Full Programs: From One Word to a Text Report

In C you wrote a marks report with char name[20] for every student. A name with a space broke scanf("%s"), and a name of twenty letters broke the array. This lesson writes that report again with strings, and the name can be as long as the person's real name. Six programs get there, each a little bigger than the last. Read the "new thing" line above each program first; it is what that program is for.

Program 1: count the words and their characters

The new thing: while (cin >> word) reads one word per turn until the input ends, and size() measures each word as it arrives.

A word here is a run of characters with no space in it, which is exactly what >> reads into a string. Lesson 01 showed one read; this program loops. Zara tests the empty input first, so it gets its own line.

#include <iostream>
#include <string>
using namespace std;

int main() {
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    string word;
    int count = 0;
    int characters = 0;
    int long_words = 0;
    while (cin >> word) {
        count++;
        characters += word.size();
        if (word.size() >= 5) {
            long_words++;
        }
    }

    if (count == 0) {
        cout << "no words\n";
        return 0;
    }
    cout << "words: " << count << '\n';
    cout << "characters: " << characters << '\n';
    cout << "5 or more characters: " << long_words << '\n';
    return 0;
}

words: 9
characters: 35
5 or more characters: 3

That output is for the input the quick brown fox and jumps over the lazy dog on two lines. The newline between the lines is just more space to >>. Each word arrives in a string sized to fit it, and size() gives its length at once: quick, brown and jumps have five. So the loop reads any number of words of any length, with no size you chose.

Program 2: a letter count with vector<int>(26)

The new thing: vector<int> count(26, 0) with tolower(u) - 'a' as the index, where u is the character cast to unsigned char.

Amara wants to know if a sentence is a pangram, a sentence that uses all 26 letters. Each letter gets a box: 'a' is 97 and 'z' is 122, so c - 'a' runs from 0 to 25. isalpha and tolower come from <cctype>; the first asks "is this a letter?", the second turns a capital into a small letter.

#include <cctype>
#include <iostream>
#include <string>
#include <vector>
using namespace std;

int main() {
    string line;
    while (getline(cin, line)) {
        vector<int> count(26, 0);
        for (char c : line) {
            unsigned char u = c;
            if (isalpha(u)) {
                count[tolower(u) - 'a']++;
            }
        }

        string missing;
        int top = 0;
        for (int i = 0; i < 26; i++) {
            if (count[i] == 0) {
                missing += char('a' + i);
            }
            if (count[i] > count[top]) {
                top = i;
            }
        }

        cout << line << '\n';
        cout << "  most used: " << char('a' + top) << ", " << count[top] << " times\n";
        if (missing.empty()) {
            cout << "  a pangram: all 26 letters\n";
        } else {
            cout << "  missing " << missing.size() << ": " << missing << '\n';
        }
    }
    return 0;
}

Sphinx of black quartz, judge my vow
  most used: a, 2 times
  a pangram: all 26 letters
Kenji fixed one bug and found two more
  most used: e, 4 times
  missing 9: chlpqsvyz

Why the cast? On GCC for a PC, char is signed, so a byte above 127 becomes a negative number. Each byte of a Bangla letter is such a byte. The <cctype> functions are defined only for values an unsigned char can hold, so a negative one is undefined behaviour. Copying c into unsigned char u first makes every byte 0 to 255, and isalpha then says no to those bytes. So the vector is the table, and the cast keeps the table safe from text that is not English.

Program 3: split a line at its commas

The new thing: two ways to cut a line into fields: find with substr, and getline(in, field, ',') on an istringstream.

Each line of Amara's class file is a name and three marks with commas between them. The first way does the cutting by hand. line.find(',', start) gives the position of the next comma at or after start, or string::npos when there is none. A field runs from start to just before that comma.

#include <iostream>
#include <string>
using namespace std;

int main() {
    string line;
    getline(cin, line);

    size_t start = 0;
    int field = 1;
    while (true) {
        size_t comma = line.find(',', start);
        if (comma == string::npos) {
            cout << field << ": [" << line.substr(start) << "] to the end\n";
            break;
        }
        cout << field << ": [" << line.substr(start, comma - start) << "] comma at " << comma << '\n';
        start = comma + 1;
        field++;
    }
    return 0;
}

1: [Amara Okafor] comma at 12
2: [85] comma at 15
3: [90] comma at 18
4: [77] to the end

That output is for the input line Amara Okafor,85,90,77. The second argument of substr is a length, not an end, so it is comma - start. The brackets show that the space inside the name stayed, which >> would have lost.

The second way lets a stream do the cutting. Lesson 02 showed getline with a third argument: it reads up to that character and drops it. Put the line in an istringstream, and each call hands you the next field.

#include <iostream>
#include <sstream>
#include <string>
using namespace std;

int main() {
    string line;
    getline(cin, line);

    istringstream in(line);
    string field;
    int k = 1;
    while (getline(in, field, ',')) {
        cout << k << ": [" << field << "]\n";
        k++;
    }
    return 0;
}

1: [Amara Okafor]
2: [85]
3: [90]
4: [77]

The same four fields, from the same input, with no position to keep track of. Here is how the two compare.

So reach for getline when you only want the fields, and for find when you need positions or a longer separator.

Program 4: numbers out of each line

The new thing: a new istringstream for every line, read with >> until it runs dry, so an empty line simply reads no numbers.

Alice's phone logs her walks. Each line is one day, with the steps of each walk on it, and a day with no walk is an empty line. She wants each day's walks and steps, and the total of all days.

#include <iostream>
#include <sstream>
#include <string>
using namespace std;

int main() {
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    string line;
    int day = 0;
    long long all_steps = 0;
    while (getline(cin, line)) {
        day++;
        istringstream in(line);
        int steps;
        int walks = 0;
        long long total = 0;
        while (in >> steps) {
            walks++;
            total += steps;
        }

        cout << "day " << day << ": ";
        if (walks == 0) {
            cout << "no walks\n";
        } else {
            cout << walks << (walks == 1 ? " walk, " : " walks, ") << total << " steps\n";
        }
        all_steps += total;
    }
    cout << "all " << day << " days: " << all_steps << " steps\n";
    return 0;
}

day 1: 3 walks, 5400 steps
day 2: no walks
day 3: 1 walk, 5000 steps
day 4: 2 walks, 5000 steps
all 4 days: 15400 steps

That output is for four input lines: 1200 3400 800, an empty line, 5000 and 2500 2500. The outer loop reads lines, so it sees the empty one; cin >> alone would skip it and lose day 2. The stream is made inside the loop, so each day starts fresh. So the line keeps the days apart, and the string stream reads the numbers inside one day.

Program 5: a receipt built in an ostringstream

The new thing: the whole body of a report written into an ostringstream with setw and setprecision, then printed once with str().

David's shop prints a receipt. Its first line says how many lines it has, a number known only after the last item is read. So the program writes the item lines into a string first, prints the heading, then prints the string. setw(n) pads the next value to n characters; left and right choose the side; fixed with setprecision(2) gives two digits after the point. All of them come from <iomanip> and work on any output stream.

#include <iomanip>
#include <iostream>
#include <sstream>
#include <string>
using namespace std;

int main() {
    ostringstream body;
    body << fixed << setprecision(2);

    string item;
    int qty;
    double price;
    double total = 0;
    int lines = 0;
    while (cin >> item >> qty >> price) {
        double cost = qty * price;
        body << left << setw(10) << item << right << setw(4) << qty
             << " x " << setw(6) << price << " = " << setw(8) << cost << '\n';
        total += cost;
        lines++;
    }
    body << string(34, '-') << '\n';
    body << left << setw(26) << "TOTAL" << right << setw(8) << total << '\n';

    cout << "David's shop, " << lines << " lines\n";
    cout << body.str();
    return 0;
}

David's shop, 4 lines
pencil       3 x   0.50 =     1.50
notebook     2 x   2.25 =     4.50
eraser       1 x   0.80 =     0.80
ruler        1 x   1.20 =     1.20
----------------------------------
TOTAL                         8.00

That output is for four input lines: pencil 3 0.50, notebook 2 2.25, eraser 1 0.80 and ruler 1 1.20. Each item line is 10 + 4 + 3 + 6 + 3 + 8 = 34 characters, so the dashes are string(34, '-'). The settings stay on the stream, so fixed is written once. So an ostringstream is a draft you can finish before anyone sees it.

Program 6: the CSV marks toolIntermediate

The new thing: the pieces together in a real tool: getline for lines, getline(in, field, ',') for fields, stoi for marks, and a column as wide as the longest name.

A CSV file, for "comma-separated values", stores a table as text: one row per line, commas between the fields. Every spreadsheet program can save a sheet as CSV, so marks lists often travel that way. Each line here is name,mark,mark,mark. The tool prints a table with each student's average, then each test's class average. stoi, from lesson 02, turns a field like "85" into the int 85.

#include <algorithm>
#include <iomanip>
#include <iostream>
#include <sstream>
#include <string>
#include <utility>
#include <vector>
using namespace std;

int main() {
    vector<pair<string, vector<int>>> rows;
    size_t width = 4; // at least as wide as the word "name"
    string line;
    while (getline(cin, line)) {
        if (line.empty()) {
            continue;
        }
        istringstream in(line);
        string name, field;
        getline(in, name, ',');
        vector<int> marks;
        while (getline(in, field, ',')) {
            marks.push_back(stoi(field));
        }
        width = max(width, name.size());
        rows.push_back({name, marks});
    }

    ostringstream table;
    table << fixed << setprecision(1);
    table << left << setw(width) << "name" << right
          << setw(5) << "T1" << setw(5) << "T2" << setw(5) << "T3" << setw(9) << "average" << '\n';
    vector<int> test_sum(3, 0);
    for (const auto& [name, marks] : rows) {
        int total = 0;
        table << left << setw(width) << name << right;
        for (int t = 0; t < 3; t++) {
            table << setw(5) << marks[t];
            total += marks[t];
            test_sum[t] += marks[t];
        }
        table << setw(9) << total / 3.0 << '\n';
    }
    table << left << setw(width) << "class" << right;
    for (int t = 0; t < 3; t++) {
        table << setw(5) << (double)test_sum[t] / rows.size();
    }
    table << '\n';

    cout << rows.size() << " students\n" << table.str();
    return 0;
}

5 students
name                 T1   T2   T3  average
Amara Okafor         85   90   77     84.0
Bob Smith            48   55   61     54.7
Maria Lopez Garcia   90   94   88     90.7
Kenji Sato           81   79   85     81.7
Zara Ali             67   72   70     69.7
class              74.2 78.0 76.2

That output is for five input lines: Amara Okafor,85,90,77, Bob Smith,48,55,61, Maria Lopez Garcia,90,94,88, Kenji Sato,81,79,85 and Zara Ali,67,72,70.

The name column is 18 wide because the longest name, Maria Lopez Garcia, has 18 characters. Nobody chose that number; the data did. In C, char name[20] would have cut a longer name, and here a name of 200 characters only makes the column wider. total / 3.0 divides by a double, so 164 / 3 prints as 54.7, not 54. So the C report's two limits, the name length and the space inside a name, are both gone.

#include <iostream>
#include <sstream>
#include <string>
using namespace std;

string initials(const string& full_name) {
    istringstream in(full_name);
    string word, result;
    while (in >> word) {
        result += word[0];
        result += '.';
    }
    return result;
}

int main() {
    cout << initials("Maria Lopez Garcia") << '\n';
    cout << initials("  Amara   Okafor ") << '\n';
    cout << "[" << initials("") << "]\n";
    return 0;
}

M.L.G.
A.O.
[]

#include <iomanip>
#include <iostream>
#include <sstream>
#include <string>
using namespace std;

string ticket_code(char counter, int number) {
    ostringstream out;
    out << counter << '-' << setw(3) << setfill('0') << number;
    return out.str();
}

int main() {
    cout << ticket_code('A', 7) << '\n';
    cout << ticket_code('B', 42) << '\n';
    cout << ticket_code('C', 123) << '\n';
    string both = ticket_code('A', 7) + ", " + ticket_code('A', 8);
    cout << both << " (" << both.size() << " characters)\n";
    return 0;
}

A-007
B-042
C-123
A-007, A-008 (12 characters)

#include <cctype>
#include <iostream>
#include <string>
using namespace std;

string shift(string text, int k) {
    for (char& c : text) {
        unsigned char u = c;
        if (islower(u)) {
            c = 'a' + (c - 'a' + k) % 26;
        } else if (isupper(u)) {
            c = 'A' + (c - 'A' + k) % 26;
        }
    }
    return text;
}

int main() {
    string secret = shift("Meet Zara at noon", 3);
    cout << secret << '\n';
    cout << shift(secret, 23) << '\n';
    return 0;
}

Phhw Cdud dw qrrq
Meet Zara at noon

size_t comma = line.find(',', start);
cout << "[" << line.substr(start, comma) << "]\n";

istringstream in;
while (getline(cin, line)) {
    in.str(line);
    int x, total = 0;
    while (in >> x) {
        total += x;
    }
    cout << total << '\n';
}

cin >> n;
for (int i = 0; i < n; i++) {
    getline(cin, line);
    istringstream in(line);
    getline(in, name, ',');
    getline(in, field, ',');
    cout << name << ' ' << stoi(field) << '\n';
}

terminate called after throwing an instance of 'std::invalid_argument'
  what():  stoi

#include <iostream>
#include <sstream>
#include <string>
using namespace std;

int main() {
    istringstream in("a,,b,");
    string field;
    int count = 0;
    while (getline(in, field, ',')) {
        count++;
        cout << count << ": [" << field << "]\n";
    }
    cout << count << " fields\n";
    return 0;
}

#include <iostream>
#include <string>
#include <vector>
using namespace std;

int main()
{
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    vector<int> count(26, 0);
    string line;
    while (getline(cin, line)) {
        // For every letter of the line, turn a capital into a small
        // letter, then add 1 to count[letter - 'a'].
    }

    // For each letter from a to z that appears, print the letter,
    // a space and its count. Print "none" if no letter appears.

    return 0;
}

#include <iostream>
#include <sstream>
#include <string>
using namespace std;

int main()
{
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    int n = 0;
    cin >> n;
    string line;
    getline(cin, line); // finish the line that held n

    for (int i = 0; i < n; i++) {
        getline(cin, line);

        // Split the line at its commas: a name, then three marks.
        // Keep the name with the highest total; the first one wins a tie.
    }

    // Print the best name on one line and its total on the next.

    return 0;
}

#include <iostream>
#include <sstream>
#include <string>
#include <vector>
using namespace std;

int main()
{
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    int n = 0;
    cin >> n;
    string line;
    getline(cin, line); // finish the line that held n

    for (int i = 0; i < n; i++) {
        getline(cin, line);

        // Read the words of the line, then print them from the last
        // to the first, separated by single spaces.
    }

    return 0;
}

#include <iostream>
#include <string>
#include <vector>
using namespace std;

int main() {
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    string line;
    while (getline(cin, line)) {
        vector<string> fields;
        // Walk the line one character at a time, with a flag that says
        // whether you are inside quotes. A comma inside quotes belongs to
        // the field; a comma outside quotes ends it. Drop the quotes.

        // Print the fields of the line separated by " | ".
    }
    return 0;
}

Next, lesson 04 asks when a string is the wrong choice, and draws the chart and the flowchart that pick a char, a string_view or a vector<char> instead.