leonardmh / namealizer Goto Github PK

The word formatting features of this tool are fairly robust and could be more widely useful if you could just call the formatter directly from the command line.

All of that functionality is already broken out into a separate function anyways so it shouldn't be too hard.

Please.

Currently the user is required to remember format styles which are defined by me. This is less than ideal. I think the best way to do this is to allow the user to define their own arbitrary separator and then specify a standard word format which would follow the currently existing formatting options.

This change would affect several things:

1. I would have to update the command line options to allow the user to specify wordstyle and separator.
2. These changes will require updates to the tests.
3. This will require a re-write of the README.

Overall this is definitely worth it and needs to be done before any further improvements are bolted on top.

This should be as easy to use as possible which means download, install, and use. Even more ideally the install step should be optional. This currently clones with the correct permissions to use right out of the box but the hashbang isn't set correctly. I think it was developed for use on a Linux machine originally.

Put together a simple GUI using PyQt.

I received the following error when just using the tool:

Traceback (most recent call last):
  File "namealizer", line 212, in <module>
    main()
  File "namealizer", line 204, in main
    string_to_print = "{} {}".format(get_random_word(dictionary), get_random_word(dictionary))
  File "namealizer", line 113, in get_random_word
    word_index = random.randint(0, len(dictionary[starting_letter]) - 1)
IndexError: list index out of range

This should be improved so that any failures show the help message.

This tool would be infinitely more useful if it could be installed directly from pip. With that said, it's not really all that useful until it has a stable API (done?) and useful functions to be called from import (done).

Since the dictionaries are currently just implemented as pure plaintext they could likely see some serious storage benefits from being stored and transmitted as compressed files.

Python offers pretty good support out of the box for interacting with compressed files so some benchmarking may be in order to see if there are any real benefits to this. If so then this will be integrated into the next release.

I think the two most likely use cases for this script are:

1. As a module that others import into their own code. In which case I'll need to work on PIP support.
2. As a web interface that people can use or tie into as an API. This will likely be a little bit of work and is worth looking at as a v1.0+ feature.

When working on the duplicate remover I noticed that a lot of extra single-letter-words were being generated. After diving into the issue a little bit with a smaller dictionary I realized that this was due to a bug in the dictionary importer.

Given a list of words like:

aardvark
antietam
antonym

These would be imported in as the following list:

['a', 'a', 'r', 'd', 'v', 'a', 'r', 'k', 'antietam', 'antonym']

The default dictionary all_en_US.dict contains several entries which aren't real words. These should be removed.

If I create two WordGenerator objects with the same seed and then call them with the same access method they will return the same results.

However, if I create two WordGenerator objects with the default seeds and then change their seeds I don't see any apparent effect.

I don't have any coverage for the command line arguments. I'm also not really thoroughly testing the sparse dictionaries.