https://bugs.documentfoundation.org/show_bug.cgi?id=156233

ady <adylo811...@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEEDINFO                    |NEW
           Keywords|                            |bibisectRequest, regression
            Summary|data validation error when  |Validation #REF! error when
                   |I duplicate sheet and later |deleting original worksheet
                   |delete original sheet       |after duplicating it
            Version|7.4.7.2 release             |4.2.0.4 release

--- Comment #8 from ady <adylo811...@gmail.com> ---
The prior attachments are less than ideal but usable (if the user/tester knows
what to look for). Further investigations confirm the bug:

Using any cell reference in any of the possible validation types (not only
custom but any and all of them), will result in a #REF! error in the source
values of the Validity dialog when following these steps (which use just one of
the possible validity types, as example)

1. A1: 1
2. A2: 3

3. A3: Menu Data > Validity:
3.1. Allow: Whole Numbers
3.2. Data: valid range
3.3. Min: =A1  (Calc will delete the "=" sign when closing the dialogue)
3.4. Max: =A2  (Calc will delete the "=" sign when closing the dialogue)
3.5. OK

4. A3: 2
5. Duplicate the current worksheet (aka "copy" worksheet in older versions).
6. Delete the original worksheet.  <- The corruption is generated at this
point!!!
6.1. A3: 3 (the value is rejected, same as trying the prior value "2" again).
7. Save the file.
8. Close the file; reopen.  <-  This is required in order to _see_ the problem.
9. A3: 3 (the value is rejected, same as trying the prior value "2" again).
10. A3: Menu Data > Validity; the Min and Max fields will show the #REF! error.

Notes:
* Step 9 is optional.
* Step 6 already causes the corruption/problem.
* Step 6.1 is optional.
* Adding a "6.2" step to show the Min and Max fields will not show the #REF!
error yet. Closing the dialogue with OK at this point "restores" the correct
values.

The problem is reproduced in LO 4.2 but not in 4.1 and older > regression.

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to