fix(fit): strip @ in compare --exclude-chains; mh-summary clean break#422
Merged
Conversation
vsbuffalo
force-pushed
the
fix/mh-clean-break-and-exclude-at
branch
from
July 10, 2026 21:43
03157ea to
937d6a9
Compare
…an break Q1. `compare --exclude-chains fit:ids` matches the fit name VERBATIM against the compared fits — a run-store-handle fit is named `@a`, a path-named one `ctl_rm.toml`; the `@` is part of a handle name, not a sigil. Following the `--help` (whose example is the handle name `@a`), a user can prepend a spurious `@` to a path-named fit and get a bare "no compared fit named '@ctl_rm.toml'". Now, when dropping a leading `@` would match, the error names the fix, and the `--help` states the name is matched verbatim. Q2. The pmmh_summary.json -> mh_summary.json rename (#415) is a clean break. read_convergence no longer falls back across sibling summary names — it reads only the method's own <algorithm>_summary.json, matching PmmhStageResult::load. A pre-rename mh fit (only pmmh_summary.json on disk) is NotAssessed and must be re-run, never read under the old name.
vsbuffalo
force-pushed
the
fix/mh-clean-break-and-exclude-at
branch
from
July 10, 2026 22:10
937d6a9 to
ca4df22
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Two convergence-tooling fixes garki hit on
75eb8def. Both verified against the installed code (they are real bugs, not a stale install).Q1 —
compare --exclude-chains @fit:idsnever matchedparse_cohort_excludesplit the token but kept the@sigil as part of the fit name, so@ctl_rm.toml:4was compared againstctl_rm.tomland always missed ("no compared fit named '@ctl_rm.toml'"). The--help-documented@fit:idsform was unusable; only the barefit:idsform worked.Fix: strip the optional
@before matching, so@fit:idsandfit:idsbind the same fit. The prior unit tests passed vacuously — the model names were literally@a/@b, so no strip was ever exercised; retargeted them to real names and assert both forms resolve identically.Q2 — mh-summary rename is a clean break (no back-compat)
The
pmmh_summary.json→mh_summary.jsonrename (#415) is a deliberate breaking change.read_convergencewas inconsistently still falling back across sibling summary names, whilePmmhStageResult::loadwas not. Removed the fallback so both read only the method's own<algorithm>_summary.json.Consequence (documented breaking change): a pre-rename mh fit (only
pmmh_summary.jsonon disk) isNotAssessedand must be re-run — it is never silently read under the old name. Fits produced by a build ≥ #415 are unaffected.Tests
parse_cohort_exclude_forms/_rejects_bad_forms— real names; both@fit:idsandfit:idsforms, error messages name the stripped fit.read_convergence_clean_break_no_pmmh_fallback— an old mh fit (pmmh_summary.jsononly) isNotAssessed.read_convergence_finds_mh_summary_for_mh_method— mh reads its ownmh_summary.json.Installed locally at this commit ahead of merge to unblock the garki agent.