[io] test for renaming with unversioned classes
This Pull request:
Changes or fixes:
Test for https://github.com/root-project/root/issues/14491 Closes https://github.com/root-project/root/issues/14491
Checklist:
- [x] tested changes locally
- [ ] updated the docs (if necessary)
Test Results
20 files 20 suites 3d 18h 54m 10s ⏱️ 3 789 tests 3 788 ✅ 0 💤 1 ❌ 73 927 runs 73 925 ✅ 0 💤 2 ❌
For more details on these failures, see this check.
Results for commit adc0ac09.
:recycle: This comment has been updated with latest results.
Instead providing root file I would add writedata.C macro and execute it. So one can clearly see that kind of data is produced.
Instead providing root file I would add writedata.C macro and execute it. So one can clearly see that kind of data is produced.
Thanks, good idea. Could you verify that I set the FIXTURES flag correctly? Since I am not used to those yet.
Fixtures configured correctly. I let @pcanal approve PR.
It seems it only fails on Windows. No idea why...
It seems it only fails on Windows. No idea why...
Looks like difference in file content. May be some variables not initialized when file is created.
@bellenot do you have by chance an educated guess of what could be failing in Windows? such as an integer type having different widths on each platform or so.
@ferdymercury I don't see anything obvious, but I'll check... The test output & diff is:
-- BEGIN TEST OUTPUT --
Processing C:/ROOT-CI/src/roottest/root/io/evolution/versions/ROOT14491readdata.cxx+...
***********************************
* Row * Instance * seg_split *
***********************************
dups 0
copy 0
dups 1
copy 1
* 0 * 0 * 0 *
***********************************
dups 0
copy 0
dups 1
copy 1
***********************************
* Row * Instance * seg_unspl *
***********************************
dups 0
copy 0
dups 1
copy 2
dups 0
copy 0
dups 1
copy 3
* 0 * 0 * 0 *
***********************************
(int) 0
-- END TEST OUTPUT --
-- BEGIN OUTDIFF OUTPUT --
--- C:/ROOT-CI/src/roottest/root/io/evolution/versions/ROOT14491readdata.ref Mon Dec 8 12:12:20 2025
+++ C:/ROOT-CI/build/roottest/root/io/evolution/versions/ROOT14491readdata.log Mon Dec 8 13:54:41 2025
@@ -23,6 +23,6 @@
copy 0
dups 1
copy 3
-* 0 * 0 * 1 *
+* 0 * 0 * 0 *
***********************************
(int) 0
-- END OUTDIFF OUTPUT --
So maybe @pcanal has an idea, before I start debugging...