dv-api-metrics 0.3.0

Python code to collect metrics from datasets in Dataverse collections

dv-api-metrics

• This module provides Python scripts to collect metrics about Dataverse datasets and collections in demo.dataverse.org or dataverse.harvard.edu.
• The scripts are primarily used to support the CAFE project and will be archived when the Dataverse Hub supports additional metrics reports.

Requirements

• Requires Python 3.10 or greater
• Uses command line ($bash shell)

Installation

• Create dataverse.harvard.edu and/or demo.dataverse.org account
• Retrieve your API key
• On the command line, create a new directory, such as dv_api_metrics_reports
• Then, type: pip install dv-api-metrics to install the module
• On the command line, set $DATAVERSE_API_TOKEN to your API token
• Execute desired script

The reports can be run with simple commands, such as:

• dv-collection-subjects CAFE
• dv-collection-citations CAFE
• dv-harvest-counts CAFE
• dv-collection-inventory CAFE
• dv-monthly-datasets CAFE
• dv-harvest-views CAFE
• dv-monthly-downloads CAFE

These commands will produce the reports in your current directory (e.g., ~./dv_api_metrics_reports). Please note, these reports may fail randomly for various reasons related to accessing these via the REST APIs - most typically a 403 forbidden result code probably due to rate limiting. If a report fails, it will eventually succeed when run later.

Detailed Usage

The module includes the following scripts to collect metrics about Dataverse datasets and collections in demo.dataverse.org or dataverse.harvard.edu.

get_collection_dataset_citations.py

• Get a collection's dataset citations. Includes its subcollections.
• % python get_collection_dataset_citations.py <collection> --installation [hdv|demo]\ --filename <filename> --verbose

get_collection_dataset_inventory.py

• Get a collection's dataset inventory. Includes its subcollections.
• % python get_collection_dataset_inventory.py <collection> --installation [hdv|demo]\ --filename <filename> --verbose

get_collection_datasets_per_subject_count.py

• Get the total count of datasets per subject in a collection and its subcollections.
• % python get_collection_harvested_dataets_count.py <collection> --installation [hdv|demo]\ --filename <filename> --verbose

get_collection_harvest_dataset_counts.py

• Get a collection's harvested dataset counts. Includes its subcollections.
• % python get_collection_harvest_dataset_counts.py <collection> --installation [hdv|demo]\ --filename <filename> --verbose

get_collection_harvest_dataset_views.py

• Get a collection's harvested dataset unique views. Includes its subcollections.
• % python get_collection_harvest_dataset_views.py <collection> --installation [hdv|demo]\ --filename <filename> --verbose

get_collection_metrics.py

• Get monthly unique dataset download metrics for a named collection. Saves metrics to a tab-delimited file.
• % python get_collection_metrics.py <installation> <collection> \ --metrics [dvm | mdc] --filename <filename> --output [records|time_series] --verbose

get_collection_unique_monthly_downloads.py

• Get monthly unique dataset download metrics for a named collection. Saves metrics to a tab-delimited file.
• % python get_collection_unique_monthly_downloads.py <installation> <collection> \ --metrics [dvm | mdc] --filename <filename> --output [records|time_series] --verbose

Limitations

• Module uses existing Dataverse Metrics API endpoints or Native API endpoints.
• Make Data Count (MDC) metrics range from 2020-09 to the present.