fix: distinguish diagnostic fields from entity identities - #135
Merged
Merged
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Fix the diagnostic regression identified after PR134: the valid scenario ID
examplescould intercept the aggregate model pathscenarios.examplesandselect one scenario instead of the whole collection.
Only the model's quoted-identity syntax now enters identity lookup. Ordinary
fields and numeric positions use their existing projection; the obsolete
numeric shortcut is removed from identity-only resolution. Admission rules,
failure codes, resource limits and public source contracts are unchanged.
Verification
model-semantic position and quoted identity at both wire orders.
unchanged code/path, and a model position1 span replaced with position0.
applicable provider CI are required before merge.
Final committed
npm run checkpassed clean at2b081df8314c74f2b5a497f255dc1f26918670a6, including all333 browser cases.Three fresh oracle-review lanes found no remaining named gap; the earlier
four-lane production review is retained for unchanged production bytes.
The first review found a real weakness in the new positional test: equal
content did not prove the original occurrence. The test-only follow-up repairs
that oracle. Existing production-review conclusions are retained separately
from the fresh oracle review; no green gate is treated as proof of perfection.
Scope
Two product files only. No new parser, public path AST, ID prohibition, source
cutover, package identity or registry publication. This closes REG-01 after
final gates; it does not complete P9-06 or the overall source-format decision.
Retro
The previous route review proved identifier/index disjointness but omitted
aggregate field names. The repair preserves the producer's existing syntactic
roles. Complete incoming-path inventory and exact occurrence oracles are the
local recurrence controls; no active skill policy was modified.