tablo is a plain text interchange format for tabular data. It is more expressive than CSV while remaining easy for people to read and write.
It adds explicit headers, datatypes, and cell formatting to address shortcomings of delimiter-separated formats like CSV or TSV. tablo solves delimiter collision issues by having well-defined quoting and escaping rules that are familiar to users of formats like JSON or common programming languages like Python or JavaScript.
the Microsoft version of CSV is a textbook example of how not to design a textual file format
—Eric S. Raymond, The Art of Unix Programming
Stated simply, there is no single CSV standard. It exists as a myriad of informal variants whose implementation varies from vendor to vendor. Character encodings and escape sequences vary from one application to the next, and the ambiguities in various edge cases means that the output of one application may not be readable by another.
tablo is designed to solve a number of ambiguities and shortcomings in CSV.
One of the first obvious differences is that header rows are optional, but well-defined. In other words, a document may or may not contain a header, but determining whether the document includes a header is always unambiguous.
A crucial aspect of the tablo format is that it doesn't make assumptions about the type of data in each cell. If a value is surrounded by quotes, it is always a string. If a value is a number without quotes, it is always a number. If a value is an ISO-8601 formatted date preceded by a hash mark, it is always a datetime.
This repository contains the tablo specification, reference implementations in select languages, and a set of example files.