Skip to content

WEST batch - #40

Closed
Simon-McIntosh wants to merge 1 commit into
mainfrom
review/v0.10.0rc1+west-task-2e
Closed

WEST batch#40
Simon-McIntosh wants to merge 1 commit into
mainfrom
review/v0.10.0rc1+west-task-2e

Conversation

@Simon-McIntosh

Copy link
Copy Markdown
Owner

A review batch of WEST standard names, exported against the released v0.9.0 grammar.

What this carries

196 standard-name identities, frozen as the batch roster artifact (78,990 bytes, sha256 b75c04706c7d9bfd6a8d2288be9398fc7bed14c58b4153c198ffe52cdf5c06aa).

The source manifest resolved 355 data-dictionary paths to 214 distinct terminal identities with 31 unmatched. Of the 214 candidates the export emitted 196 and withheld 18, with 2 validation failures and no gate failures.

Grammar provenance

The catalog records the grammar it was actually built against:

grammar_version: 0.9.0
isn_model_version: 0.9.0

That version is a public release, checked four ways before the cut: the grammar checkout was clean with HEAD exactly at annotated tag v0.9.0; the tag is a plain major.minor.patch release rather than a candidate; it is present on the upstream remote; and 0.9.0 is published on the package index as a final release. A development, candidate or fork-only grammar fails each of those checks by name.

Quality gates

All eight export gates passed with none skipped, and the approved-baseline check passed:

Gate Result
catalog_status pass
graph_tests pass
score_thresholds pass
cross_field_consistency pass
divergence_detection pass
identity_token_collision pass
exclusion_accounting pass
manifest_source_accounting pass

Every emitted name also renders canonically under this grammar: parsing and re-composing the accepted cohort gives 2352 stable identities and zero that render under a different spelling, so no name in this batch carries a pre-ruling form.

What was withheld, and why

Two identities were withheld on validation rather than admitted as passing entries: total_electron_count and vertical_coordinate_of_ece_channel, both carrying the invalid physics-domain value unscoped.

The 31 unmatched source paths fall into two groups. Sixteen name no physical quantity — fit weights, bare and nested time axes, a solver iteration count, contour-tree bookkeeping coordinates and instrument dimensions — and were never candidates for a standard name. Fifteen do denote physical quantities whose candidate name is currently unreachable at its lifecycle stage, most of them having exhausted their review attempts; barometry/gauge/pressure, ece/channel/optical_depth and ic_antennas/antenna/module/strap/width_phi are representative. Those return in a later batch once their names are regenerated, and each is listed with its blocked candidate in the accompanying exclusion accounting.

Reviewing this

The roster artifact is the reproducible batch identity: it lists every emitted name, so a disagreement about membership is settled by reading it rather than by re-running the export. The grammar pin is a published release, so the same catalog can be rebuilt from v0.9.0 alone.

Published 196 entries for the west-task-2e review batch and withheld 18 from this release.
The frozen review cohort is identified by v0.10.0rc1+west-task-2e.
@Simon-McIntosh Simon-McIntosh changed the title WEST review batch: 196 standard names against the v0.9.0 grammar WEST batch Sep 7, 2026
@Simon-McIntosh

Copy link
Copy Markdown
Owner Author

Closing: this batch was cut against the grammar repository in error. The standard-name catalog belongs in imas-standard-names-catalog, and the batch will be re-cut there.

@Simon-McIntosh
Simon-McIntosh deleted the review/v0.10.0rc1+west-task-2e branch September 7, 2026 09:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant