To comment on the following update, log in, then open the issue:
http://www.openoffice.org/issues/show_bug.cgi?id=33851





------- Additional comments from [EMAIL PROTECTED] Fri Aug 18 22:13:58 -0700 
2006 -------
As Eike (er) points out, to be "like Excel" is not a sufficient specification 
of the requirement. When 
people say "like Excel" this might be a considered view covering all scenarios, 
but in other cases the 
person only sees a small part of the behaviour and does not know about other 
things done by Excel. So 
"like Excel" is only a guide but it is not sufficient for definition of a 
requirement. 

I think it is useful first to consider how a filtered spreadsheet is used. It 
is different from other uses of a 
spreadsheet. The filter is a visual tool, that changes a view of the values, 
while the underlying cell 
contents including formulas are not changed. The filtered spreadsheet mainly 
has significance for the 
user looking at the values. Typical uses of a filter would be:
- view rows where a cell in some columns meets some criteria
- clean up data - change certain values (like re-code of data)
- fill cells with data according to some criteria
- extract (copy) values from rows that meet some criteria
Notice that the user in these cases is only concerned with values. In my 
experience any other uses, 
would be better done a different way (not using filters).

Also consider how other features are implemented when there is ambiguity about 
the user intention. 
Copy/paste is a good example. Does the user intend to copy the formula or the 
value? There are plenty 
of users that have no idea about this, and the feature defaults to paste the 
formulas. Should formulas 
be replicated as relative reference or absolute? What is the intention of the 
user? My point here is that 
ambiguity about features is common and does not prevent implementaation, it is 
only necessary that a 
decision is made. 

I would never recommend a user (either Excel or OOo) to adopt scenario 2,3 or 
5, as I do not consider 
those to be safe or good practice, however I accept others may feel 
differently.  So I will comment only 
on 1 & 4. If anyone explains their reasons for using 2,3 or 5 I will try and 
suggest a different method.

Scenario 1 copy/paste - in my experience this is useful only for pasting 
values. I cannot imagine why 
anyone would like to paste formulas copied from a filtered range, or how the 
users intention could be 
described. The decision in Excel to paste values seems a reasonable solution, 
and would at least be an 
improvement over the current behaviour. I would be interested if anyone can 
explain the reasons for 
pasting formulas from a filtered copy.

Scenario 4 fill - for me this is the feature that I have used frequently in 
Excel and it is my "most wanted" 
for OOo. I think this feature alone would be large improvement. The requirement 
in this case seems 
clearly specified.

Regarding the Excel implementation, I would not search for logic there. My 
recollection is the same bug 
existed in early Excel filters and was fixed about 13 years ago, possibly some 
compromise was 
necessary. 

---------------------------------------------------------------------
Please do not reply to this automatically generated notification from
Issue Tracker. Please log onto the website and enter your comments.
http://qa.openoffice.org/issue_handling/project_issues.html#notification

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to