Summer Code Jam 2020: Qualifier

To qualify for the upcoming Summer Code Jam, you'll have to complete a qualifier assignment. For the assignment, you'll have to write an Article class that could be used to represent an article published on a website.

Please read the instructions carefully and submit your solution before the deadline using the sign-up form. Also, note that we've included a test suite you can use to test your solution before you submit it.

• The deadline for signing up has passed! It's no longer possible to sign up for the Summer Code Jam.

Table of Contents

• Qualifying for the Code Jam
• Rules and Guidelines
• Qualifier Assignment
  • Basic Requirements
  • Intermediate Requirements
  • Advanced Requirements
• Test Suite
  • Running the Test Suite

Qualifying for the Code Jam

The qualifier assignment has three sections of increasing difficulty:

1. Basic Requirements
2. Intermediate Requirements
3. Advanced Requirements

To qualify for the Code Jam, your solution has to pass the basic requirements. However, we urge you to at least try the intermediate and/or advanced requirements if you think you can tackle them. We will publish our solution for the qualifier after the deadline has passed.

Rules and Guidelines

• Your submission will be tested using a Python 3.8.3 interpreter without any additional packages installed. You're allowed to use everything included in Python's standard library, but nothing else. Please make sure to include the relevant import statements in your submission.

• Use qualifier.py as the base for your solution. It includes stubs for the two classes you need to write: Article and ArticleField.

• Do not change the names of the two classes included in qualifier.py. The test suite we will use to judge your submission relies on these two classes. Everything else, including the docstrings of the classes, may be changed.

• You can leave the ArticleField class as-is if you do not wish to tackle the advanced requirements.

• Do not include "debug" code in your submission. You should remove all debug prints and other debug statements before you submit your solution.

• This qualifier task is supposed to be an individual challenge, so try to solve it on your own. You should not discuss (parts of) your solution in public (including our server). You are still allowed to do research and ask questions about Python as they relate to your qualifier solution, but try to use general examples if you post code along with your questions.

Qualifier Assignment

For this assignment, you'll write an Article class to represent articles published to a blog. Instances of this class will have attributes like title and author. You will also write a few methods to work with these classes. We've added a "stub" for your Article class in the qualifier.py file for you fill in.

Please read the requirements below carefully; it's important for your code to meet those requirements exactly. Also, note that there's a test suite available for you to test your code before you submit it.

For all the examples below, assume we've created an instance of Article like this:

>>> fairytale = Article(
...     title="The emperor's new clothes",
...     author="Hans Christian Andersen",
...     content="'But he has nothing at all on!' at last cried out all the people. The Emperor was vexed, for he knew that the people were right.",
...     publication_date=datetime.datetime(1837, 4, 7, 12, 15, 0),
... )

Basic Requirements

The requirements listed in this section only apply to the Article class.

1. Write an __init__ method that stores the arguments for the four parameters (title, author, content, and publication_date) as attributes. The attributes should be publicly available using the same names as the parameters.

   >>> fairytale.title
   "The emperor's new clothes"
   >>> fairytale.publication_date
   datetime.datetime(1837, 4, 7, 12, 15, 0)

2. To make debugging easier, implement a __repr__ method that returns a string representation of the class. It should exactly match the following format:

   >>> print(repr(fairytale))
   <Article title="The emperor's new clothes" author='Hans Christian Andersen' publication_date='1837-04-07T12:15:00'>

   • The value for publication_date is formatted using datetime.datetime.isoformat().
   • Make sure to use the repr of the values for title, author, and publication_date.isoformat().

3. As it's nice to know how long an article is, implement support for the built-in function len. It should return the length of content.

   >>> fairytale.content
   "'But he has nothing at all on!' at last cried out all the people. The Emperor was vexed, for he knew that the people were right."
   >>> len(fairytale)
   128

4. Blogs often feature a short section of an article on their front page. Write a method called short_introduction that has an int parameter named n_characters. The method should return a short introduction that contains at most n_characters from the start of the article's content. To avoid awkwardly cutting off text in the middle of a word, find and "cut" the text on the last space or newline character within the first n_characters + 1. You may assume there's always at least one space or newline character within the first n_characters + 1.

   >>> fairytale.short_introduction(n_characters=60)
   "'But he has nothing at all on!' at last cried out all the"

   • The value returned by Article.short_introduction should not include the space or newline character you used to break up the text.

5. It's often interesting to have some statistics to show on your blog. Write a method called most_common_words that has an int parameter named n_words. The method should return a dictionary of the n_words most common words in the content of the article. If words have the same frequency, order them in the same order in which they first appeared in the content. The method should also be case-insensitive (for example, "The" and "the" count as the same word).

   >>> fairytale.most_common_words(5)
   {'the': 3, 'he': 2, 'at': 2, 'all': 2, 'people': 2}
   >>> fairytale.most_common_words(3)
   {'the': 3, 'he': 2, 'at': 2}

   • Output the words in lowercase in the dictionary.
   • Every non-alphabet (ASCII only) character counts as a space or word break. For example, "It's" counts as two "words": "it" and "s".

Intermediate Requirements

The requirements listed in this section only apply to the Article class. Please make sure the changes you make for the requirements this section don't break any of the requirements listed in the previous section.

1. A common way to uniquely identify an article is to give it a unique number. Add a feature to the class that gives each new Article a unique id number. The numbers should be sequential and, in good Python tradition, the first article should get an id of 0.

   >>> article_one = Article(title="PEP-8", author="Guide van Rossum", content="Use snake_case", publication_date=datetime.datetime(2001, 7, 5))
   >>> article_one.id
   0
   >>> article_two = Article(title="Fluent Python", author="Luciano Ramalho", content="Effective Programming", publication_date=datetime.datetime(2015, 8, 20))
   >>> article_two.id
   1

   • You should not define anything outside of the class definition to accomplish this; do not use "global" variables.

2. Making mistakes is human and so is trying to fix them. Add a feature to keep track of when the most recent change was made to the article's content. Create a new attribute named last_edited and set its initial value to None. When a change is made to the content, obtain the current date and time with datetime.datetime.now() and assign it to last_edited. To avoid making breaking changes to the class's "API", content should still be accessed and changed with normal attribute access.

   >>> fairytale.last_edited
   None
   >>> fairytale.content = "I'm making a change to the content of this article"
   >>> fairytale.last_edited
   datetime.datetime(2020, 5, 26, 19, 41, 10)  # My local time at the time of writing

3. A common operation on a collection of articles is to sort them by their publication date. Add support for sorting Article objects directly without having to use a key function for sorted or list.sort. The sorting order should be based solely on the publication_date attribute. Sorting in ascending order (the default) should result in the oldest article being first.

   >>> articles = [
   ...     Article(..., publication_date=datetime.datetime(2001, 7, 5)),
   ...     Article(..., publication_date=datetime.datetime(1837, 4, 7)),
   ...     Article(..., publication_date=datetime.datetime(2015, 8, 20)),
   ... ]
   >>> sorted(articles)
   [<Article ... publication_date="1837-04-07T...">, <Article ... publication_date="2001-07-05T...">, <Article ... publication_date="2015-8-20T...">]

Advanced Requirements

The requirements in this section will ask you to implement the ArticleField class. It is not necessary to make changes to the Article class, but if you do, make sure the tests for the requirements in previous sections still pass.

While duck typing is a common practice in Python, type checking can come in handy. For example, articles will probably have to be saved to a database with a rigid data type scheme. The types of the article's attributes can be checked upon assignment to catch type errors early rather than late. In this section, you'll implement a simple descriptor that checks if the value assigned to an attribute has the correct type.

1. Implement a descriptor, ArticleField, that checks if the value assigned to an instance attribute has the correct type. To make the descriptor reusable for different types, its __init__ method should have a field_type parameter, which is the type it should check against. To not be too rigid, the type check should allow instances of subclasses of the type in addition to the type itself.

   If the value has the correct type, the assignment should happen normally; if not, the descriptor should raise a TypeError:

   >>> class Article:
   ...     attribute = ArticleField(field_type=int)
   >>> article = Article(...)
   >>> article.attribute = 10
   >>> article.attribute
   10
   >>> article.attribute = "some string"
   Traceback (most recent call last):
       ...
   TypeError: some message here
   >>> article.attribute
   10

2. Whenever you raise an exception, it's important to give the developer enough information to debug the error. Modify your descriptor's exception message to include the name of the attribute being assigned, the name of type that was expected, and the name of the type that was received instead. You should not change the function signature of the __init__ method to do this; do not pass the name of the attribute to __init__. The message should exactly match the following format:

   >>> class Article:
   ...     age = ArticleField(field_type=int)
   >>> article = Article(...)
   >>> article.age = "some string"
   Traceback (most recent call last):
       ...
   TypeError: expected an instance of type 'int' for attribute 'age', got 'str' instead

Test Suite

We've written a basic test suite that tests if your code passes the requirements above. We strongly suggest you run these tests before you submit your solution. In principle, if your solutions passes the tests for the basic requirements, you should qualify for the Code Jam.

It's perfectly fine to have a look at the tests (see test_qualifier.py). This is the same test suite we will use to judge your solution. However, note that we will run the tests with different data to ensure that solutions were not written to only work with the exact data provided.

Running the Test Suite

To run the test suite, first download the files test_qualifier.py and run_tests.py. Place these files in the same directory as the file containing your solution and make sure the file containing your solution is called qualifier.py. Then, open a terminal/command window and change the current directory to your solution's directory. Finally, run the following command:

python run_tests.py

Note: You may have to replace python with the command you use to run Python from the command line. If you're using Windows and python doesn't work, try py instead. If you're using Linux, you may have to use python3 instead.

The test suite requires at least Python 3.7. It has also been confirmed to work with 3.8.