manx-corpus-search

A corpus search for primarily bilingual manx to english texts.

Deployed at https://corpus.gaelg.im/

To add/modify documents, see: manx-search-data

Installation

1. Clone the source
2. Copy the OpenData folder from manx-search-data into CorpusSearch/OpenData folder
3. dotnet run

Tech Stack

• React
• C# (ASP.NET Core, both WebAPI and content server)
• Document Searching: Apache Lucene.NET
• Query Search Syntax: csly
• CSV: CsvHelper
• JSON: Newtonsoft.Json

Aims

• Run in RAM on a cheap (<$20/m) droplet
• No expectation of scaling up for a large number of users
• Expected corpus size is unlikely to exceed 10MM words of Manx (and 10MM words of English)
• Stateless

Deployment

Deployable on a $5 DigitalOcean droplet. See GitHub actions

Analytics

• Uses https://app.segment.com/ anonymously - tracking the count of searches

Server requirements

• git
• dotnet-sdk-6.0
• TODO