/metatab-packages

Data Packages in Metatab format

Primary LanguageJupyter NotebookMIT LicenseMIT

Metatab Packages

Data Packages in Metatab format

This repository holds Metatab files for data packages for which we don't control the data -- all of the resource URLs are off site. You can use the metatab files to create local data packages with the metapack application.

For a guide to getting started using the metatab programs, see the tutorial for the Metatab Python module.

Creating Packages In This Repo

After cloning this repo, the typical process is:

  1. Create a subdirectory for the data origin
  2. In the origin directory, create a new metadata.csv file.
  3. Update the name elements in the metadata, and regenerate the name.
  4. Create a new directory with the dataset name, excluding the version.
  5. Move the metadata.csv file into the new dataset directory.

For example, to create the dataset for the California School Dashboard

$ mkdir cde.ca.gov
$ metapack -c

Open the metadata.csv file and make these changes:

Name.Dataset: accountability-dashboard
Name.Origin: cde.ca.gov

Then save the file and run:

$ metapack -u 
Changed Name
Name is:  cde.ca.gov-accountability_dashboard-1
$ mkdir cde.ca.gov/cde.ca.gov-accountability_dashboard
$ mv metadata.csv cde.ca.gov/cde.ca.gov-accountability_dashboard