-
Notifications
You must be signed in to change notification settings - Fork 25
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
ENH: Bump for ITK v5.4rc01 #236
Conversation
Bump CI and package dependencies for ITK v5.4 release candidate 1. Expected to include update to C++17 build tools.
Upgraded ITK on the CI to version 5.4rc01, which was released by Matt McCormick on 28 July 2023: https://github.com/InsightSoftwareConsortium/ITK/releases/tag/v5.4rc01 Following ITKElastix pull request InsightSoftwareConsortium/ITKElastix#236 commit InsightSoftwareConsortium/ITKElastix@c711bbd "ENH: Bump for ITK v5.4rc01", Tom Birdsong, July 24, 2023. Follow-up to pull request #762 commit 353a095 "COMP: Upgrade ITK from v5.3rc04 to v5.3.0", August 25, 2022. ITK v5.4rc01 includes an upgrade from C++14 to C++17.
Upgraded ITK on the CI to version 5.4rc01, which was released by Matt McCormick on 28 July 2023: https://github.com/InsightSoftwareConsortium/ITK/releases/tag/v5.4rc01 Following ITKElastix pull request InsightSoftwareConsortium/ITKElastix#236 commit InsightSoftwareConsortium/ITKElastix@c711bbd "ENH: Bump for ITK v5.4rc01", Tom Birdsong, July 24, 2023. Follow-up to pull request #762 commit 353a095 "COMP: Upgrade ITK from v5.3rc04 to v5.3.0", August 25, 2022. ITK v5.4rc01 includes an upgrade from C++14 to C++17.
I'm just trying to upgrade ITK from v5.3.0 to v5.4rc01 at the SuperElastix/elastix CI as well: https://github.com/SuperElastix/elastix/tree/Upgrade-ITK-to-v5.4rc01 However, to my surprise, it appears to trigger some regression failures: From both Windows and Ubuntu, at https://dev.azure.com/kaspermarstal/Elastix/_build/results?buildId=4292&view=results
And from macOS, at https://my.cdash.org/viewTest.php?onlyfailed&buildid=2395262
Details from https://my.cdash.org/test/90114840 (elastix_run_3DCT_lung.Kappa.bspline.ASGD.001_COMPARE_LANDMARKS):
And from https://my.cdash.org/test/90114841 (elastix_run_3DCT_lung.Kappa.bspline.ASGD.001_COMPARE_OVERLAP):
I do not know if these differences are of any importance, I mean I don't know if these are really buggy, but they certainly suggest that the behavior has changed from ITK v5.3.0 to v5.4rc01. Please let me know if you have a clue about those changes! |
Update: I found that the behavior change between ITK v5.3.0 to v5.4rc01 was caused by pull request InsightSoftwareConsortium/ITK#3929 commit InsightSoftwareConsortium/ITK@4e46cb6 "STYLE: Default default-constructor of ImageRandom ConstIterator classes", merged on February 24, 2023 (My 😧 fault!) It should be fixed (restoring the v5.3.0 behavior) by InsightSoftwareConsortium/ITK#4183 "BUG: ImageRandomIteratorWithIndex should not assign data in constructor". I hope this fix can still be included with the next ITK release (candidate). |
Bump CI and package dependencies for ITK v5.4 release candidate 1.
Expected to include update to C++17 build tools.