The key take away from sgold’s message is that a simple test case makes a big difference.
…and from experience, 9 times out of 10, things will work fine in the simple test case and the issue-reporter has something to look into on their end.
Without a test case, no matter how hard someone looks into it, there will be this nagging thought in the back of their mind “Maybe it’s just a problem with ndebruyn’s editor setup”… which even if untrue will prevent the brain from really considering all solutions.
…and it might be true.