vein graph: legacy case-variant labels no longer shadow a type; wfbench survives a failed roster write - #1656
Merged
Conversation
…irst
On swarm38 every jarvis-typed write failed UNKNOWN_TYPE ("no such Schema
in the graph") although the EvalSet Schema exists: the graph also
carries legacy labels Evalset / Evaltrigger / Evalrequirement /
Evaltriggeroutput with zero nodes, and resolveType matched labels
case-insensitively with LIMIT 1, canonicalizing to a spelling that has
no Schema. Resolve against Schema.type first, then labels, preferring
an exact-case match in both. Live test registers the legacy labels and
proves an EvalSet write still lands under the schema-backed label.
(cherry picked from commit d1a8377)
…errored run graph/* steps return error strings (fail-soft), so a roster that did not land shows up as a missing ref_id; the neighbors hop then rejected the undefined ref_id and the run errored after the callback had already gone out (swarm38). Gate hop/edge/link on roster_ok; the record chain and result run either way and carry the write results as diagnostics. (cherry picked from commit 153d49d)
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Follow-up to #1654 (merged). On swarm38 the strut runs scored and Hive accepted the callback, but the lab marked both runs
errorand nothing reached the graph. Two causes.vein resolves types via
Schema.typebefore labels (engine)Every jarvis-typed write on swarm38 failed
UNKNOWN_TYPE: EvalSet: unknown node type (not a Vein type and no such Schema in the graph)— whileMATCH (s:Schema {type:'EvalSet'})returns the schema and Jarvis lists it. That graph also carries legacy labelsEvalset,Evaltrigger,Evalrequirement,Evaltriggeroutput(zero nodes; Neo4j keeps the tokens).resolveTypematcheddb.labels()case-insensitively withLIMIT 1, canonicalized to the spelling that has no Schema, and the write died. Now:Schema.typefirst, labels as fallback, exact case preferred in both. Live test registers the legacy labels and proves anEvalSetwrite lands under the schema-backed label, never the legacy one.wfbench: a failed roster write degrades to no-roster, never an errored run (harness)
graph/*steps are fail-soft (error strings), so a missing roster surfaced as an undefinedref_id, the neighbors hop's schema rejected it, and the run errored — after the callback had gone out.roster_okgate on hop/edge/link; the record chain and result run either way, carrying the write results as diagnostics.Verification
tscclean;npm test665/665; livenpm run test:graph141/141 against a throwaway Neo4j (incl. the new legacy-label case).tscclean;npx tsx src/lab/wfbench/smoke.tsgreen (35-step workflow validates).default, the lab run endssuccess, and Hive's rubric reader finds the EvalSet.Not addressed: Hive's own roster upsert also fails on swarm38 (Jarvis has no EvalSet for the slug;
rosterUpsertOutcome: okis the return-null path). Likely the same label pair on the Jarvis side.