Dynamic metadata block generation and DVUploader integration

[JR-1991]

Overview

This pull request introduces the dynamic generation of metadata blocks for Dataverse version >= 5.14, which is based on IQSS/dataverse#9213. The metadata schemes are retrieved from a Dataverse instance and converted into metadata block objects that can be filled with metadata. When uploaded, the objects are transformed into compliant Dataverse JSON and sent through pyDataverse's standard dataset creation/update methods. Additionally, this pull request includes python-dvuploader as a file upload solution, allowing for parallel native/direct uploads to Dataverse. Furthermore, file downloads are also parallelized.

TLDR

• Generates metadata block objects on the fly
• Integrates python-dvuploader for file uploads
• Extension of the unit/integration test framework (Coverage: 76%)
• Parallel file downloads as well as subtractive selection using patterns
• List metadata block configurations for an overview
• Migration to pyDantic V2

Example

from easyDataverse import Dataverse

# Connect to a Dataverse installation
dataverse = Dataverse(
  server_url="https://demo.dataverse.org",
  api_token="MY_API_TOKEN",
)

# Initialize a dataset
dataset = dataverse.create_dataset()

# Fill metadata blocks
dataset.citation.title = "My dataset"
dataset.citation.subject = ["Other"]
dataset.citation.add_author(name="John Doe")
dataset.citation.add_dataset_contact(name="John Doe", email="john@doe.com")
dataset.citation.add_ds_description(value="This is a description of the dataset")

# Upload files or directories
dataset.add_file(local_path="./my.file", dv_dir="some/dir")
dataset.add_directory(dirpath="./my_directory", dv_dir="some/dir")

# Upload to the dataverse instance
dataset.upload("my_dataverse_id")