Every wordmaxxing

Following the trend of accounts like F.ck every word or every word is trans, this is a repository containing the code for Every wordmaxxing, which gives it a Gen Alpha spin.

While the project's theme is unserious, the code provided should be considered an easy to follow example of how to achieve the following goals:

• set up an Azure Function with a time trigger, in Node.js
• access data in the Azure Cosmos DB and update it
• read a plain text shipped together with the function code, in Node.js
• use axios to access 3rd party APIs (in this case, Profanity filter by Ninja API and Bluesky Official API are used)

How to run

First, set up a Microsoft Azure Functions Node.js v4 project.

Rename local.settings.json.sample to local.settings.json and fill the appropriate values, such as your Bluesky credentials or Azure Cosmos DB connection strings.

In your Cosmos DB Container, include only a single line with the following data: {id: "line_id", pkey: "pkey", data: 0}. pkey is used as a partition key (not important in this example), and data will point to the index of the line in words.txt which should be processed next.

Finally, place a words.txt file in the project's root directory. This file should contain one word per line.

Now you can run your function either locally, or you can upload it to Azure and have it run at a specified time. In this example, the schedule argument value "0 17 * * * *", means that it will run on the 17th minute of each hour of each day.

Content source

In the implementation Every wordmaxxing, words were sourced from The Natural Language Toolkit (NLTK) and filtered down to nouns using the nltk package.

from nltk.corpus import wordnet as wn

def write_all_nouns():
    # get all words and filter them down to nouns ('n')
    all_nouns = [x.name() for x in wn.all_synsets(pos='n')]
    # sort alphabetically
    all_nouns.sort()
    with open('words.txt', 'w') as f:
        for noun in all_nouns:
            # transform word variant (dog.n.01) to the word itself
            parts = noun.split('.')
            # skip duplicates (such as dog.n.02)
            if parts[2] != '01':
                continue
            # nltk uses '_' instead of ' '
            f.write(parts[0].replace('_', ' ') + '\n')

This database contains some very bad words (including racial slurs). This is why Profanity filter by Ninja API is implemented to check every word and skip the rude ones.