Contents

This repository contains scripts for

1. uploading to a Dataverse server (dvpost)
2. cloning a Dataverse setup (sub-dataverses, groups and roles) to another Dataverse server (dvclone)
3. gathering statistics about file types and sizes (dvstats)
4. converting a statistics file (SPSS or SAS) to a ZIP file that contains the data in CSV form, plus a codebook (stats2scv.py)

dvstats --filesize prints to standard output a CSV file with the file contents of all datasets. This can be used to calculate storage sizes per dataverse, for example.

dvstats --status prints to standard output a CSV file with the status and authors of all datasets. This can be used to find unpublished datasets, for example.

All scripts use a simple interface class dave (dataverse) that uses the Dataverse native API.

Installation

There is one dependency: requests, a Python module that offers a simple and elegant HTTP interface. Install this with (in a virtual environment, or globally):

pip install requests

Configuration

All scripts assume that the API keys can be found in a configuration file ~/.config/dataverse.json in the home directory of the user. The user should generate the API keys in the web interface of Dataverse. Note: the API keys have a limited lifetime! The attribute root can be useful if you can to run tests on the Harvard demo dataverse, for example.

This configuration file should look like this:

{
"production": {"url": "https://dataverse.nl",      "root": "UMCU", "key": "..."},
"demo":       {"url": "https://demo.dataverse.nl", "root": "UMCU", "key": "..."}
}