In LibO/OOo normally you have to use a semicolon as parameters separator, not a comma:

=IF((G19+H19)>0;(G19-H19);"")

That is correct. But when you try to change the comma to a semicolon it is
not accepted and changes back to a comma (and the error).
The funny thing is that when I open the document in Excel on a WINDOWS
machine, I see the comma as a semicolon.
Regards, Denham

On Sun, Jun 5, 2011 at 12:54 PM, <bugzilla-daemon@freedesktop.org> wrote:

> https://bugs.freedesktop.org/show_bug.cgi?id=37942
>
> --- Comment #1 from vitriol <vitriol_vitriol@katamail.com> 2011-06-05
> 00:24:22 PDT ---
> In LibO/OOo normally you have to use a semicolon as parameters separator,
> not a
> comma:
>
> =IF((G19+H19)>0;(G19-H19);"")
>
> --
> Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
> ------- You are receiving this mail because: -------
> You reported the bug.
>

(In reply to comment #2)

> That is correct. But when you try to change the comma to a semicolon it is
> not accepted and changes back to a comma (and the error).

I'm not able to reproduce this behavior with version 3.4.0 under Win7.

I am using Ubuntu 11.4. It  must be that the bug is somewhere in Ubuntu. I
also have noticed that there is no problem when running the same document
through Excel on Windows XP.
Denham

On Sun, Jun 5, 2011 at 1:09 PM, <bugzilla-daemon@freedesktop.org> wrote:

> https://bugs.freedesktop.org/show_bug.cgi?id=37942
>
> --- Comment #3 from vitriol <vitriol_vitriol@katamail.com> 2011-06-05
> 00:39:04 PDT ---
> (In reply to comment #2)
>
> > That is correct. But when you try to change the comma to a semicolon it
> is
> > not accepted and changes back to a comma (and the error).
>
> I'm not able to reproduce this behavior with version 3.4.0 under Win7.
>
> --
> Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
> ------- You are receiving this mail because: -------
> You reported the bug.
>

Maybe relater to Bug 37860?

@Denham
Please, attach the sample document that you have mailed me. I confirm that is a regression.

Created attachment 47546 [details]
A spreadsheet produced in LibreOffice which shows the formulas as faulty.

Created attachment 47547 [details]
Appearance of spreadsheet when using LibreOffice

I nominate this bug for "LibreOffice 3.4 most annoying bugs".

Well, when you have to fish out an ancient computer running WINDOWS XP from the bottom of some cupboard just to send out an invoice, I guess you could call that 'annoying' ;-)
Denham

Kohei, could you please have a look?

I'm afraid this is just another manifestation of the unfortunate event as outlined in Bug 37860.  In short, this is not a bug, and the reason and rationale is explained in Bug 37860.

To fix this, please change the "" to 0 in columns G and H.  Then the formulas in column F should work as expected (if I understand the intent of those formulas correctly).

Not a regression.

(In reply to comment #12)

> Not a regression.

Not a regression? Old documents don't work anymore as expected, therefore *is* a regression!
This behavior (bug, for me...) breaks compatibility with OOo and previous versions of LibO.

(In reply to comment #13)
> (In reply to comment #12)
> 
> > Not a regression.
> 
> Not a regression? Old documents don't work anymore as expected, therefore *is*
> a regression!
> This behavior (bug, for me...) breaks compatibility with OOo and previous
> versions of LibO.

Then I challenge you to come up with a solution (or best compromise) that best serves a wide variety of use cases.  Just saying "fix this for me and break it for others" is a not a great solution I would go for.

@Kohei Yoshida
What's really wrong in previous behavior?

See Bug 37860 for my explanation.  I won't repeat it here.

Thanks for letting me know that the bug was not a bug. The work-around (to replace the "" with a zero in the If arguments) worked well so I have no complaints. If anyone else is bugged by this non-bug it may be useful to them too.
Entomology was never my strong-point, so you must forgive me from being a bit confused by the difference between bugs and non-bugs.
Denham