Automated localization of embedded databases

What about the thousands of records in my apps SQLite database?

Problem

Content rich apps often contain embedded SQL databases which are not detected by traditional localization systems. There might be one screen used to show, for example, a single selection out of one thousand possible choices. The translator must translate all one thousand menu selections, but to the app developer, it is a single screen. String extraction, visual display of the data, and text overrun detection of embedded SQL database content can be a very manual process for the developer.

Solution

The solution for translating an embedded SQL database is to extract the data by doing a SQL export to a text file of SQL statements that will recreate the database. The strings from the SQL export are displayed visually to the translators in a browser, in context, with text overruns detected. Once all of the content is translated, language specific SQL databases are recreated using the translated text and added back into the app.

Impact

The technology takes a single screen and expands it out into all possible versions of that screen for a translator to review and translate. All the visual context and text overrun capabilities are presented to the translator. The developer then receives back all the data in a translated embedded SQL database, one for each translated language. This makes it easier for translators to provide accurate translations the first time, while still allowing the developer to use embedded SQL databases.

Status

US Patent Application Number 62065717
Patent Filed 19-OCT-2015