Repository navigation
unstructured2graph: decouple label promotion from ontology enforcement - #251
Merged
Merged
Conversation
Closes #249. enforce_ontology previously conflated two concerns: promoting entity_type to a real Memgraph label, and restricting/validating that promotion against a fixed vocabulary. Splits them into two independent parameters: - promote_labels: bool = False -- promote every entity_type to a label (sanitized to a safe Cypher identifier, e.g. "natural object" -> :NaturalObject), no fixed vocabulary, no ontology_conformant flagging. - enforce_ontology: bool = False -- restrict promotion to ontology_path's vocabulary and flag anything outside it ontology_conformant=false. Takes precedence over promote_labels when both are set. New promote_all_entity_types_to_labels() in memgraph.py implements the unrestricted path: discovers distinct entity_type values under the workspace label (no static list to iterate, since there's no ontology), sanitizes each into a PascalCase label via _entity_type_to_label(), skipping values that can't be turned into a safe identifier. Wires sessions-graph's reconcile_session() to accept and forward the same three parameters, and updates its CLI (the real production entry point) to pass enforce_ontology=True, so reconciled session content actually gets typed labels going forward -- previously it called from_texts() with neither, so it got no promotion at all despite #248 landing. Note: unstructured2graph's real-LLM e2e suite (test_e2e_lightrag.py) and sessions-graph's test_e2e_reconciliation.py currently fail against a live Memgraph for an unrelated, pre-existing reason -- see #250 (MemgraphDocStatusStorage missing abstract methods required by lightrag-hku 1.5.5). Verified this reproduces identically on a clean main checkout before touching any code here. All non-e2e tests pass (60 in unstructured2graph, 24 in sessions-graph; 218 across the full 5-package suite).
antejavor
force-pushed
the
unstructured2graph-decouple-label-promotion
branch
from
August 4, 2026 08:19
ede3918 to
1f349e1
Compare
nltk 3.10.1 (2026-08-01) added a CWD-import security check (nltk/inisec.py) that raises ImportError during pytest collection -- uv pip install -e (what CI's test-unstructured2graph job runs) resolves nltk live rather than from the lockfile, so it picked up the brand-new release immediately. Legitimate upstream security hardening, not a compromise -- verified nltk/inisec.py is real code in nltk/nltk's own GitHub repo -- just not something to adapt to under time pressure here. Verified against the exact CI install path (uv venv + uv pip install -e .[test] in a throwaway venv): nltk now resolves to 3.9.4 and all 60 tests pass, where they previously failed on collection.
This was referenced Aug 10, 2026
Closed
Merged
3 tasks done
antejavor
added a commit
that referenced
this pull request
Aug 18, 2026
…try_points registry) (#295) PyPI has been stale for months relative to main -- these four packages haven't been released since before most of maps #275/#288's work landed, so the actual subagent-nesting model (Agent node, SPAWNED inference, HAS_AGENT, USED_SKILL-on-Agent attachment) isn't published anywhere yet. - actions-graph 0.1.2 -> 0.2.0: first-class Agent node model, SPAWNED inference rule, and the agent_spawning_tool_names default-name fix, all verified live against a real Claude Code session. - skills-graph 0.1.3 -> 0.2.0: USED_SKILL attaching to the specific Agent (not just the flat Session) when skill usage happens inside a subagent. - sessions-graph 0.4.0 -> 0.5.0: reconcile_session() now writes episodic Session.summary, plus config propagation fixes to the reconcile subprocess. (0.4.0 itself was already unreleased -- the ontology/label- promotion work from #248/#251.) - agent-context-graph 0.1.9 -> 0.2.0: entry_points-based runtime plugin registry. (main's 0.1.9 already diverged from the last published 0.1.9 build without a version bump -- #273 landed after the last release.) No code changes -- version bumps only, verified compatible with existing consumers (all internal >= constraints, no upper bounds).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Closes #249.
enforce_ontology(introduced in #248) previously conflated two different concerns under one flag: promotingentity_typeto a real Memgraph label, and restricting/validating that promotion against a fixed vocabulary. This splits them into two independent parameters onfrom_texts()/from_unstructured():promote_labels: bool = False— promote everyentity_typeto a real Memgraph label, sanitized into a safe Cypher identifier (e.g."natural object"→:NaturalObject), with no fixed vocabulary and noontology_conformantflagging (there's nothing to be non-conformant relative to).enforce_ontology: bool = False— restrict promotion toontology_path's vocabulary (or the bundled default), flagging anything outside itontology_conformant: false. Takes precedence overpromote_labelswhen both are set — it's the stricter, ontology-gated mode.New
promote_all_entity_types_to_labels()inmemgraph.pyimplements the unrestricted path: since there's no fixed vocabulary to iterate over, it discovers distinctentity_typevalues actually present under the workspace label, then sanitizes each into a PascalCase label via_entity_type_to_label()— skipping (and logging) any value that can't be turned into a safe identifier.sessions-graph wiring:
reconcile_session()now accepts and forwardspromote_labels/enforce_ontology/ontology_path. Its CLI (the actual production entry point) now passesenforce_ontology=True— previously it calledfrom_texts()with neither, so reconciled session content got zero label promotion despite #248 landing.Also opened #250: while verifying this, found that
unstructured2graph's andsessions-graph's real-LLM e2e suites currently fail against live Memgraph for an unrelated, pre-existing reason —MemgraphDocStatusStorageis missing 4 abstract methods thatlightrag-hku==1.5.5(already pinned onmain) requires. Confirmed this reproduces identically on a cleanmaincheckout before touching any code here, so it's not something this PR introduced or something in scope to fix here.Test plan
unstructured2graphunit tests —_entity_type_to_label()sanitization (multi-word, snake_case, digit-leading rejection),promote_all_entity_types_to_labels()(distinct-value discovery, per-value promotion, unsanitizable-value skipping, noontology_conformanttouching),promote_labels/enforce_ontologyprecedence in the loader layer.sessions-graphunit tests —reconcile_session()forwards all three new kwargs tofrom_texts(); CLI passesenforce_ontology=True.scripts/dev-memgraph.sh test) — 218 passed across all five packages; the only failures are the 6 real-LLM e2e tests blocked by lightrag-memgraph: MemgraphDocStatusStorage missing 4 abstract methods required by lightrag-hku 1.5.5 #250 (verified pre-existing, unrelated).