Added IPAEndRingStops to track stopped muons in the MuBeamResampler.fcl

[Etho-b02]

• Dependent on Offline/Mu2eG4/geom/protonAbsorber_cylindrical_v05.txt
• geom_run1_a.txt shuold move to v05 to reflect correct IPAEndRing dimensions
• Currently EndRingPath is disabled from trigger and out paths in the fcl file
  • To enable paths add IPAEndRingStopPath to the end of physics.trigger_paths and IPAEndRingStopOutput to physics.outPath

[FNALbuild]

Hi @Etho-b02,
You have proposed changes to files in these packages:

• JobConfig

which require these tests: build.

@Mu2e/write, @Mu2e/fnalbuild-users have access to CI actions on main.

📝 The author of this pull request is not a member of the Mu2e github organisation.

About FNALbuild. Code review on Mu2e/Offline.

[AndrewEdmonds11]

Looks good to me. I'll leave it up to the Production team to decide whether the extra IPAEndRingStops output wants to be on by default

[AndrewEdmonds11]

I would change this back to @nil since we just hard-coded it to 0 for our tests

[oksuzian]

PR Review Summary — #562 "Add IPAEndRingStops to track stopped muons in MuBeamResample.fcl"

Reviewed at head f57078ab

Decision

• 🔴 request changes

Scope understood

• JobConfig/pileup/prolog.fcl: adds IPAEndRingMuonFinder (StoppedParticlesFinder, stoppingMaterial: "IPAPolystyrene"), IPAEndRingStopFilter (FilterG4Out) and compressPVIPAEndRingStops, mirroring the existing IPA chain.
• JobConfig/pileup/MuBeamResampler.fcl: adds IPAEndRingStopPath to trigger_paths, IPAEndRingStopOutput (a 6th RootOutput) to outPath.
• Also in JobConfig/pileup/prolog.fcl and not described by the PR title/body: beamResampler and neutralsResampler have MaxEventsToSkip: @nil replaced by MaxEventsToSkip: 0, with the @nil lines left commented out.

Carry-forward from the existing review

• @AndrewEdmonds11 (APPROVED, 2026-08-03) left an inline request on JobConfig/pileup/prolog.fcl:262: "I would change this back to @nil since we just hard-coded it to 0 for our tests." → UNADDRESSED at f57078ab (the PR still has a single commit, authored before the comment). Carried forward as finding 1, and it applies to both occurrences, not just line 262.
• @AndrewEdmonds11's approval note — "I'll leave it up to the Production team to decide whether the extra IPAEndRingStops output wants to be on by default" — is still an open question. Carried forward as finding 3, with evidence on why the default matters.

Findings

1. 🔴 [S0] Leftover test hack: MaxEventsToSkip: @nil → 0 in the shared pileup prolog (carried over, unaddressed)

   • Evidence: JobConfig/pileup/prolog.fcl:235-236 (beamResampler) and :261-262 (neutralsResampler) at f57078ab:

     //MaxEventsToSkip: @nil
     MaxEventsToSkip: 0
     

     Pileup.filters.beamResampler / .neutralsResampler are consumed by MuBeamResampler.fcl, EleBeamResampler.fcl, PiBeamResampler.fcl and NeutralsResampler.fcl, so this is not scoped to the IPA end-ring work at all.
   • Impact: @nil is the Mu2e "you must supply this" sentinel — note fileNames: @nil two lines above is deliberately left as @nil in the same block. Offline/EventMixing/src/ResamplingMixer_module.cc declares fhicl::Atom<unsigned> maxEventsToSkip{..., 0u} and eventsToSkip() draws std::uniform_int_distribution<size_t> uniform(0, maxEventsToSkip_) once, to pick the starting offset into the first secondary file. Replacing the sentinel with a legal value 0 converts a hard, loud failure ("you forgot the offset") into a silent one: every job that does not carry an explicit override starts at event 0 of its resampling file, i.e. all jobs resample the same events. No released dataset is affected today — the prodtools job-definition path appends physics.filters.<resampler>.mu2e.MaxEventsToSkip: <N> after the entry's fcl_overrides, and Validation/nightly/pileup_09.fcl sets it explicitly (8934) — which is precisely why the nightly would not catch a regression here either. The value of the @nil is the guard, and this removes it.
   • Suggested fix: restore MaxEventsToSkip: @nil in both blocks and delete the commented-out lines rather than leaving them in (no commented-out code). If a hard-coded 0 is wanted for local testing, put it in the test's own top-level fcl (as Scripts/ListPileupProducts.sh and Validation/nightly/pileup_0*.fcl already do), not in the shared prolog.

2. 🟠 [S1] stoppingMaterial: "IPAPolystyrene" matches nothing in any released geometry, and nothing in the geometries production pins — the new stream will select zero events

   • Evidence: StoppedParticlesFinder::materialAccepted() (Offline/CommonMC/src/StoppedParticlesFinder_module.cc) is an exact string compare material == stoppingMaterial_; a non-matching name is silently zero-acceptance, no exception. Tracing the name:
     • IPAPolystyrene is defined in Offline/Mu2eG4/src/ConstructMaterials.cc:1337 and consumed by exactly one geometry file, Offline/Mu2eG4/geom/protonAbsorber_cylindrical_v05.txt (protonabsorber.ipa.endRingMaterial = "IPAPolystyrene"). Both are on Offline main only.
     • protonAbsorber_cylindrical_v05.txt is included only by geom_run1.txt and geom_run1_a_stickman.txt (and therefore geom_common.txt) — on main.
     • Every currently released SimJob backing release lacks it entirely: protonAbsorber_cylindrical_v05.txt is absent from /cvmfs/mu2e.opensciencegrid.org/Musings/Offline/{v13_30_00,v13_31_00,v13_32_10,v13_33_00} (v13_33_00 backs the newest musing, SimJob/MDC2025au). In those releases geom_common.txt → geom_run1_a_stickman.txt → protonAbsorber_cylindrical_v04.txt, i.e. protonabsorber.ipa.endRingMaterial = "CorrugatedPolypropylene".
     • Even once main is released, the production MuBeamResampler entries override services.GeometryService.inputFile to Offline/Mu2eG4/geom/geom_run1_a.txt (MDC2025 / MDC2030 campaigns) or geom_run1_b_v01.txt / geom_run1_b_v40.txt (Run1B), and all of those resolve to protonAbsorber_cylindrical_v04.txt at line 66 — CorrugatedPolypropylene, not IPAPolystyrene.
   • Impact: IPAEndRingMuonFinder produces an empty SimParticlePtrCollection every event, IPAEndRingStopFilter rejects, IPAEndRingStopPath never passes, and IPAEndRingStopOutput writes an events-free sim.*.IPAEndRingStops.*.art for every job. The failure is completely silent — no exception, no non-zero exit — so it would only surface as an empty dataset after a full campaign.
   • Suggested fix: either (a) use the material actually configured for the end rings in the geometry this is meant to run against — CorrugatedPolypropylene for the _v04-based geom_run1_* family — or (b) state in the PR body which Offline version + geometry file this requires (protonAbsorber_cylindrical_v05.txt, i.e. geom_common.txt/geom_run1.txt on main) and confirm the production entries will be repointed. Please also post the StoppedParticlesFinder end-of-job line (accepted = N) from a run against the intended geometry as validation evidence — a non-zero accepted is the one-line proof this finding is closed.

3. 🟡 [S2] The 6th output stream is unconditional in a shared production fcl (@AndrewEdmonds11's open question)

   • Evidence: MuBeamResampler.fcl:41-42 puts IPAEndRingStopPath in trigger_paths and IPAEndRingStopOutput in outPath unconditionally; MuBeamResampler.fcl:74-81 adds the RootOutput. This fcl is the production beam-flash workhorse (MDC2025, MDC2030, Run1B all drive it), and each job now writes one more art file.
   • Impact: every future MuBeamFlash job in every campaign gains a 6th output that the production tooling will register in SAM. With finding 2 unresolved, those are O(10^4) events-free files per campaign. There is also no downstream consumer yet — nothing cats or selects IPAEndRingStops the way IPAStopsCat / Scripts/ListPileupProducts.sh handle IPAStops — so the dataset is write-only until a follow-up lands.
   • Suggested fix: either confirm with the Production team that the stream should be on by default and add the matching downstream entry, or keep IPAEndRingStopPath / IPAEndRingStopOutput out of the default trigger_paths / outPath and let a campaign entry opt in via fcl_overrides. JobConfig/pileup/IPAStopSelector.fcl is the precedent for a separately-driven IPA stream.

4. ⚪ [S3] Formatting of the new output block

   • Evidence: MuBeamResampler.fcl:74-81 — module_type : / SelectEvents : / fileName : use a space before the colon and the last three lines are indented 5 spaces, where the four sibling output blocks in the same file use module_type: and consistent 4-space bodies.
   • Impact: none functionally; it just reads as a different hand in an otherwise uniform file.
   • Suggested fix: match the surrounding IPAStopOutput block verbatim.

Verified — no action needed

• 🟢 Producer/filter/compressor wiring is internally consistent and mirrors the IPA chain exactly: IPAEndRingMuonFinder → IPAEndRingStopFilter.mainSPPtrInputs → compressPVIPAEndRingStops.particleInputs: ["IPAEndRingStopFilter"] → IPAEndRingStopOutput.SelectEvents: [IPAEndRingStopPath]. Every label referenced by the new path is defined in the new prolog blocks.
• 🟢 No duplicated simulation cost from the new path: IPAEndRingStopPath shares @sequence::Pileup.beamResamplerSequence and @sequence::Common.g4Sequence with the five existing paths, and art executes each module once per event regardless of how many paths name it.
• 🟢 @sequence::Pileup.SimKeptProducts used by the new output exists (JobConfig/pileup/prolog.fcl:356) and is the same sequence the sibling stop outputs use.
• 🟢 Build systems: checked both. Mu2e/Production has no CMakeLists.txt anywhere in the tree and exactly one SConscript (JobConfig/ensemble/SConscript, unrelated to fcl); pileup fcl is resolved at run time via FHICL_FILE_PATH, so there is no build file to mirror for this change and no configure_file/install list to update.
• 🟢 Whitespace-only deletions of the two blank lines at prolog.fcl (before compressPVIPAStops and before IPAStopFilter) are cosmetic and harmless.

Validation check

• Build/tests run: none by me; no validation evidence is given in the PR body either. FNALbuild lists build as the required test for JobConfig changes.
• Config contract check: partial. FHiCL structure and module/label wiring are coherent (finding "Verified" above); the stoppingMaterial value does not resolve against any released geometry (finding 2); MaxEventsToSkip no longer enforces its supply contract (finding 1).
• Cross-repo consistency: needs follow-up. The change depends on Offline protonAbsorber_cylindrical_v05.txt / IPAPolystyrene, which exist on Offline main but in none of the SimJob backing releases up to v13_33_00, and which the production-pinned geom_run1_* geometries do not include.

Residual risk

• A silently-empty IPAEndRingStops dataset produced at campaign scale before anyone notices (finding 2 has no loud failure mode).
• A future resampler entry that omits the MaxEventsToSkip override silently resampling identical events across all its jobs (finding 1).
• One extra art file per MuBeamFlash job across MDC2025 / MDC2030 / Run1B storage and SAM bookkeeping (finding 3).

Author follow-ups

1. Revert MaxEventsToSkip to @nil in both beamResampler and neutralsResampler, and delete the commented-out //MaxEventsToSkip: @nil lines (finding 1 — already requested inline by @AndrewEdmonds11 and still open).
2. Confirm the intended stopping material: CorrugatedPolypropylene (what protonAbsorber_cylindrical_v04.txt, and hence every production geometry today, sets for the IPA end rings) vs IPAPolystyrene (only in protonAbsorber_cylindrical_v05.txt on Offline main). Fix the value or document the Offline/geometry requirement in the PR body (finding 2).
3. Post validation evidence: the StoppedParticlesFinder end-of-job summary for IPAEndRingMuonFinder (accepted = ...) from a short MuBeamResampler run against the geometry you intend, plus the resulting file's event count.
4. Decide with the Production team whether the 6th output should be on by default, and if so, say which downstream Cat/selector entry will consume IPAEndRingStops (finding 3).
5. Best practice reminder: please add a meaningful PR description covering intent, scope, and validation evidence, and keep the PR to a single topic — the MaxEventsToSkip edit is unrelated to the IPA end-ring stops. (Also, the title says MuBeamResample.fcl; the file is MuBeamResampler.fcl.)

[oksuzian]

@FNALbuild run build test

[FNALbuild]

⌛ The following tests have been triggered for ab80665: build (Build queue - API unavailable)

[oksuzian]

PR Review Summary — re-review of #562 "Add IPAEndRingStops to track stopped muons in MuBeamResample.fcl"

Reviewed at head ab80665d (2026-08-10). Second pass. Prior reviews: @AndrewEdmonds11 APPROVED at f57078ab with one inline request, and mine 🔴 at f57078ab (4887595941, 4 findings). The delta is one commit, ab80665d "Adressed PR comments:", touching both files. Every prior finding is verified in the code at this head, not read off the commit message.

Decision

• 🔴 Request changes — for one thing only, and it is narrow. Three of the four findings are properly addressed, and the design decision you made on the output stream is the right one. But the MaxEventsToSkip fix landed on only one of the two resamplers, and the one still holding the hard-coded 0 is beamResampler — the one this PR's own MuBeamResampler.fcl actually runs. One line, and I flip to approve.

Scope understood (unchanged, plus this pass's changes)

• IPAEndRingMuonFinder / IPAEndRingStopFilter / compressPVIPAEndRingStops are added to the shared pileup prolog, and IPAEndRingStopPath / IPAEndRingStopOutput to MuBeamResampler.fcl.
• New this pass: the path and output are deliberately kept out of trigger_paths and outPath; neutralsResampler's MaxEventsToSkip is restored to @nil; the commented-out //MaxEventsToSkip: @nil lines are deleted from both blocks; the output block's key : spacing is normalised.

---

Findings

1. 🔴 [S0] MaxEventsToSkip: 0 is still in beamResampler — the fix landed on neutralsResampler only (carried over, PARTIAL).

   • Evidence: JobConfig/pileup/prolog.fcl at ab80665d:

     228:    beamResampler: {
     235:        MaxEventsToSkip: 0          <-- still hard-coded
     253:    neutralsResampler: {
     260:        MaxEventsToSkip:  @nil      <-- restored
     

     main has @nil at both sites (:213, :238), so the 0 is introduced by this PR and survives at beamResampler. The commit message states "changed MaxEventsToSkip: nil in pileup/prolog : -neutralsResampler -beamResampler" — the diff (f57078ab...ab80665d) shows beamResampler only lost its commented-out line; the value was not restored.
   • Impact: this is the more consequential of the two. Pileup.beamResamplerSequence : [genCounter, protonTimeOffset, beamResampler] (prolog.fcl:370) is what MuBeamResampler.fcl runs on every one of its paths, and EleBeamResampler.fcl / PiBeamResampler.fcl use it too. As in the first review: @nil is the "you must supply this" sentinel, and replacing it with the legal value 0 turns a loud startup failure into a silent one — any job without an explicit override starts at event 0 of its resampling file, so all such jobs resample the same events. Nothing breaks today (prodtools appends the override, and Validation/nightly/pileup_09.fcl sets it explicitly), which is exactly why nothing would catch the regression either.
   • Suggested fix: MaxEventsToSkip: @nil at :235, matching what you already did at :260.
   • Credit where due: @AndrewEdmonds11's inline request pointed at line 262, i.e. the neutralsResampler occurrence, and that one is fixed. My finding extended it to both blocks; this is the half that remains.

2. 🟡 [S2] stoppingMaterial: "IPAPolystyrene" still will not match under any geometry production actually runs — now dormant rather than harmful (carried over, downgraded from 🟠 S1, PARTIAL).

   • What changed: with IPAEndRingStopPath out of trigger_paths, IPAEndRingMuonFinder is not scheduled, so the zero-acceptance behaviour cannot produce empty datasets today. The severity drop reflects that, not a change in the underlying facts.
   • What you answered: the commit message states this needs protonAbsorber_cylindrical_v05.txt and the latest geometry files. That is correct and it is the option-(b) answer I asked for — but it is in the commit message, not the PR body, and one gap remains open underneath it.
   • Evidence, re-verified at Offline main today:
     • IPAPolystyrene end rings come only from protonAbsorber_cylindrical_v05.txt, included by exactly two files: geom_run1.txt:66 and geom_run1_a_stickman.txt (hence geom_common.txt, the default).
     • geom_run1_a.txt:66 is still protonAbsorber_cylindrical_v04.txt — CorrugatedPolypropylene. And geom_run1_b_v01.txt:4 simply #includes geom_run1_a.txt, so Run1B inherits v04 too.
     • Every production entry that drives this fcl overrides the geometry to one of those: services.GeometryService.inputFile : Offline/Mu2eG4/geom/geom_run1_a.txt for MDC2025 and MDC2030 MuBeamFlash, and geom_run1_b_v01.txt for Run1B.
     • No released Offline musing carries v05 at all: absent from v13_30_00, v13_31_00, v13_32_10 and v13_33_00 (the newest), where geom_run1_a_stickman.txt still includes v04.
   • Impact: the dependency on unreleased Offline main is real but tracked. The part that is not addressed by "wait for the Offline release" is that geom_run1_a.txt — the file production actually pins — does not pick up v05 even on main. Whoever enables this path later inherits a silently zero-acceptance stream, because StoppedParticlesFinder::materialAccepted() is an exact string compare with no diagnostic.
   • Suggested fix: say in the PR body which Offline version and geometry file this requires, and state explicitly whether geom_run1_a.txt is expected to move to v05 or whether the production entries will be repointed at geom_run1.txt. When you enable the path, post the StoppedParticlesFinder end-of-job accepted = N for IPAEndRingMuonFinder — a non-zero count is the one-line proof this is closed.

3. ⚪ [S3] Indentation of the new output block is still off by one (carried over, PARTIAL).

   • Evidence: MuBeamResampler.fcl:74-81 — the key : → key: spacing is fixed, but the closing ] and the SelectEvents / fileName lines are still indented 5 spaces where IPAStopOutput (:65-72) and the other four blocks use 4.

4. ⚪ [S3] Nothing documents how to turn the stream on, and the PR body is still one line.

   • Enabling it means overriding physics.trigger_paths and physics.outPath, which are sequences — an fcl_overrides entry has to restate the whole list, not append to it. A one-line comment above IPAEndRingStopPath giving the two override lines would save the next person the reverse-engineering.
   • The PR body still does not mention the Offline/geometry dependency, the validation status, or that the stream ships disabled. The title still says MuBeamResample.fcl; the file is MuBeamResampler.fcl.

---

Carry-forward accounting (vs review 4887595941 at f57078ab)

1. 🔴 [was S0] MaxEventsToSkip — PARTIAL, see finding 1. neutralsResampler fixed and both commented-out lines removed; beamResampler value not restored.
2. 🟡 [was S1] stoppingMaterial — PARTIAL, downgraded, see finding 2. Documented in the commit message and defused by disabling the path; the geom_run1_a.txt gap is still open.
3. 🟢 [was S2] The 6th output stream is no longer on by default — FIXED in ab80665d, verified, and this is the right call. trigger_paths and outPath at :41-42 are back to the five original entries. This also resolves @AndrewEdmonds11's open question from his approval note ("I'll leave it up to the Production team to decide whether the extra IPAEndRingStops output wants to be on by default") in the conservative direction: the machinery is staged, the decision stays with the Production team, and no campaign gains a file it did not ask for.
4. 🟡 [was S3] Output-block formatting — PARTIAL, see finding 3.

---

Verified 🟢 — no action needed

• 🟢 Leaving IPAEndRingStopPath and IPAEndRingStopOutput defined but unscheduled is safe — art ignores both silently. Tested, not assumed. I ran a minimal job under Offline v13_33_00 (art v3_15_00) with a path defined in physics but absent from trigger_paths, and a RootOutput defined in outputs but absent from outPath, including a SelectEvents pointing at the unscheduled path. Result: Art has completed and will exit with status 0, only the scheduled output's file was created, the unscheduled one was never opened, and no warning or error naming either was emitted. So the staged blocks are genuinely inert — they cost nothing until someone lists them.
  • Worth knowing for finding 4: because art never validates the unscheduled config, a typo in IPAEndRingStopPath or in SelectEvents will not surface until the day the stream is enabled.
• 🟢 The producer/filter/compressor wiring still mirrors the IPA chain exactly and every label the new path names is defined: IPAEndRingMuonFinder → IPAEndRingStopFilter.mainSPPtrInputs → compressPVIPAEndRingStops.particleInputs: ["IPAEndRingStopFilter"] → IPAEndRingStopOutput.SelectEvents: [IPAEndRingStopPath].
• 🟢 The new prolog blocks land in Pileup.producers / Pileup.filters, so they are now configured into every fcl including JobConfig/pileup/prolog.fcl — EleBeamResampler, NeutralsResampler, PiBeamResampler, MuStopPileup and the selectors. Unscheduled everywhere, so harmless; called out because the blast radius of that prolog is wider than this PR's title suggests.
• 🟢 @sequence::Pileup.SimKeptProducts used by the new output exists and is the same sequence the sibling stop outputs use.
• 🟢 No build file to update: Mu2e/Production has no CMakeLists.txt in the tree and one unrelated SConscript; pileup fcl resolves at run time via FHICL_FILE_PATH.
• 🟢 neutralsResampler's restored @nil is spelled exactly as main has it, two-space form included.

Validation check

• Build/tests run: mu2e/buildtest had no result at ab80665d ("This test has not been triggered yet"), so I triggered one — this is the required test for JobConfig changes, and FNALbuild notes the author is not a Mu2e org member, so it needs a member to start it. Result pending at ab80665d as of this review; please check it before merging. Separately, I ran the unscheduled-path/output test described above myself.
• Config contract check: partial — FHiCL structure and label wiring are coherent and the disabled stream is verified inert; MaxEventsToSkip still does not enforce its supply contract at beamResampler (finding 1).
• Cross-repo consistency: needs follow-up — depends on Offline main (protonAbsorber_cylindrical_v05.txt), which is in no released musing through v13_33_00, and geom_run1_a.txt — what production actually pins — does not include it even on main (finding 2).

Residual risk

• A future resampler entry that omits the MaxEventsToSkip override silently resampling identical events across all its jobs (finding 1).
• The IPA end-ring stream is now staged but unvalidated and unvalidatable by art until it is switched on; the first person to enable it inherits both the material mismatch and any typo in the unscheduled config.

Author follow-ups

1. MaxEventsToSkip: @nil at prolog.fcl:235 (beamResampler) — the one blocking item (finding 1).
2. Put the Offline/geometry dependency in the PR body, and say whether geom_run1_a.txt moves to protonAbsorber_cylindrical_v05.txt or the production entries get repointed (finding 2).
3. Optional: 4-space indent on the new output block (finding 3), and a comment giving the two overrides needed to enable the stream (finding 4).
4. Check the mu2e/buildtest result at ab80665d before merging — I triggered it, but it had not reported when this review was written.

[FNALbuild]

☀️ The build tests passed at ab80665.

Test	Result	Details
test with	✅	Command did not list any other PRs to include
merge	✅	Merged ab80665 at 20c93ed
build (prof)	✅	Log file. Build time: 04 min 20 sec
ceSimReco	✅	Log file.
g4test_03MT	✅	Log file.
transportOnly	✅	Log file.
POT	✅	Log file.
g4study	✅	Log file.
cosmicSimReco	✅	Log file.
cosmicOffSpill	✅	Log file.
ceSteps	✅	Log file.
ceDigi	✅	Log file.
muDauSteps	✅	Log file.
ceMix	✅	Log file.
rootOverlaps	✅	Log file.
g4surfaceCheck	✅	Log file.
FIXME, TODO	✅	TODO (0) FIXME (0) in 0 files
clang-tidy	✅	0 errors 0 warnings

N.B. These results were obtained from a build of this Pull Request at ab80665 after being merged into the base branch at 20c93ed.

For more information, please check the job page here.
Build artifacts are deleted after 5 days. If this is not desired, select Keep this build forever on the job page.

[oksuzian]

@FNALbuild run build test

[FNALbuild]

⌛ The following tests have been triggered for ce9614f: build (Build queue - API unavailable)

[FNALbuild]

☀️ The build tests passed at ce9614f.

Test	Result	Details
test with	✅	Command did not list any other PRs to include
merge	✅	Merged ce9614f at 0c81d36
build (prof)	✅	Log file. Build time: 04 min 21 sec
ceSimReco	✅	Log file.
g4test_03MT	✅	Log file.
transportOnly	✅	Log file.
POT	✅	Log file.
g4study	✅	Log file.
cosmicSimReco	✅	Log file.
cosmicOffSpill	✅	Log file.
ceSteps	✅	Log file.
ceDigi	✅	Log file.
muDauSteps	✅	Log file.
ceMix	✅	Log file.
rootOverlaps	✅	Log file.
g4surfaceCheck	✅	Log file.
FIXME, TODO	✅	TODO (0) FIXME (0) in 0 files
clang-tidy	✅	0 errors 0 warnings

N.B. These results were obtained from a build of this Pull Request at ce9614f after being merged into the base branch at 0c81d36.

For more information, please check the job page here.
Build artifacts are deleted after 5 days. If this is not desired, select Keep this build forever on the job page.