https://bugs.documentfoundation.org/show_bug.cgi?id=99625
--- Comment #12 from Wolfgang Jäger <[email protected]> ---
No harm meant!
(In reply to Eike Rathke from comment #8)
> ... Pointers to the ODF OpenFormula definition
> don't help here, CONCAT and TEXTJOIN are not covered there.
Pointers to the OpenFormula document cannot help with a specification for
CONCAT and TEXTJOIN.
However, they can help to understand that the actual behaviour under
array-evaluation is improper with respect to general mandatory requirements
(subchapter 3.3). This would apply even if Excel 2016 did it otherwise - and
also was not based on any specification of the new functions in advance. In
addition: If Excel 2016 actually returns an array of identical elements under
the mentioned conditions, It will most likely also be in disorder concerning
general principles applying to Excel itself. Thus MS will have to fix a bug
themselves in that case.
Is somebody capable of checking for my above claim referring to the (should-be
existing) SPECIFICATIONS concerning array-evaluation in Excel (XLSX defining
documents)? I found not to be. I though feel sure that the authors of
OpenFormula did not fail to be compatible with Excel concerning the
fundamentals of array-evaluation.
Or - does somebody assume an accumulating function might be declared to return
1-1-array in the standard case? In this case subchapter 3.3, statement 2.2.1)
of OpenFormula would apply. Also consider the analogy with SUM.
To get finally convinced, try something like
{=TEXTJOIN(", ";FALSE();A1;A1:B4;A7:B8)}
(Appologies for my stubbornness. This touches a relevant question of priniples,
imo.)
--
You are receiving this mail because:
You are the assignee for the bug._______________________________________________
Libreoffice-bugs mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs
