Skip to content

Latest commit

 

History

History
35 lines (24 loc) · 1.88 KB

README.md

File metadata and controls

35 lines (24 loc) · 1.88 KB

LanguageTool with Tailscale Sidecar Configuration

This Docker Compose configuration sets up LanguageTool with Tailscale as a sidecar container to enhance secure networking.

LanguageTool

LanguageTool is a powerful grammar and spell-checking tool available for various languages. It can be used in various applications, including web browsers, office suites, and as a standalone server for integration with other services.

Configuration Overview

In this setup, the tailscale-adguardhome service runs Tailscale, which manages secure networking for LanguageTool. The languagetool service utilizes the Tailscale network stack via Docker's network_mode: service:. This setup ensures that LanguageTool's service is only accessible through the Tailscale network (or locally, if preferred), providing an extra layer of security and privacy for your LanguageTool deployment.

Using n-gram datasets

LanguageTool can make use of large n-gram data sets to detect errors with words that are often confused, like their and there.

Source: https://dev.languagetool.org/finding-errors-using-n-gram-data

Download the n-gram dataset(s) onto your local machine and unzip them into a local ngrams directory:

home/
├─ /
│  ├─ ngrams/
│  │  ├─ en/
│  │  │  ├─ 1grams/
│  │  │  ├─ 2grams/
│  │  │  ├─ 3grams/
│  │  ├─ nl/
│  │  │  ├─ 1grams/
│  │  │  ├─ 2grams/
│  │  │  ├─ 3grams/

Mount the local ngrams directory to the /ngrams directory in the Docker container using the -v configuration and set the languageModel configuration to the /ngrams folder.