Utility functions to handle Latvian letters with accents

This library will help you to:

• create diacritics-insensitive regular expressions with createDiacriticsInsensitiveRegExp
• remove accents from Latvian texts with unaccent

Possible use cases

Running diacritics-insensitive database queries

Goal: Finding accented text by unaccented user input, e.g. Raiņa bulvāris database entry by Raina bulvaris user input

SQLite

SQLite supports neither diacritics-insensitive, neither regular expression search by default.

Solution: Use unaccent utility function to create a separate column to store normalized (unaccented) text value and use it to perform queries.

MongoDB

Disclaimer: MongoDB comes with a text index, that allows running diacritics-insensitive $text queries. Unfortunately you can't combine text index with 2dsphere. In order to make diacritics-insensitive search still possible, the regular expression should be used.

MongoDB supports querying by a regular expression, like:

Address.find({ street: /Raina bulvaris/ })

In order enable diacritics-insensitive search and to make Raiņa bulvāris found, the diacritics-insensitive regular expression should be created using createDiacriticsInsensitiveRegExp:

Address.find({ street: createDiacriticsInsensitiveRegExp('Raina bulvaris') })

This will result in the following regular expression being queried:

Address.find({ street: /(?:R)(?:a|[Āā])(?:i|[īĪ])(?:n|[Ņņ])(?:a|[Āā])(?: )(?:b)(?:u|[Ūū])(?:l|[Ļļ])(?:v)(?:a|[Āā])(?:r)(?:i|[īĪ])(?:s|[Šš])/ })