https://issues.apache.org/ooo/show_bug.cgi?id=125743

Jim DeLaHunt <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Version|4.0.1                       |OOo 3.3 or older

--- Comment #8 from Jim DeLaHunt <[email protected]> ---
Thank you for the clarifications. 

>> why change the "version" field
>
>Look at the help for the field Version:
>https://issues.apache.org/ooo/page.cgi?id=fields.html#version
>"The oldest version of the software the issue can be found in."
>As I know the function was never implemented in an other way, so it is "OOo 
>3.3 >or older" (OpenOffice.org before Apache OpenOffice).

I had not read this definition. Thank you for showing me. Based on your
understanding that the function was never implemented in another way, the
correct setting is indeed "OOo 3.3 >or older". I have changed it back.

> Expectations are not a bug of AOO.
> A bug occurs if a function does not what it should do by the given
> implementation.

I agree. So what is important in this discussion is the specified correct
behaviour for the function. 

I opened a forum post on this topic, "What is correct INDEX() behaviour in
array context?"
<https://forum.openoffice.org/en/forum/viewtopic.php?f=9&t=72871>. 

A valuable pointer there was to the specification for Apache OpenOffice's
functions,
<http://docs.oasis-open.org/office/v1.2/OpenDocument-v1.2-part2.html> (actually
the related .odt file is authoritative).

Specifically, the relevant sections seem to be:
* 6.14.6INDEX
<http://docs.oasis-open.org/office/v1.2/os/OpenDocument-v1.2-os-part2.html#__RefHeading__1018424_715980110>
* 3.3Non-Scalar Evaluation (aka 'Array expressions')
<http://docs.oasis-open.org/office/v1.2/os/OpenDocument-v1.2-os-part2.html#__RefHeading__1017866_715980110>

There is also a helpful comment from forum ocntributor 'acknak', " think the
problem with the bug report is that there is no clearly correct behavior. If
you find a situation where Calc adds 2 and 3 to get 6, then obviously there is
a problem, and the correct behavior is clear as well. I'm not seeing where
that's the case here. Calc's behavior may not make sense but there's no way to
say that it's wrong."

I'm interpreting the situation as, Calc does not have a specification for
correct behaviour of INDEX() returning a non-scalar value in an array
expression context, and therefore the reported behaviour is not wrong. It may
be unexpected or unsatisfactory, but it is not presently 'wrong'.

-- 
You are receiving this mail because:
You are the assignee for the issue.
You are watching all issue changes.

Reply via email to