https://bugs.documentfoundation.org/show_bug.cgi?id=140760
--- Comment #13 from Eike <[email protected]> --- (In reply to Heiko Tietze from comment #12) > Whether this is a bug is at least questionable. Read through the comments. That is exactly what I did and I realized that the discussion there lacks a proper understanding of the top-down approach of Autofilter. It is like dealing with multi-dimensional conditional probabilities - if that helps. It is interesting to see, that the reporting comment of bug 117276 already mentions the problem: it says in step 3: Wert=7000,9000 are not shown and not selectable and after removing Wert=8000 and adding Tesla, suddenly Wert=7000 (and 9000 by the way) are in the selection list. There is no row in this dataset for Tesla with Wert=7000,9000 so why should adding Tesla allow for such a selection? If you think the current behavior is questionable, why don't you stay professional and explain the strategy Autofilter is currently adhering to and explain why my comment doesn't hold instead of being vage as you did? Many thanks! -- You are receiving this mail because: You are the assignee for the bug.
