Neptune GSP: Use UUID graph URIs instead of explicit default graph

[Copilot]

Neptune's GSP implementation was inserting all triples into an explicit default graph (via ?default parameter), breaking graph-aware semantics for TriG and other multi-graph formats. Named graphs were being flattened instead of preserved.

Changes

• GSP operations (publish, put, post): Generate temporary UUID graph URI (urn:uuid:...), POST/PUT to that graph, then DELETE it. This forces Neptune to process graph-aware RDF correctly.
• Configuration: Added _use_temp_graph option (default: True) for opt-out backward compatibility
• Error handling: Cleanup DELETE always executes, even on POST/PUT failure
• Documentation: Updated Neptune plugin README with feature explanation and config examples

Usage

# In system.conf
KNOWLEDGE_TYPE = 'neptune'
KNOWLEDGE_ENDPOINT = 'https://neptune.cluster.amazonaws.com:8182/sparql'
KNOWLEDGE_REGION = 'us-east-1'
KNOWLEDGE_USE_TEMP_GRAPH = True  # Default, enables graph-aware semantics

The UUID graph approach triggers Neptune's union graph semantics instead of explicit default graph insertion, preserving named graph structure from TriG data.

Note

Code review identified pre-existing serialization behavior in put()/post() that may serialize entire store instead of target graph. This matches existing database_utils.py implementation and is outside scope of this fix.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• neptune.example.com
  • Triggering command: /usr/bin/python3 python3 -m pytest tests/unit/test_neptune_plugin.py::TestNeptuneGSPOperations -v (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

The current GSP implementation (including the publish() call) inserts triples into an explicit default graph, as signified by the ?default parameter. This is incorrect because all RDF passed to neptune by Whyis is graph aware (usually as TRIG), and all triples end up in an explicitly named default graph, rather than using union semantics on the default graph. Change the neptune GSP implementation (publish, put, post) to provide a UUID-generated graph URI (prefereably in the uuid URI namespace) as the "default" graph to insert to, and then call to delete that default graph after the main call is complete. Make this a configurable option on the neptune database driver, but make the default to use this functionality.

Provide appropriate documentation and unit tests.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.