https://bugs.documentfoundation.org/show_bug.cgi?id=166275

--- Comment #8 from Robert Lacroix <[email protected]> ---
(In reply to Heiko Tietze from comment #7)
> Perhaps we find a solution to the actual use case. I doubt that you need to
> see every tab resp. to click on any. For example, if you feed the sheet from
> external sources we could flash the tab or highlight it, if you refer in a
> formula to sheet$foo this could become hyperlinky, etc.

Since we are talking about use cases, the following should actually end up in a
new UX bug request for Pivot Tables. Feel free to fork one from this
conversation.

What you see above is actually simplified my use case. My file also has
"control" tabs which affect what calculations do in the data tabs. The several
control sheets have unique formulas and layouts.

The 71 data sheets have identical layout except for the number of rows
containing actual data. Each data sheet has a set of fixed header rows which
are identical across all data sheets, containing such things as a reference to
a control sheet cell, and a list of valid values for data in that spreadsheet
column to appear in the Pivot table. The Pivot Table data range extends into
the header above the actual data to add the minimum required categories for
that spreadsheet column. Locking the Pivot Table column field to a subset of
these header values fixes the Pivot Table to a static set of columns regardless
of what data appears in the data rows.

It would be ideal for Pivot Tables to have optional "validation" sheet ranges
in the layout dialog attached to column fields, which are separate from the
data. Then I would not need these header rows in every data sheet - they could
point to a control sheet. A validation range is an attribute of a column field
in the Pivot Table so that it provides a list of canonical values which may
appear in data of that spreadsheet column, even if the data range is empty. The
goal is to have a set of static Pivot Table columns. Using a validation range
for a column field would automatically provide an initial set of values to lock
the categories of the field.

I am working around this ideal structure by extending the Pivot Table up into
the header rows so that validation values are included, and manually locking
the desired category values so that valid values which are not used, or invalid
values which may appear in the data, do not cause extra columns to appear in
the Pivot Table. This was the motivation to ask for the "Lock" attribute of
category fields, as well as protecting adjacent spreadsheet cells from being
overwritten by the Pivot Table.

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

Reply via email to