Data API documentation

JIPipe communicates data with external applications via a standardized file system hierarchy. The specifications of this format are explained in this page / sub-pages.

The specification is used in following circumstances:

• Pipeline output
• Exported cache results
• Communication with external script languages (R, Python)

The data API is modular due to the different uses. It incorporates following elements:

• Data table row
• Data table
• Pipeline output
• Exported cache

Additionally, the data API incorporates elements from the JSON API to store metadata.

The following graph gives an overview how these relate to tech other:

The smallest unit is the data table row. It stores the files of a data type in its own standardized way. Please use the

Detailed descriptions

You will find detailed descriptions on the other pages.

• Data table

Organizes data into an annotated table

• Row folder

Data storage of a JIPipe data type in the file system

• Data types

Current list of data types available in JIPipe

• Pipeline output

A pipeline output folder is generated on running a pipeline using the Run function or via other means of running the pipeline and saving the results to the hard drive. The generated folder follows a hierarchical structure that is based on the location of data within graph compartments, nodes, and slots. It also contains a copy of the project that generated the data in a file project.jip. graph LR; Root["/"] --> DataTable["