Hello! This commit needs way more text than that.
This is just too misterious now. What asserts fails? What has the test case to do with the fix? Why modified_non_trans_table even matters without replication? Why it's specific to versioning and partitioning? Where trans_safe becomes false?
You are right. To get the answer I had to run the debugger and read the discussion with midenok. Though I think one can always GET jira.mariadb.org/browse/MDEV-17891 to see some details about what was the exact assertion, what was the stack trace and so on. I have updated the comment and have added one more test case from the ticket. See bb-10.3-nikita, or simply https://github.com/MariaDB/server/commit/688b9de0fedfcf13c4d33e7aa98e03e6e37... -- Yours truly, Nikita Malyavin