence to convert to pig latin
cout << "Please enter a string to convert to pig latin:\n";
getline(cin, userSentence);
// Directs to Word * splitSentence function
Word* tempptr = splitSentence(userSentence, size);
delete [] tempptr;
return 0;
}
//PT 1. Analyze the sentence
Word* splitSentence(const string words, int &size)
{
bool flag = true;
int num = 0;
char ch;
for (int i = 0; i < words.length(); ++i)
{
ch = words[i];
if (isalpha(ch))
{
if (flag)
{
flag = false;
++num;
}
}
else if (isspace(ch))
{
flag = true;
}
}
Word *sentence = new Word[num];
int index = -1;
flag = true;
num = 0;
for (int i = 0; i < words.length(); ++i)
{
ch = words[i];
if (isalpha(ch))
{
if (flag)
{
flag = false;
++num;
++index;
}
if (isupper(ch))
{
ch = tolower(ch);
}
sentence[index].english += ch;
}
else if (isspace(ch))
{
flag = true;
}
}
size = num;
return sentence;
}
Live Demo
That being said, this will be much easier to implement splitSentence() if you could use std::istringstream and std::vector and other C++ idioms, instead of using C idioms, eg:
#include <iostream>
#include <sstream>
#include <cctype>
#include <string>
#include <vector>
#include <algorithm>
using namespace std;
struct Word
{
string english; // English sentence
string piglatin; // Pig latin sentence
};
// PT 1. Function prototype
Word* splitSentence(const string words, int &size);
int main()
{
string userSentence;
int size;
// Get the users sentence to convert to pig latin
cout << "Please enter a string to convert to pig latin:\n";
getline(cin, userSentence);
// Directs to Word * splitSentence function
Word* tempptr = splitSentence(userSentence, size);
delete [] tempptr;
return 0;
}
//PT 1. Analyze the sentence
Word* splitSentence(const string words, int &size)
{
istringstream iss(words);
vector<string> vec;
string s;
while (iss >> s)
{
remove_if(s.begin(), s.end(),
[](unsigned char ch){ return !isalpha(ch); });
if (!s.empty())
{
transform(s.begin(), s.end(), s.begin(),
[](unsigned char ch){ return tolower(ch); });
vec.push_back(s);
}
}
Word *sentence = new Word[vec.size()];
transform(vec.begin(), vec.end(), sentence,
[](const string &s){
Word w;
w.english = s;
return w;
}
);
size = vec.size();
return sentence;
}
Live Demo