Remark: I wrote this post long time ago but never published it. Unfortunately BatchRefine is still on the old 2.x codebase of OpenRefine but some might still find it useful.  In this post I will show how you can use a headless OpenRefine instance called BatchRefine to process large amounts of CSV data without memory problems. A normal OpenRefine instance is in our experience not really suited for processing large files. If you ever tried to load several hundred thousand or even millions of lines...