feat: grant DAKP sparse qualifier stack on pinned association classes - #188
Merged
Merged
Conversation
Grants anatomical_context_qualifier, sex_qualifier, population_context_qualifier, frequency_qualifier, and temporal_context_qualifier to EntityToDiseaseAssociation / EntityToPhenotypicFeatureAssociation via CLASS_FIELD_OVERRIDES, so DAKP assertion rows carrying the sparse qualifier stack keep those qualifiers on the edge instead of losing them in prune_to_class. species_context_qualifier stays ungranted (disabled); temporal_interval_qualifier and severity_qualifier stay ungranted (unsatisfiable).
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Advanced Run ID: Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
SkyeAv
changed the title
Compresses the CURIE-ranged qualifier sentence so the expanded slot list fits the budget the agent-speed guard enforces.
SkyeAv
added a commit
that referenced
this pull request
Sep 15, 2026
Cut 19.0.0 and bump the package version in pyproject.toml, CITATION.cff, and uv.lock. Major: five breaking changes ship. The `tablassert` console command now requires the optional `[cli]` extra, since `cyclopts` and `rich` left the base install (#187). Level-one normalization redefines fullmap keys as cleaned Unicode-lowercase, Porter2-normalized, byte-ordered token sets and moves the database to schema v6, so every schema-v5 fullmap is rejected and must be rebuilt (#171, #172, #174, #175, #176). `build-fullmap --aria2c` is gone -- aria2c is used automatically whenever the `[aria2]` extra is installed (#178) -- and `--taxon-allowlist` is gone because the built-in top-100 experimental-taxon allowlist now applies to every build, guarding both reuse paths by `META.taxon_allowlist` identity (#153). Logging now requires the `[log]` extra and is fully disabled without it (#152). Features: `tablassert.net` is a new stdlib-only transient/permanent classification and retry seam shared by the agent and the BABEL downloader (#163, #164), with one bounded jittered retry layer across the inner agent, reflexion, and judge (#167) and machine-readable `error_code` values on skipped checkpoint records (#166). `build-kg` gained an automatic ephemeral shared-prefix TCode cache keyed by content-addressed XXH64 op digests (#155-#159, #161). The `--distill` corpus became schema-uniform v2 with a sibling `outcomes.ndjson`, a deterministic `RewardConfig`-tunable reward, and the new zero-dependency `tablassert distill-weigh` command (#180-#183). `CLASS_FIELD_OVERRIDES` grants the DAKP sparse qualifier stack on the pinned association classes (#188). Fixes: `distill-export` partitions its corpus by content and unions the schema across record files, so a mixed v1/v2 directory no longer CastErrors or silently stringifies a column (#184); BABEL retry warnings survive the loguru sink; PMC downloads are idempotent, atomic, and bounded-parallel (#165); reflexion and judge prompts are hard-bounded (#168). Changelog: - versioned the Unreleased section as 19.0.0 - 2026-09-15 - merged three duplicate `### Added` blocks into one - moved the two `BREAKING:` entries from `### Changed` into `### Breaking Changes` and added a `**Migration:**` note to all five - added PR links to all 17 entries, which carried none - added the missing entries: the distill v2 corpus/reward/distill-weigh stack (#180-#183), the distill-export schema-drift fix (#184), the TCode run-cache detail (#155-#161), and the README badge removal (#154) Docs: none needed here. Every shipped doc change landed with its own PR, and the docs source-of-truth gate passes against the bumped tree. Testing: - make check -> exit 0 - uv run ruff check . -> All checks passed! - uv run pyright -> 0 errors, 0 warnings, 0 informations - uv run pytest -> 1622 passed, 52 skipped - cargo test --manifest-path rust/Cargo.toml -> 156 passed, 0 failed - cargo clippy --all-targets -- -D warnings -> clean - uv lock --check -> resolved 169 packages, lock current - uv run mkdocs build --strict -> exit 0 - docs SSOT + CLI coverage after the changelog edit -> 204 passed
This was referenced Sep 15, 2026
SkyeAv
added a commit
to glusman-team/dakp
that referenced
this pull request
Sep 15, 2026
Take the release that grants DAKP's sparse qualifier stack (SkyeAv/Tablassert#188) and absorb the three breaking changes that ride with it. - `pyproject.toml`: `tablassert[cli,log,qc]>=19.0.0,<20`. 19.0 moved the `tablassert` console script behind the optional `[cli]` extra (its launcher preflights the extra and exits before importing the app) and made logging a silent no-op without `[log]`. `TablassertRunner` shells out to that console script and reads its stdout/stderr, so both extras are load-bearing here, not optional niceties. - Fullmaps must now be `tablassert.fullmap.v6`. 19.0 rebased level-one normalization onto stemmed, deduped, byte-sorted tokens, which changed the RECORDS key SPACE: a v5 database is not an older layout of the same data, its keys can never be hit by a normalized query, and v1-v5 are rejected on read. The runner's missing-fullmap error and the `FULLMAP_DEFAULT` docs now say v6 and explain why it is not a relayout. - No emitted-config change lands here; the qualifier wiring the grant enables is the next commit. Verification: `uv run pytest` -> 1131 passed (coverage 99.84%, misses are the pre-existing aria2c/loopback network-gated lines, unchanged by this commit); `ruff check` + `ruff format --check` clean; `uv run tablassert --help` exits 0 against the installed 19.0.0, and `CLASS_FIELD_OVERRIDES` reports all five qualifier grants.
SkyeAv
added a commit
to glusman-team/dakp
that referenced
this pull request
Sep 15, 2026
Take the release that grants DAKP's sparse qualifier stack (SkyeAv/Tablassert#188) and absorb the three breaking changes that ride with it. - `pyproject.toml`: `tablassert[cli,log,qc]>=19.0.0,<20`. 19.0 moved the `tablassert` console script behind the optional `[cli]` extra (its launcher preflights the extra and exits before importing the app) and made logging a silent no-op without `[log]`. `TablassertRunner` shells out to that console script and reads its stdout/stderr, so both extras are load-bearing here, not optional niceties. - Fullmaps must now be `tablassert.fullmap.v6`. 19.0 rebased level-one normalization onto stemmed, deduped, byte-sorted tokens, which changed the RECORDS key SPACE: a v5 database is not an older layout of the same data, its keys can never be hit by a normalized query, and v1-v5 are rejected on read. The runner's missing-fullmap error and the `FULLMAP_DEFAULT` docs now say v6 and explain why it is not a relayout. - No emitted-config change lands here; the qualifier wiring the grant enables is the next commit. Verification: `uv run pytest` -> 1131 passed (coverage 99.84%, misses are the pre-existing aria2c/loopback network-gated lines, unchanged by this commit); `ruff check` + `ruff format --check` clean; `uv run tablassert --help` exits 0 against the installed 19.0.0, and `CLASS_FIELD_OVERRIDES` reports all five qualifier grants.
SkyeAv
added a commit
to glusman-team/dakp
that referenced
this pull request
Sep 15, 2026
…nfigs Tablassert 19.0.0's CLASS_FIELD_OVERRIDES grant (SkyeAv/Tablassert#188) keeps DAKP's five sparse qualifier slots on the pinned association classes through prune_to_class — the upstream unblock the qualifier columns were waiting for. Wire them now that the grant exists. - `_TABLE_QUALIFIERS`: approved-treats and contraindication tables declare `anatomical_context_qualifier` / `sex_qualifier` / `population_context_qualifier` / `frequency_qualifier` / `temporal_context_qualifier`, each backed by the assertion column its sentence-hosted attachment (contexts.attach_qualifiers) already populates. The FAERS table declares none: its writer populates no qualifier column (a FAERS row has no sentence to host attachment), and an always-blank backing column would encode nothing. - Every qualifier emits `nullable: true`: the columns are sparse by construction, and Tablassert's nullable semantics keep the edge when a cell is blank or unresolved, omitting only the qualifier. - New `_QUALIFIER_GUARD` replaces the hardcoded Disease-only guard: class-ranged slots carry a category allow-list matching their Biolink range (Disease / AnatomicalEntity / BiologicalSex / PopulationOfIndividualOrganisms), while the type-ranged slots (frequency_qualifier: UO `frequency value`; temporal_context_qualifier: `xsd:string` `time type`) carry none — no Categories member names an honest allow-list for their values, so resolution falls through the fullmap with nullable keeping the edge on misses. - species_context_qualifier stays unwired: Tablassert deliberately disables it (DISABLED_EDGE_FIELDS — config load rejects a manual declaration); taxon context is auto-derived from node metadata instead. - Grant tripwire widened: the test now pins all six granted slots on both pinned classes, so a Tablassert downgrade fails at test time instead of silently pruning the stack off every edge at build time. - tables/*.yaml regenerated (the suite enforces byte-equality; graph.yaml's RIG consideration now names the #188 grant). Verification: `uv run pytest` -> 1131 passed (coverage 99.89% locally, misses are the pre-existing aria2c/loopback network-gated lines CI executes); `ruff check` + `ruff format --check` + `pyright` clean; each regenerated Section validates against the installed tablassert 19.0.0 (`Section.model_validate`) with the expected nullable/guard shape on all three tables.
SkyeAv
added a commit
to glusman-team/dakp
that referenced
this pull request
Sep 15, 2026
…nfigs (#29) Tablassert 19.0.0's CLASS_FIELD_OVERRIDES grant (SkyeAv/Tablassert#188) keeps DAKP's five sparse qualifier slots on the pinned association classes through prune_to_class — the upstream unblock the qualifier columns were waiting for. Wire them now that the grant exists. - `_TABLE_QUALIFIERS`: approved-treats and contraindication tables declare `anatomical_context_qualifier` / `sex_qualifier` / `population_context_qualifier` / `frequency_qualifier` / `temporal_context_qualifier`, each backed by the assertion column its sentence-hosted attachment (contexts.attach_qualifiers) already populates. The FAERS table declares none: its writer populates no qualifier column (a FAERS row has no sentence to host attachment), and an always-blank backing column would encode nothing. - Every qualifier emits `nullable: true`: the columns are sparse by construction, and Tablassert's nullable semantics keep the edge when a cell is blank or unresolved, omitting only the qualifier. - New `_QUALIFIER_GUARD` replaces the hardcoded Disease-only guard: class-ranged slots carry a category allow-list matching their Biolink range (Disease / AnatomicalEntity / BiologicalSex / PopulationOfIndividualOrganisms), while the type-ranged slots (frequency_qualifier: UO `frequency value`; temporal_context_qualifier: `xsd:string` `time type`) carry none — no Categories member names an honest allow-list for their values, so resolution falls through the fullmap with nullable keeping the edge on misses. - species_context_qualifier stays unwired: Tablassert deliberately disables it (DISABLED_EDGE_FIELDS — config load rejects a manual declaration); taxon context is auto-derived from node metadata instead. - Grant tripwire widened: the test now pins all six granted slots on both pinned classes, so a Tablassert downgrade fails at test time instead of silently pruning the stack off every edge at build time. - tables/*.yaml regenerated (the suite enforces byte-equality; graph.yaml's RIG consideration now names the #188 grant). Verification: `uv run pytest` -> 1131 passed (coverage 99.89% locally, misses are the pre-existing aria2c/loopback network-gated lines CI executes); `ruff check` + `ruff format --check` + `pyright` clean; each regenerated Section validates against the installed tablassert 19.0.0 (`Section.model_validate`) with the expected nullable/guard shape on all three tables.
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.
Grants
anatomical_context_qualifier,sex_qualifier,population_context_qualifier,frequency_qualifier, andtemporal_context_qualifiertoEntityToDiseaseAssociation/EntityToPhenotypicFeatureAssociationviaCLASS_FIELD_OVERRIDES, so DAKP assertion rows carrying the sparse qualifier stack keep those qualifiers on the edge instead of losing them inprune_to_class.Qualifier grants
src/tablassert/biolink.py, alongside the existingdisease_context_qualifier/regulatory_approvalsgrants.species_context_qualifierstays ungranted (inDISABLED_EDGE_FIELDS, never emittable);temporal_interval_qualifierandseverity_qualifierstay ungranted (inUNSATISFIABLE_EDGE_FIELDS, attached to no class, so a grant could never validate).extra_forbiddensurface;_validation_recordalready strips granted fields from its in-memory copy, and the existing tripwire test fails if a future biolink-model release attaches a granted slot.Agent guidance
src/tablassert/agent.pynow namespopulation_context_qualifier,frequency_qualifier, andtemporal_context_qualifieralongside the existing three in the CURIE-ranged qualifier sentence.Docs
docs/configuration/table.md:category_overriderow and allow-list section now document the sparse-stack grants and the three deliberate exclusions.CHANGELOG.md: Unreleased entry describing the grants and exclusions.Testing
uv run pytest tests/test_biolink.py tests/test_lib.py::test_prune_to_class_keeps_class_field_override_grants tests/test_lib.py::test_prune_to_class_keeps_override_only_slots tests/test_models.py tests/test_rig.py -p no:cacheprovider -o addopts="" -q->180 passeduv run pytest tests/test_docs_source_of_truth.py tests/test_agent_compact.py tests/test_agent_docs.py -p no:cacheprovider -o addopts="" -q->221 passeduv run pytest tests/ -p no:cacheprovider -o addopts="" -q --ignore=tests/test_qc.py->1607 passed, 52 skipped, 2 failed; both failures are pre-existing environment gaps (missing[qc]extra packages; a log-capture assertion), unrelated to this change.uv run ruff check-> clean;uv run ruff format --check-> clean after one applied reformat.