https://github.com/heckj/warbler

Warbler language server implementation to identify the use of passive voice within Markdown documents
https://github.com/heckj/warbler

lsp markdown nlp python3

Last synced: 15 days ago
JSON representation

Warbler language server implementation to identify the use of passive voice within Markdown documents

• Host: GitHub
• URL: https://github.com/heckj/warbler
• Owner: heckj
• License: apache-2.0
• Created: 2020-09-23T19:25:02.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2021-03-16T23:56:18.000Z (over 3 years ago)
• Last Synced: 2024-10-10T19:41:16.141Z (about 1 month ago)
• Topics: lsp, markdown, nlp, python3
• Language: Python
• Homepage:
• Size: 235 KB
• Stars: 3
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Warbler

This is a [language server] implementation for markdown, specifically focusing

on using the LSP to provide writer feedback for content, rather than concrete validation of the markdown structure to any specification.

[language server]: https://langserver.org/

This is an experimental implementation, intended to leverage modern NLP systems to identify and highlight the use of passive in written content.

Code for the LSP python implementation is derived from [Benten](https://github.com/rabix/benten) - another LSP that's implemented in python.

The name was chosen as a [common small songbird](http://www.pacificnorthwestbirds.com/category/warbler-summer/) from the pacific northwest. Something to look over your shoulder and call out...

## Dev Notes

This setup uses tox for most of it's build and such, so you'll need it as a prerequisite.

If you don't already have it installed:

    pip3 install tox

or

    python3 -m pip install --user tox

Run tests:

    tox

Development work:

- lint the code

    tox -vv -e lint

- use your the editor happily

    tox -vv -e dev

    code .

- run stand-alone on the CLI

    python warbler

- local virtualenv outside of .tox

    python3 -m venv .venv

    python3 -m pip install --upgrade pip

    source .venv/bin/activate

    pip3 install .

## dev notes

General thesis of this setup:

[client] ---> [LanguageServer]

Processing the docs:

1. convert the incoming markdown file into plain text (w/ markdown & beautifulsoup)

2. run the resulting text through textacy, filtering for passive voice tagging

3. link the tags back to locations within the markdown to reference where the issue is happening.