Spreadsheet tables are often labeled, and these labels effectively constitute types for the data in the table. In such cases tables can be considered to be built from typed data where the placement of values within the table is controlled by the types used for rows and columns.
We present a new approach to the transformations of spreadsheet tables that is based on transformations of row and column types.
We illustrate the basic idea of type-based table construction and transformation and lay out a series of research questions that should be addressed in future work.
This example shows actual table transformations needed by an Excel user:
- (a) A table showing the earnings of three companies in three different quarters.
- (b) The same data in linearized form.
- (c) The data for each company aggregated (maximum in a quarter) over all quarters.
Tables are the fundamental data structure of spreadsheets. Investigating the properties of tables as well as their transformations should thus be a priority of spreadsheet research.
2018, 5th International Workshop on Software Engineering Methods in Spreadsheets, October