DSPortal

DSPortal is a Python-based utility that simplifies ingesting content into a DSpace 7/8 repository via the REST API. It handles authentication, CSRF protection,

Designed for repeatable and scriptable metadata and file ingestion workflows.

Example methods for item creation, bundle setup, and bitstream uploads.

---

Features

• Authenticated REST session with CSRF & token handling
• Easy integration with external metadata extraction tools

---

Requirements

• Python 3.8+
• requests
• Access to a DSpace 7/8 instance with REST API enabled

---

Example Usage

1. Authenticate & Create Session

session = DSpaceSession(api_base="https://carleton-dev.scholaris.ca/server/api")
session.authenticate(user="dspace_admin", password="dspace_password")

2. Create a New Item

collection_id = "uuid-of-target-collection"

metadata_payload = {
    "name": "Title of Work",
    "metadata": "Insert structured metadata here",
    "inArchive": True,
    "discoverable": True,
    "withdrawn": False,
    "type": "item"
}

item_uuid, item_handle = item_creation(session, collection_id, metadata_payload)

3. Create Bundles for the Item

og_bundle_id = bundle_creations(session, item_uuid)

4. Upload Original Files

file = path/to/files
upload_files(session, package_data, og_bundle_id, file, OG_BITSTREAM_PAYLOAD)

---

Developer Notes

• Token Expiration: Sessions expire after ~30 mins. Call ensure_auth_valid(user, password) to auto-refresh.
• CSRF Handling: Automatically managed after login and per request.
• Error Handling: Logs basic failures; expand logging if needed.
• Large Files: Files >1GB are uploaded with MultipartEncoder (commented out in this sample; plug in your encoder module as needed).

---

Furture Features

• CLI Wrapper via click
• Retry logic for flaky uploads
• Logging system integration
• Better metadata schema validation before submit

---