This is (to be) a set of command-line tools for manipulating and transforming data. For the most part, the tools take in data from a file or STDIN as a tab-separated table of numbers, without at header, and returns the data in the same format.
Final names and tools TBD
Probably these are sub commands on a single command.
- select or cut- cut, select or manipulate columns
- sort - reorder rows
- aggr - group by and aggregate, or do window functions
- narrow - switch the data from wide to narrow format (a.k.a., stack, melt, normalize)
- widen - switch the data from narrow to wide format (a.k.a., unstack, pivot, denormalize)
- where - select rows
- format - outputs as human-readable text table, markdown, markup, etc.
- parse - parse into table from various formats
- plot - generate a plot of the data
- ...
The general format each command expects and outputs is tab-separated numbers without a header.
This can be altered in a couple ways. First, certain options will allow all commands to read (or output?) code in a slightly different format (e.g., with a header, CSV).
Second, special characters will provide additional information, such they are unlikely to show up unexpectedly, e.g., header and types. In particular, if the first character is !, the first row (excluding the leading !) is treated as a header row.