README.md

Using datasette to browse a CLDF dataset

Basics

The cldf command, which can be installed with the Python package pycldf provides functionality to load any CLDF dataset into a SQLite database. datasette is a powerful tool for exploring data in SQLite database.

Thus, exploring CLDF with datasette is as simple as running

pip install pycldf
pip install datasette
cldf createdb PATH/TO/CLDF/metadata.json PATH/TO/dataset.sqlite
datasette PATH/TO/dataset.sqlite

and visiting http://localhost:8001 in your browser.

Customization

But datasette provides a full-featured plugin system, which allows adding custom visualisations for the data. This can be paired with CLDF's metadata to extend the usability of datasette. Geographic coordinates for languages - for example - are well specified in CLDF, thus can generically be made available for datasette's cluster map plugin:

1. Install the plugin:

   pip install datasette-cluster-map

2. Write a datasette configuration file, with the following content:

   {
       "title": "",
       "plugins": {
           "datasette-cluster-map": {
               "latitude_column": "cldf_latitude",
               "longitude_column": "cldf_longitude"
           }
       }
   }

   This will make sure the appropriate columns in the SQLite database are detected.
3. Run datasette with the configuration:

   datasette PATH/TO/dataset.sqlite -m PATH/TO/datasette_metadata.json

Navigating to the LanguageTable of the database in datasette's interface will show a map with clusters of language locations.

datasette-cldf

A datasette setup which is even more tuned to CLDF data is available with the datasette-cldf Python package.