chore: version line to 1.1, comment cleanup, and mission.md request sources - #64
Merged
Conversation
This was referenced Aug 17, 2026
…ect reference # Conflicts: # .claude/mission.md
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Replaces #61, #62 and #63, which are closed in favour of this one. No library code changes — three project-level corrections, one commit each.
1. Version line 1.0 → 1.1
MAJOR_MINOR: '1.1'inbuild.yml. No1.1.*tag exists, so the counter starts clean and the next master release publishes 1.1.0; the1.0.xtags are untouched. Verified on CI:Computed version: 1.1.0 (previous: ).The reason is the registration-semantics change in #60, more than the dependency removal.
AddCacheregistrations no longer accumulate process-wide across service collections, so a host that calledAddCacheon one collection and expected the types on another silently loses that — a behavioural change with no compile error to warn anyone. In a three-or-more-call sequence, a call that doesn't mention a type now inherits the most recent value rather than the first-ever one.The FluentAssertions removal is the weaker half of the argument: it can break a consumer who was unknowingly compiling against the transitive reference, but #56 shows consumers were fighting the dependency (
NU1605from pinning an older version), not relying on it.This is the one change here with externally visible effect — it sets the version every consumer sees.
2. Remove multi-line comment blocks added in #60
Coding Guidelines allow single-line inline comments only, no multi-line blocks. Two of mine broke that:
build.yml,Test with coverage— I appended two lines about the Microsoft.Testing.Platform migration to an existing one-line comment, making a three-line block. Removed; the flags are self-describing andglobal.jsondeclares the runner, so it was history rather than something the file needs. The original line about skipping Integration and TimeCritical tests stays.AddCacheConcurrencyTests.cs— the note onMarkerTypewas two lines, now one. This one earns its single line: the nested-generic trick genuinely isn't readable from the code.Not touched:
build.ymlhas three pre-existing multi-line blocks (Compute version,Resolve build version,Push to NuGet). They predate this work, so removing them is a separate decision.3.
.claude/mission.mdrequest sourcesTwo corrections to agent instructions — no build or runtime effect.
Incoming requestsused a literalc:\Users\...path while every other reference in the file already used$DOC_ROOT. A hard-coded path resolves on one machine and silently finds nothing elsewhere — indistinguishable from "nothing was pending", which is the failure mode that matters for a startup check.Verification
Release build clean. CI-filtered suite: 467 tests, 465 passed, 2 skipped.
Note the suite contains two known timing-flaky tests, tracked as backlog items 9 and 13. Item 13 (
TimeToLiveCacheTests.DropEvenIfUsed) carries noTimeCriticaltrait, so unlike item 9 it is not filtered out of CI and can redden this PR at random. If a run fails on that test, it is not this change.