1.6.0

Additions

• insertSetIndexColumn for partitioning a column's values into sets and assigning a unique id to each. Useful for spreading a table comprising only key and value columns.

Bug Fixes

• Better handling of ragged tables. Some transposing operations that encountered empty columns previously resulted in empty tables; they now transpose all non empty rows and columns. All ragged tables now padded with empty strings.

Other Changes

• Some code cleanup to reduce download footprint.

1.5.0

Additions

• splitAt function for bisecting data values at any position.
• disaggregate for flexible disaggregation of data values using regular expressions.
• normalize for splitting a table into a key table and value table
• moveColumnToEnd for reordering of table columns

Refactoring

• Computation of gather and rowFilter operations now much more efficient and suitable for larger data tables.

Minor

• Minor improvements to API documentation.
• Table normalization added to examples.
• Additional tests for new functions.

1.4.0

Additions

• toCSV and toDelimited for converting a table into CSV and other text delimited strings. Useful when saving a table as a file.

Minor

• Minor improvements to API documentation.

1.3.0

Additions

• headTail convenience function for splitting a String into a tuple separating its first and remaining characters.

Minor

• Add example of multi column.-group gathering to API doc (uses headTail for more compact bisection of column values)

1.2.0

Additions

• insertIndexColumn for adding a column containing row-unique indices. Useful for relational joining.

1.1.0

Additions

• insertColumnFromJson to extract values from JSON arrays of objects as table columns.

Minor

• Minor improvements to API documentation formatting.

Initial Release

Data Shaping and Tidying

Leaning heavily on the principles of the tidyverse, this package makes it easy to reshape and tidy tabular data for data analysis and visualization.

Use cases include:

• Importing tabular data from a CSV file for analysis.
• Joining relational tables for data query.
• Editing, filtering and mapping rows and columns of data in a table.
• Creating tidy data for visualization with elm-vegalite, elm-vega, elm-visualization and other visualization packages.
• Building data science applications in Elm.

Tidy data are tables of data where

• each variable is in a column.
• each observation is a row.
• each value is a cell.

Unfortunately, many datasets are not in this format; the tidy package allows you to transform data into tidy tables so you spend less time fighting with data tools and more time working on your analysis and visualization. For a more complete description of tidy data, see Wickham (2014).