Python history quiz program

Question

This program allows users to manage questions and quiz users from years of historical events. I'm looking for suggestions on how can I better organize my code, maybe apply OOP for functions and avoid using global variable data_filename.

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

from enum import Enum
import os
import pickle
import random


data_filename = 'program.obj'


def read_int(prompt='> ', errmsg='Invalid number!'):
    number = None
    while number is None:
        try:
            number = int(input(prompt))
        except ValueError:
            print(errmsg)
    return number


def display_menu():
    print('What do you want to do?')
    print('[1] List all historical events')
    print('[2] Add event')
    print('[3] Remove event')
    print('[4] Quiz')
    print('[5] Statistics')
    print('[6] Clear statistics') # TODO: move to options submenu
    print('[7] Exit')


def clear_screen():
    print(chr(27) + "[2J")


def pause():
    input('Press any key to continue...')


def yes_or_no(prompt='Proceed? [y|n]\n> ', errmsg='Valid answers are y and n.'):
    answer = input(prompt).strip().lower()
    while answer != 'y' and answer != 'n':
        print(errmsg)
        answer = input(prompt).strip().lower()
    return answer


def read_data_file():
    file = open(data_filename, 'a+b')
    file.seek(0)
    data = {'events': [], 'statistics': {'total_successes': 0, 'total_failures': 0}}
    if os.path.getsize(data_filename) > 0:
        data = pickle.load(file)
    file.close()
    return data


def list_events(data):
    events = data['events']
    if len(events) == 0:
        print('Event list is empty!')
    else:
        print('Historical events:')
        for event in events:
            print(event)


def add_event(data):
    events = data['events']
    year = read_int(prompt='Enter year: ')
    description = input('Enter description: ')
    events.append({'year': year, 'description': description})
    print('Successfully added a new historical event!')


def remove_event(data):
    events = data['events']
    if len(events) == 0:
        print('Event list is empty!')
    else:
        for event in events:
            print(events.index(event), event)
        index = read_int('Which event do you want to delete? ')
        try:
            events.pop(index)
            print('Successfully deleted event!')
        except IndexError:
            print('Number out of range!')


def quiz(data):
    events = data['events']
    stats = data['statistics']
    if len(events) == 0:
        print('Event list is empty!')
    else:
        num = read_int(prompt='How many questions should I ask? ')
        if 0 < num <= len(events):
            for event in random.sample(events, num):
                print(event['description'])
                year = read_int('In which year was following event occurred? ')
                if year == event['year']:
                    stats['total_successes'] += 1
                    print('Good answer!')
                else:
                    stats['total_failures'] += 1
                    print('Bad answer!')
        elif num < 0:
            print('Number of questions can\'t be negative!')
        else:
            print('Too much questions!')


def display_stats(data):
    stats = data['statistics']
    tries = stats['total_successes'] + stats['total_failures']
    if tries == 0:
        total_successes = 0
        total_failures = 0
    else:
        total_successes = stats['total_successes']/tries * 100
        total_failures = stats['total_failures']/tries * 100
    print('Statistics')
    print('Total: {0:10.2f}% successes, {1:10.2f}% failures'.format(total_successes, total_failures))


def clear_stats(data):
    answer = yes_or_no('Are you sure you want to clear statistics? [y|n]\n> ')
    if answer == 'y':
        data['statistics'] = {'total_successes': 0, 'total_failures': 0}
        print('Successfully cleared statistics!')
    else:
        print('Statistics left unchanged.')


# TODO: settings submenu
# def settings(data):
#     print('-' * 10)
#     print('Program settings')
#     print('-' * 10)
#     print('[1] Clear statistics')
#     print('[2] Back')
#     user_choice = read_int()
#     while user_choice != 2:
#         if user_choice == 1:
#             clear_stats(data)


def update_data_file(data):
    file = open(data_filename, 'wb')
    pickle.dump(data, file)
    file.close()


class Choices(Enum):
    list_events = 1
    add_event = 2
    remove_event = 3
    quiz = 4
    statistics = 5
    clear_stats = 6
    exit = 7


program_data = read_data_file()
choice = None
while choice != Choices.exit.value:
    clear_screen()
    display_menu()
    choice = read_int()
    if choice == Choices.list_events.value:
        list_events(program_data)
    elif choice == Choices.add_event.value:
        add_event(program_data)
    elif choice == Choices.remove_event.value:
        remove_event(program_data)
    elif choice == Choices.quiz.value:
        quiz(program_data)
    elif choice == Choices.statistics.value:
        display_stats(program_data)
    elif choice == Choices.clear_stats.value:
        clear_stats(program_data)
    elif choice == Choices.exit.value:
        print('Good bye!')
    else:
        print('Invalid choice!')
    update_data_file(program_data)
    pause()

Answer 1

Constants

data_filename should be capitalized, since it's a global constant.

Early-return

There's no need to use number as the condition on your while; instead:

number = None
while number is None:
    try:
        number = int(input(prompt))
    except ValueError:
        print(errmsg)
return number

can be

while True:
    try:
        return int(input(prompt))
    except ValueError:
        print(errmsg)

Menu management

display_menu could use a sequence of tuples, better yet a sequence of named tuples or @dataclasses, each having a title string attribute and a callable attribute. Then your display_menu could be

print('What do you want to do?')
print('\n'.join(f'[{i}] {item.title}' for i, item in enumerate(menu, 1)))

I see that you also have a Choices enum. That's not bad, and you could potentially use both an enum and the suggestion above, so long as you have a dictionary of enum-choices-to-function-references.

Membership tests

answer != 'y' and answer != 'n'

can be

answer not in {'y', 'n'}

Loop structure

Avoid having to call input twice; this:

answer = input(prompt).strip().lower()
while answer != 'y' and answer != 'n':
    print(errmsg)
    answer = input(prompt).strip().lower()
return answer

can be

while True:
    answer = input(prompt).strip().lower()
    if answer in {'y', 'n'}:
        return answer
    print(errmsg)

File operations

file = open(data_filename, 'a+b')
file.seek(0)

has a few issues:

• Newly-opened file handles do not need an initial seek-to-beginning; that's redundant if you open in read-binary mode
• You should be using with and avoiding an explicit close()
• You should only bother to initialize your default data when needed
• You want to check whether the file exists, not its size

So:

if os.path.exists(data_filename):
    with open(data_filename, 'rb') as file:
        return pickle.load(file)
return {'events': [], 'statistics': {'total_successes': 0, 'total_failures': 0}}

Redundant predicates

if 0 < num <= len(events):
    ...
elif num < 0:
    print('Number of questions can\'t be negative!')
else:
    print('Too much questions!')

needs to be re-thought. First, too much questions should be too many questions. Also, what if the user enters 0? Surely that's not "too many questions", but that's what will be printed. Suggested:

if num < 1:
    print('Not enough questions.')
elif num > len(events):
    print('Too many questions.')
else:
    ...

Stronger-typed, stronger-structured data

stats['total_successes'] + stats['total_failures']

is an example of what I've seen called "data pasta". Dictionaries are being abused, here, when something like a @dataclass containing type hints is more appropriate.

String interpolation

'Total: {0:10.2f}% successes, {1:10.2f}% failures'.format(total_successes, total_failures)

is more easily expressed as

(
    f'Total: {total_successes:10.2f}% successes, '
    f'{total_failures:10.2f}% failures'
)

The line split is optional but better for legibility.

String escapes

'Number of questions can\'t be negative!'

is more easily written as

"Number of questions can't be negative!"