https://bugs.documentfoundation.org/show_bug.cgi?id=157268
Stéphane Guillou (stragu) <stephane.guil...@libreoffice.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Blocks| |90486
Resolution|--- |NOTABUG
CC| |stephane.guillou@libreoffic
| |e.org
--- Comment #14 from Stéphane Guillou (stragu)
<stephane.guil...@libreoffice.org> ---
Thank you all.
(In reply to m.a.riosv from comment #6)
> Seems to me the issue comes from the data.
> There are no hours from the same date, but from different dates.
Thanks for spotting that, m.a.riosv.
(In reply to MikeD from comment #7)
> However, I did create a new set of time stamps using fractional numeric
> values, and then format them to time., and created a new chart, this gets
> the axis right.
Glad that works now.
(In reply to ady from comment #8)
> I guess _that_ is the key part of this ticket. I mean, users might want to
> have an axis of the chart displayed in a different format than the format
> used for the data itself. Such expectation is reasonable: an adequate
> formatting for a chart might be different than an adequate formatting in
> cells/table.
We do have that in axis formatting > Numbers: we can use "source format" or
untick that and pick any format.
(In reply to MikeD from comment #9)
> Is it a BUG: yes in that what appears to be correct data does not produce
> the logical result.
As m.a.riosv mentioned, the underlying data is not what you thought it was. We
can not expect functions to base their calculation on how the data is formatted
instead of the actual value it holds. Just as one example, different people
will interpret one single formatting as different things. Is 1,234 a thousand
marker or a decimal marker? Is 3/12 a division or a date? etc. So "logical"
will be different from one person to the other. We have to rely on actual,
underlying data.
If you are forced to work with "bad data" for whatever reason, there are
functions that might help get to clean data, for example extracting time data
with hour(), minute(), second().
> It is also, to some extent a feature enhancement requirement: In that time
> formatting is so variable,one needs to specify both an input and output
> format - possibly also with a test to translate the applied format on the
> source data.
If there is an enhancement request, please open a new report as this report has
already become too convoluted. (And link it here.)
In any case, the original issue is "not a bug" as it was a case of bad data
hidden by the formatting.
Referenced Bugs:
https://bugs.documentfoundation.org/show_bug.cgi?id=90486
[Bug 90486] [META] Chart bugs and enhancements
--
You are receiving this mail because:
You are the assignee for the bug.
