A command line tool for categorizing products by the flavor (flavour) of English used.

• Python 3
• Virtualenv

make
source venv/bin/activate
english parse_products --file <path_to_file_with_product_ids>

or

make
./venv/bin/english parse_products --file <path_to_file_with_product_ids>

make develop
make test

The entry point of this tool is the 'parse_products' function in the main.py file. The parse_products function takes a command line argument file path pointing to a text file containing product_ids to be checked. The function first grabs updated word list definitions from a github gist. The product_ids are then read in and used to make a request to a graphql endpoint to retrieve more information about the products. Each product's name and description are parsed and checked against the word lists to see how many words can be considered "American English" vs "British English". Based on the numbers of words for each, the product is categorized as either American English, British English, Mixed British and American English, or Unknown. The product_ids are printed to stdout with the determined categorization.

The text parsing is done with a case-insensitive regex. It is unsophisticated and by no means robust.

A simple suite of tests can be found in the test.py file.

The input file of product_ids is well formed with a single product_id per line.

The word lists retrieved from the github gist are also well formed with a single word per line.

The products endpoint is assumed to return json data in the form of a dictionary (hash) with a top level key called 'data', with a sub-key called 'products' that is a list of products containing at minimum an id, name, and description.

A product is considered to be American English if at least one word was found to be from list of American English words and no words were found from the list of British English words.

A product is considered to be British English if at least one word was found to be from list of British English words and no words were found from the list of American English words.

A product is considered to be Mixed British and American English if at least one word was found to be from list of American English words and at least one word was found from the list of British English words.

A product is considered to be Unknown if no words were found from either the American or British list of words.