[Libreoffice-bugs] [Bug 136621] CALC INDIRECT() Function does not work correctly if local function separator for formulae is defined

[bugzilla-daemon at bugs.documentfoundation.org]

https://bugs.documentfoundation.org/show_bug.cgi?id=136621

--- Comment #3 from Colin <that.man.colin at gmail.com> ---
(In reply to Eike Rathke from comment #2)
> 0. the Swedish currency symbol certainly is not ':' colon but 'kr' for
> Svensk krona instead.
> 
> If the function parameter separator is ';' semicolon then the formula
> =SUM(D1;INDIRECT(C1))
> returns the sum of two individual arguments, namely D1 and the reference
> returned by the call to INDIRECT(C1).
> 
> In the formula
> =SUM(D1:INDIRECT(C1))
> the ':' colon is the range operator so the range passed to sum is the result
> of D1 as one corner and the second corner is the reference returned by the
> call to INDIRECT(C1). So if C1 contains "D7" the resulting range is D1:D7
> passed to SUM().
> 
> There is no bug and not an esoteric issue, the function parameter separator
> is not misbehaving and there is no malfunction. It is simply that you were
> not aware of the syntax.

Hi Eike,
I inadvertently referred to the currency separator as the currency symbol.
The currency symbol is in fact the two character ISO country code for Sverige
which is "SE" plus the single character currency initial "K" hence "SEK". The
currency separator for numerical representations of the value is the colon with
two potential presentations - 123:25 when there are fractional portions and
123:- when it is a simple integer. The fractional representation is sometimes
also supplemented with "k" or "kr". As we have three local Kronor to contend
with (Danska, Norska och Svenska)monetary represetations become slightly
complicated.
Something has definitely gone wrong. Further experiments have identified that
if I use the "drag select" and Sigma [sum] function or alternatively the Sigma
[sum] function in an empty cell adjacent to "populated" data cells then the
system automatically writes two different formulae - one each with the colon
and semicolon separators. It also produces inconsistency when some of the
selected or inferred cells contain the results of their own formulae.
My assumption that it was the INDIRECT() function was indeed erroneous. I will
endeavour to clearly define the inconsistencies and submit a new report

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/libreoffice-bugs/attachments/20200910/2a9d6fe1/attachment.htm>