english-words

english-words-py

Returns sets of English words created by combining different words lists together. Example usage: to get a set of English words from the "web2" word list, including only lower-case letters, you write the following:

>>> from english_words import get_english_words_set
>>> web2lowerset = get_english_words_set(['web2'], lower=True)

Usage

From the main package, import get_english_words_set as demonstrated above. This function takes a number of arguments; the first is a list of word list identifiers for the word lists to combine and the rest are flags. These arguments are described here (in the following order):

• sources is an iterable containing strings corresponding to word list identifiers (see "Word lists" subsection below)
• alpha (default False) is a flag specifying that all non-alphanumeric characters (e.g.: -, ') should be stripped
• lower (default False ) is a flag specifying that all upper-case letters should be converted to lower-case

Each word list is pre-processed to handle the above flags, so using any combination of options will not cause the function to run slower.

Note that some care needs to be used when combining word lists. For example, only proper nouns in the web2 word list are capitalized, but most words in the gcide word list are capitalized.

Word lists

Name/URL	Identifier	Notes
GCIDE 0.53 index	gcide	Words found in GNU Collaborative International Dictionary of English 0.53. Most words capitalized (not exactly sure what the capitalization convention is). Contains some entries with multiple words (currently you must use the alpha option to exclude these). Unicode characters are currently unprocessed; for example <ae/ is present in the dictionary instead of æ. Ideally, these should all be converted.
web2 revision 326913	web2

Adding additional word lists

To add a word list, say with identifier x, put the word list (one word per line), into a plain text file x.txt in the raw_data directory at the root of the repository. Then, to process the word list (and all others in the directory) run the script process_raw_data.py.

Installation

Install this with pip with

pip install english-words

This package is unfortunately rather large (~20MB), and will run into scaling issues if more word lists or (especially) options are added. When that bridge is crossed, word lists should possibly be chosen by the user instead of simply including all of them; word lists could also be preprocessed on the client side instead of being included in the package.