Repository navigation
Fix two claims broken on main (pick_ingested_probe normalized arm; stage0 layout overlap) - #12406
Merged
Merged
Conversation
…age0 layout overlap (1) pick_probe_normalized_then_arm_has_magnitude_child_holds handed an Outcome<NormalizedTree> to the shared Outcome<Node> reader (also used by the resolved-arm claim). #8256 sealed normalize's result as NormalizedTree and updated the producer helper but not this call site; the call site now projects .root. The reader stays as is. (2) scaffold_generated_artifact_layout_overlap_repo_paths_is_exactly_v1_interpreter_dispatch compared the overlap against a two-path literal copied from the tree. The layout did not regress: the wet-actuator family grew (#9981, #10142, #10829, #12183) and every member is a registry artifact. The claim now joins the overlap against wet_actuator_generated_stage0_files by identity (DESIGN §5: a tree measurement is not an oracle). Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…malized claims, enrolled warm The normalized claim newly executes and exceeded the new-witness step budget (127,367 of 72,300), with its cost dominated by normalize over the probe source. The resolved claim already ran normalize on the same source, so the shared demand is one nullary producer (pick_probe_normalized_module), enrolled in floor_cross_claim_pure_producers_warm. Both claims still run the real route. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…e' is ambiguous with extdeps.tools.resolve) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…mporting it An import turns off the file's bare channel, which then refused pick_ingested_find_arrow_in_module (UnimportedBareProvider). Qualifying the one ambiguous call keeps the file import-free. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
github-merge-queue
Bot
removed this pull request from the merge queue due to a conflict with the base branch
Sep 27, 2026
# Conflicts: # src/v2/workflow/floor_pure_producer_share.dag
github-merge-queue
Bot
removed this pull request from the merge queue due to a conflict with the base branch
Sep 28, 2026
# Conflicts: # src/v2/workflow/floor_pure_producer_share.dag
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.
(1) pick_probe_normalized_then_arm_has_magnitude_child_holds handed an
Outcome to the shared Outcome reader (also used by the
resolved-arm claim). #8256 sealed normalize's result as NormalizedTree and
updated the producer helper but not this call site; the call site now projects
.root. The reader stays as is.
(2) scaffold_generated_artifact_layout_overlap_repo_paths_is_exactly_v1_interpreter_dispatch
compared the overlap against a two-path literal copied from the tree. The
layout did not regress: the wet-actuator family grew (#9981, #10142, #10829,
#12183) and every member is a registry artifact. The claim now joins the overlap
against wet_actuator_generated_stage0_files by identity (DESIGN §5: a tree
measurement is not an oracle).
Evidence still owed: both modules need to exit 0 with every claim PASS under
gunbc run --claim-run. The BuildBuddy runner refuses with HostBudgetUnreadable (no cgroup limit), so the srv1 run has been requested from neat-boar-16.🤖 Generated with Claude Code