grel

GREL, short for Google Refine Expression Language, is a scripting language embedded in OpenRefine (formerly Google Refine) that enables data transformations, cleansing, and extraction on tabular data. It is designed to be embedded in the OpenRefine workflow, allowing users to compute new values from existing ones, parse strings, and restructure rows during column transformations or facet operations.

The language operates on per-cell values and can reference the current row and basic metadata such as column name. It provides a typed set of operations for strings, numbers, dates, booleans, and arrays, along with a library of built-in functions. Common tasks include trimming whitespace, case conversion, splitting and joining strings, numerical calculations, date parsing, and regex matching and substitution. GREL also supports conditional expressions, as well as array and object manipulation, enabling more complex transformations without external scripting.

Usage examples appear in the OpenRefine expression editor, where users create transform expressions to populate new columns, replace values, or filter rows based on computed criteria. GREL expressions can be combined with OpenRefine features such as facets and clustering to clean and reshape data efficiently.

Development and status: GREL was developed for the Google Refine project and is now maintained as part of OpenRefine by the community. It remains a core component for data cleaning in OpenRefine, with ongoing updates and documentation provided by the project and user community.