language_tool_python

language_tool_python: A Grammar Checker for Python 📝

The language_tool_python library is a powerful Python wrapper designed to integrate with LanguageTool, an open-source grammar correction tool often referred to as the OpenOffice spellchecker. By leveraging this library, developers can detect grammar errors and spelling mistakes within their Python scripts or through a command-line interface.

Local and Remote Server Options

Using language_tool_python Locally

By default, language_tool_python downloads a LanguageTool server .jar file to run on your local machine, allowing you to detect grammar errors without internet restrictions. To get started, initialize a LanguageTool object in your Python script with the desired language:

import language_tool_python
tool = language_tool_python.LanguageTool('en-US')

Using language_tool_python with the Public API

Alternatively, you can utilize LanguageTool's public HTTP Proofreading API, avoiding the download process:

import language_tool_python
tool = language_tool_python.LanguageToolPublicAPI('es')

Using With Another Remote Server

You can also specify your own remote server if you have one set up:

import language_tool_python
tool = language_tool_python.LanguageTool('ca-ES', remote_server='https://language-tool-api.mywebsite.net')

Customizing Matches with utils.correct

This library allows you to fine-tune grammar corrections by putting decision power in your hands. You can generate a list of problematic matches with tool.check() and then apply specific corrections using language_tool_python.utils.correct(). For instance, if you want to ignore uppercase spelling suggestions, you can filter out such matches:

s = "Department of medicine Colombia University closed on August 1 Milinda Samuelli"
is_bad_rule = lambda rule: rule.message == 'Possible spelling mistake found.' and len(rule.replacements) and rule.replacements[0][0].isupper()

import language_tool_python
tool = language_tool_python.LanguageTool('en-US')
matches = tool.check(s)
matches = [rule for rule in matches if not is_bad_rule(rule)]
corrected_text = language_tool_python.utils.correct(s, matches)

Example Usage

Here's a typical use case from within a Python interpreter:

import language_tool_python
tool = language_tool_python.LanguageTool('en-US')
text = 'A sentence with a error in the Hitchhiker’s Guide tot he Galaxy'
matches = tool.check(text)
corrected_text = tool.correct(text)
tool.close()

From the command line, you can execute:

$ echo 'This are bad.' > example.txt
$ language_tool_python example.txt

This will provide suggestions directly in the console.

Client-Server Model

language_tool_python supports a client-server model, making it feasible to run LanguageTool on a server and connect from a client on a different machine:

Server Code:

import language_tool_python
tool = language_tool_python.LanguageTool('en-US', host='0.0.0.0')

Client Code:

import language_tool_python
lang_tool = language_tool_python.LanguageTool('en-US', remote_server='http://0.0.0.0:8081')

Configuration Options

To enhance flexibility, language_tool_python offers numerous configuration options. For example, enabling caching might improve performance:

import language_tool_python
tool = language_tool_python.LanguageTool('en-US', config={ 'cacheSize': 1000, 'pipelineCaching': True })

You can also set a text length limit:

import language_tool_python
tool = language_tool_python.LanguageTool('en-US', config={ 'maxTextLength': 100 })

A full range of configuration options is documented, allowing customization to suit specific requirements.

Installation and System Requirements

To install language_tool_python, simply use pip:

pip install --upgrade language_tool_python

Ensure you have Python 3.6 or higher, and LanguageTool requires Java 8.0 or newer. The library handles LanguageTool downloads for you, but manual installation is also supported for specific needs.

Conclusion

language_tool_python serves as an accessible, robust tool for implementing grammar checking within Python environments. Its adaptability to both local and remote servers, coupled with its array of configuration options, makes it ideal for developers seeking precise control over their language corrections. Whether through a simple script or within a larger server-based framework, this library effectively integrates LanguageTool’s powerful capabilities into Python workflows.