Summary: | Improper display of boolean cell value stats in LibreOffice Calc status bar | ||
---|---|---|---|
Product: | LibreOffice | Reporter: | Valery <kval> |
Component: | Calc | Assignee: | Not Assigned <libreoffice-bugs> |
Status: | UNCONFIRMED --- | ||
Severity: | minor | CC: | erack, stephane.guillou |
Priority: | medium | ||
Version: | Inherited From OOo | ||
Hardware: | All | ||
OS: | All | ||
Whiteboard: | |||
Crash report or crash signature: | Regression By: | ||
Attachments: | sample ODS |
Description
Valery
2024-03-21 09:57:52 UTC
I am unable to reproduce this. Can you please attach screenshots of the error? Version: 24.2.0.3 (X86_64) / LibreOffice Community Build ID: da48488a73ddd66ea24cf16bbc4f7b9c08e9bea1 CPU threads: 4; OS: Windows 10.0 Build 19045; UI render: Skia/Vulkan; VCL: win Locale: en-IN (en_IN); UI: en-US Calc: threaded In Excel, there is such a thing as a boolean type of value, which is (usually, not always) automatically interpreted as a number when such conversion is required; in some cases, some function might be required. In Calc, boolean values _are_ numeric values. FALSE() is equivalent to zero; zero is equivalent to FALSE(). TRUE() is equivalent to one; any non-zero numeric value is equivalent to TRUE(). There might be cases in which users might take advantage of showing "Average:TRUE; Sum:TRUE". As described in comment 0, this already happens. Perhaps the similar "Average:FALSE; Sum:FALSE" case might have some advantage for some users. My only concern (as a user) would be a potential misinterpretation, either from what such status bar text means (a user that has no knowledge about boolean might misinterpret it as some kind of error/crash), or from misinterpreting other "kinds" of zero as "FALSE()" (including but not limiting [empty] cases). Created attachment 193545 [details]
sample ODS
The behaviour is inherited from OOo, I see the same in OOo 3.3.
From what ady explained, I think the current distinction makes some sense.
However, there is some inconsistency between the result of the function in cell and the status bar: Average() and Sum() always return numeric data, even for a single TRUE value as input. Shouldn't the status bar always match that?
Another inconsistency is that the status bar returns boolean values if a range is selected with the cell cursor on a FALSE cell (see attached spreadsheet, select range B1:B2 starting from B2). Why not numeric, if a 0 is shown when a single FALSE cell is selected?
I think I am leaning toward the most transparent, consistent and useful: always show a numeric value in the status bar.
(e.g. M1:N2 sum in sample ODS should be 4 in status bar)
erack, what's your opinion? (In reply to Stéphane Guillou (stragu) from comment #3) > (e.g. M1:N2 sum in sample ODS should be 4 in status bar) I disagree. If the original request in comment 0 (i.e. show FALSE when relevant) cannot be implemented, I would _not_ change the current behavior of showing TRUE in order to always show a number. There is no point in changing a behavior that might be useful to some users (and it is already implemented) just because the counter-part cannot be implemented too. The status bar displays the result in the format of the currently active cell of the selection. That is desired. e.g. if in the attached sample N1:N4 are selected starting from N1 downwards then status is "Average: TRUE; Sum: TRUE". If the selection instead is started from N4 upwards then status is "Average: 1.25; Sum: 5". Now we could start treating boolean formats differently, and while at it then date(+time) formats because they don't make much sense either, but keep for time formats because durations are summed, what else? |