Releases: LDMX-Software/ldmx-sw
v4.1.3 : Bump ACTS and Compiler Static Analysis Cleanup
After a truly herculean effort by @tvami , we are now compiling ldmx-sw with a lot more strict requirements on the compiler. This will help insure us against future bugs and enables us to use common and well-tested tools like clang-tidy during development. Additionally, @tvami update the CI to use just+denv in more workflows and introduced a new workflow to automatically apply clang-format to non-trunk branches.
On top of this work, @bloodyyugo updated Tracking to use ACTS v36.0.0 and @tomeichlersmith updated the logging to allow setting the log-level by processor.
⚡ Highlights
- Add fiducial & BDT score cut options to EcalPreselectionSkimmer by @tvami in #1456
- Update tracking to use Acts v36.0 by @bloodyyugo in #1454
- logging fix and the long-awaited setting log-level by processor by @tomeichlersmith in #1458
- Auto-commit clang format by @tvami in #1464
- Update tracking CI, add new CI action for LTO and modify other actions to use just by @tvami in #1462
🐛 Fixes
- Change tracking timing precision, fixes to DQM histo names by @tvami in #1448
- dereference boost.log's attribute pointer only if defined by @tomeichlersmith in #1463
- Add condition for not being on trunk when pushing in format CI by @tvami in #1466
🧹 Cleanup
- Misc code improvements: override, destructor, small mistakes by @tvami in #1447
- Trailing ldmx command refs by @tomeichlersmith in #1450
- Part 2 of static analysis code improvements by @tvami in #1453
- Part 3 of static analysis: clang-tidy by @tvami in #1460
🤖 Bot
- New Gold for Generated by release v4.1.2 by @github-actions in #1446
- Bump peter-evans/create-pull-request from 6 to 7 by @dependabot in #1451
Full Changelog: v4.1.2...v4.1.3
Submodule State
Contributors
Assets 2
v4.1.2 : Patches to Help with BDT Skim
While having a tagged version is helpful for a large scale BDT skim, this release is also necessary to generate new golden histograms introduced by #1440 .
⚡ Highlights
- Add validation installation to the just commands and rename the dir by @tvami in #1441
- Modify EcalVeto skim settings to keep non-fiducial and be more inclusive by @tvami in #1444
- Add more features to ECAL DQM & validation module by @tvami in #1440
🐛 Fixes
- check for corrupted zombie files in ldmx sw by @tomeichlersmith in #1443
🧹 Cleanup
- Fix narrowing conversion in QIEDecoder by @tvami in #1426
- Move logger open into Process constructor by @tvami in #1445
🤖 Robot
- New Gold for Generated by release v4.1.1 by @github-actions in #1437
Full Changelog: v4.1.1...v4.1.2
Contributors
Assets 2
v4.1.1 : No-Cals Detector and UB Patches
The main trigger for this patch release is #1432 which altered the golden histograms slightly by fixing the definitions of epAng and epSep (which depend on epDot). We also have some other patch PRs and a no-cals detector for tracking studies included in this release.
⚡ Highlights
- Add TaggerHitFilter and no-cals detectors by @bloodyyugo in #1424
🐛 Patches
- Deprecate ECal cellxy.txt and use EcalGeometry instead by @tvami in #1425
- Introduce epdot in EcalVeto by @tvami in #1432
- Add missing defaults in CalorimeterHit.h by @tvami in #1434
- Fix UB in DetectorIDTest.cxx by @tvami in #1436
🤖 Robot
- New Gold for Generated by release v4.1.0 by @github-actions in #1430
Full Changelog: v4.1.0...v4.1.1
Contributors
Assets 2
v4.1.0: Busy Bee
After many patches, fixes, and new features relative to 4.0, we are cutting a new minor release to recognize the plethora of contributions!
Below are all of the updates relative to 4.0.0 most of which are included in the subsequent patch releases that were made due to the CI validation tests needing updates. In association with this release, the documentation website has received a pretty substantial rewrite as well.
⚡ Highlights
- Introduce variable for total events with no tries limit by @tvami in #1322
- Introduce DeepEcalProcessFilter by @tvami in #1289
- Added multi-electron tagger truth tracking capability by @EBerzin in #1374
- Introduction of SegMIP BDT + other improvement on ECAL veto by @tvami in #1383
- Introduce EcalPreselectionSkimmer by @tvami in #1389
- Update G4DarkBreM with configurable A' scaling and decay by @RobMina in #1386
- Including ratio subplot within differ when comparing by @tomeichlersmith in #1401
- just adopt a task runner by @tomeichlersmith in #1380
🐛 Fixes
- Fix 8gev detector name in GDML by @tvami in #1358
- Fix MIP tracking implementation issues by @tvami in #1356
- Enable StatBox in DQM histograms by @tvami in #1363
- add ability to link to additional libraries for
from_fileprocessors by @tomeichlersmith in #1366 - Make all truth tracking collections even with 0 SimHits by @bloodyyugo in #1372
- forgot to call
items()when looping over a dict by @tomeichlersmith in #1368 - dont recurse into submodules when checking the state by @tomeichlersmith in #1390
- copy in positron-deduced positions and directions from the TS scoring planes by @tomeichlersmith in #1396
- Fix deprecated functions in ONNX and boost + type conversion for ConfigurePythonTest by @tvami in #1393
- Update README for Validation by @tvami in #1397
- Fix underflow/overflow in EcalVeto's closest cell calc by @tvami in #1410
- Patching to ECAL lin-reg tracking by @tvami in #1407
🧹 Cleanup
- Python config cleanup and enable single-file standalone processors by @tomeichlersmith in #1310
- trim dead code in simulator failing to add parameters to the runheader by @tomeichlersmith in #1352
- Archive unused geometries by @tvami in #1357
- Make tracking digitization seed random & apply clang format by @tvami in #1359
- Clang format ldmx-sw by @tvami in #1362
- add comment about needing
git lfsby @tomeichlersmith in #1375 - Move validation configs to 8 GeV, intro reduced LDMX validation, add tracking to ecal pn by @tvami in #1385
- Bump up TruthTrack and Track versioning by @tvami in #1404
- Fix the precision of the tracking performance printouts by @tvami in #1408
- Move cout msg to logging in DetectorConstruction, ParallelWorld, TS DQM by @tvami in #1415
- Add kaon validation + fix tracking seeder settings in ecal validation by @tvami in #1421
- Bump up class versions for HcalHit, TestBeamHit, TrigScintHit by @tvami in #1429
🤖 Bot
- New Gold for Generated by release v4.0.0 by @github-actions in #1296
- New Gold for Generated by release v4.0.1 by @github-actions in #1360
- Bump actions/checkout from 3 to 4 by @dependabot in #1364
- Bump docker/build-push-action from 5 to 6 by @dependabot in #1367
- New Gold for Generated by release v4.0.4 by @github-actions in #1391
- New Gold for Generated by release v4.0.5 by @github-actions in #1399
New Contributors
- @bloodyyugo made their first contribution in #1372
- @EBerzin made their first contribution in #1374
- @RobMina made their first contribution in #1386
Full Changelog: v4.0.0...v4.1.0
Contributors
Assets 2
v4.0.5 : Visible Dark Photon Decays and More Cleanup
One of each type in this patch release! We have a new feature within the signal simulation that allows for more easily and realistically simulating visible dark photon decays. #1396 is the main culprit - moving the primary starting location has changed most of the histograms and so we need to re-generate the gold. #1393 also contributed by altering the logs a smidge.
⚡ Highlights
🐛 Fixes
- copy in positron-deduced positions and directions from the TS scoring planes by @tomeichlersmith in #1396
- Update README for Validation by @tvami in #1397
🧹 Cleanup
- Fix deprecated functions in ONNX and boost + type conversion for ConfigurePythonTest by @tvami in #1393
🤖 Bot
- New Gold for Generated by release v4.0.4 by @github-actions in #1391
New Contributors
Full Changelog: v4.0.4...v4.0.5
Contributors
Assets 2
v4.0.4 : Reduce Terminal Output of ecal_pn Validation Sample
Again, similar to v4.0.3, see v4.0.2 for actually helpful notes.
Only one commit separates this release from the last one and it is attempting to avoid the large file issues that arise from the wealth of printouts at debug level within the Tracking modules.
The error was seen when attempting to create a PR from the newly-generated golden histograms.
remote: error: Trace: 6a310c1456671eedcc7c0152b13d263de0924c62a64eb0ad3ace7ce2730ba02a
remote: error: See https://gh.io/lfs for more information.
remote: error: File .github/validation_samples/ecal_pn/gold.log is 635.06 MB; this exceeds GitHub's file size limit of 100.00 MB
remote: error: GH001: Large files detected. You may want to try Git Large File Storage - https://git-lfs.github.com/.
To https://github.com/LDMX-Software/ldmx-sw
! [remote rejected] new-gold-v4.0.3 -> new-gold-v4.0.3 (pre-receive hook declined)
error: failed to push some refs to 'https://github.com/LDMX-Software/ldmx-sw'
Error: The process '/usr/bin/git' failed with exit code 1
Instead of attempting to figure out git lfs, I am just requiring there to be less printouts.
Full Changelog: v4.0.3...v4.0.4
v4.0.3 : Patch Reduced LDMX Validation Sample Config
The main updates are still highlighted in the previous release, this is just necessary to get the new sample config running.
What's Changed
- dont recurse into submodules when checking the state by @tomeichlersmith in #1390
- Needed to patch the new reduced ldmx validation sample config with
import osin 640596b- The reason this isn't a PR is because I tried to sneak it in and re-tag v4.0.2 so we wouldn't need a whole different release. Unfortunately, the re-generation of golden histograms was not done, so I'm just giving up and creating a new release.
Full Changelog: v4.0.2...v4.0.3
Contributors
Assets 2
v4.0.2 : Prep ldmx-sw for BDT Skim
This patch released is focused on integrating improvements to the ECal BDT as well as ramp up support for multi-electron studies, 8 GeV testing, and preparation for larger-scale skimming. Specifically, #1385 and #1383 require that new golden histograms and logs are generated for the PR Validation test suite.
⚡ Highlights
- Added multi-electron tagger truth tracking capability by @EBerzin in #1374
- Introduction of SegMIP BDT + other improvement on ECAL veto by @tvami in #1383
- Move validation configs to 8 GeV, intro reduced LDMX validation, add tracking to ecal pn by @tvami in #1385
- Introduce EcalPreselectionSkimmer by @tvami in #1389
🐛 Bug Fixes
- add ability to link to additional libraries for
from_fileprocessors by @tomeichlersmith in #1366 - Enable StatBox in DQM histograms by @tvami in #1363
- Make all truth tracking collections even with 0 SimHits by @bloodyyugo in #1372
- forgot to call
items()when looping over a dict by @tomeichlersmith in #1368
🧹 Cleanup
- Make tracking digitization seed random & apply clang format by @tvami in #1359
- Clang format ldmx-sw by @tvami in #1362
- add comment about needing
git lfsby @tomeichlersmith in #1375
🤖 Bot
- New Gold for Generated by release v4.0.1 by @github-actions in #1360
- Bump actions/checkout from 3 to 4 by @dependabot in #1364
- Bump docker/build-push-action from 5 to 6 by @dependabot in #1367
New Contributors
- @bloodyyugo made their first contribution in #1372
- @EBerzin made their first contribution in #1374
Submodule State
- SimCore/G4DarkBreM v2.0.1
- Tracking/acts ff2f9761
- Trigger/HLS_arbitrary_Precision_Types 200a9aec
- Trigger/ruckus 42282f7b
- docs/doxygen.conf/doxygen-awesome-css v2.2.1
Full Changelog: v4.0.1...v4.0.2
Contributors
Assets 2
v4.0.1 : New Simulation Filter and Standalone Processors
⚡ Highlights
- Enable single-file standalone processors by @tomeichlersmith in #1310
- Request for total events with no limit on number of tries by @tvami in #1322
- Introduce DeepEcalProcessFilter by @tvami in #1289
🐛 Bug Fixes
- Fix MIP tracking implementation issues by @tvami in #1356
- Requires new gold set for PR Validations since it triggers changes to logs.
🧹 Cleanup
- Python config cleanup by @tomeichlersmith in #1310
- trim dead code in simulator failing to add parameters to the runheader by @tomeichlersmith in #1352
- Fix 8gev detector name in GDML by @tvami in #1358
- Archive unused geometries by @tvami in #1357
🤖 Bot Changes
- New Gold for Generated by release v4.0.0 by @github-actions in #1296
Full Changelog: v4.0.0...v4.0.1
Contributors
Assets 2
v4.0.0: No More Submodules*
No More Submodules1
This release marks the version that, while behaving identically to v3.4.0 (besides differences documented below), is free of the submodule shackles constraining development. If you have a branch of ldmx-sw (or one of the prior submodules) that you wish to update to this new version, please follow the instructions on the documentation website.
Updating
Due to the extremely technical nature of this change (and the fact that many of our submodules had submodules of their own), the easiest way to update to the newer version of ldmx-sw locally is to create a new clone.
# Step 1: Acknowledge that you have no local changes you want to update to the new structure
rm -rf ldmx-sw
# Step 2: Copy down the new version of ldmx-sw
git clone --recursive git@github.com:LDMX-Software/ldmx-sw.git
In all honesty, you may want to keep the old version of ldmx-sw around for some time. In this case, I would suggest removing the trunk branch so that you do not inadvertently get that repository into a broken state.
cd ldmx-sw-old
git branch -D trunk # will only work if you are not on the trunk branch
git fetch --tags
The updating instructions (and all future work) reference the tag v3.4.0 as the base off which to check your pre-transition changes.
What's Changed
- New Gold for Generated by release v3.4.0 by @github-actions in #1292
- These gold histograms were different from the prior release v3.3.6 due to #1245 updating the configurations to use 8GeV trigger thresholds.
- transition submodules from LDMX-Software into sub-directories of ldmx-sw
- These few merge commits (produced from thousands of plain commits) were pushed directly to
trunksince a normal merge/rebase was not possible within the context of such a technical transition.
- These few merge commits (produced from thousands of plain commits) were pushed directly to
- cleanup and test de-submodule-ification by @tomeichlersmith in #1293
- This PR discovered that the
Ecal/datafiles were inadvertently ignored during the transition. Thecellxy.txtandgabrielle.onnxfiles were reintroduced toEcal/dataafter the transition within this PR. The older files are left to the submodule to be archived since they apply to prior versions of ldmx-sw and the detector that are no longer in use.
- This PR discovered that the
Submodule Status
While this release is titled "No More Submodules" ™️ , we still technically have submodules. But they are all projects that are distinct from our work on LDMX!
./scripts/state
- SimCore/G4DarkBreM v2.0.1
- SimCore/G4DarkBreM/docs/doxygen-awesome-css v2.1.0
- Tracking/acts ff2f9761
- Tracking/acts/thirdparty/OpenDataDetector v3.0.0
- Trigger/HLS_arbitrary_Precision_Types 200a9aec
- Trigger/ruckus 42282f7b
- docs/doxygen.conf/doxygen-awesome-css v2.2.1
Full Changelog: v3.4.0...v4.0.0
-
There will still be submodules, but they will only be repositories that are a distinct project from LDMX. ↩