keeferrourke / pyapa

A Python module for checking APA style. Not a papaya.

pyapa

A Python module for checking APA style in writing.

This module provides an abstraction class, mostly composed of regular expressions. This project does not aim to cover the entire APA style guide, but it does attempt to cover several common errors with style.

To make errors easier to visually locate, 5 characters of context are sometimes given, depending on the detected error.

Installation

Clone this repository and run:

make install

or install from Pypi:

pip3 install pyapa

Information

Pypi: https://pypi.python.org/pypi/pyapa/

Github: https://github.com/keeferrourke/pyapa

Contact: https://krourke.org/contact

License: ISC License

pyapa.ApaCheck object class

This class documents regular expressions to match style errors.

The pyapa.ApaCheck.match() method generates an array of ApaMatch objects and stores them in the ApaCheck.Matches list for easy access.

pyapa.ApaMatch object class

This class contains the following members:

• ApaMatch.feedback a unicode string intended to hold a brief explanation of an associated error
• ApaMatch.see a URL string intended to hold an external reference to an APA Style guide
• ApaMatch.end the position of the character in a text which marks the end of the target character span
• ApaMatch.start the position of the character in a text which marks the beginning of the target character span
• ApaMatch.target the matching string that contains an error
• ApaMatch.suggestions a list of suggested replacement strings; at this time, the list is only ever one element long

There are also two helper functions to make printing these objects easy:

• ApaMatch.print() prints the object to stdout
• ApaMatch.sprint() builds a string containing information about the object

Example usage

From the interpreter:

>>> from pyapa import pyapa
>>> a = pyapa.ApaCheck()
>>> text = u'Papaya are delicious fruit, it was concluded (Author, et al. 2017).'
>>> a.match(text)
[<pyapa.pyapa.ApaMatch object at 0x000000000000>]
>>> len(a.Matches)
1
>>> a.Matches[0].print()
Match from 46 to 65 for:
Target: Author, et al. 2017
Feedback: Do not put a comma before 'et al.'
Suggestion: Author et al. 2017

Calling the module itself:

python3 -m pyapa -i input.txt
python3 -m pyapa -h
Usage:
pyapa [-h] [--version] [-o OUTPUT_FILE] -i INPUT_FILE

pyapa's Issues

Add citations for each error/pattern/rule

Using the new ApaMatch.see attribute, add url strings to external resources that validate each pattern.

Add more regexes

Current regexes are inspired from the common errors cited by @JonathanAquino.

ApaStyle.org has a rather extensive FAQ, from which other patterns can likely be derived. The class structure in pyapa.py is very simple, so it should be trivial for new contributors add on to this.

Email regex allows illegal domains

The current email regex allows for the following emails, none of which are correct.

[email protected]
[email protected]
[email protected]
[email protected]

I hardly see a point in fixing them, as this is not an email-verifying service - what are your thoughts?

Reference-dot match captures too much

When run on input

Hans & Yorke (2006) contradict existing literature (Hans and Yorke, 2006).

the program outputs

Match from 9 to 22 for:
Target: Yorke (2006)
Feedback: References go at the end of sentences; end your sentence after the reference.
Suggestion: Yorke (2006).

which is not correct (from my understanding, and as referenced here.