Bug 141590 - Crash if i insert values in the function FORECAST.ETS.PI.ADD
Summary: Crash if i insert values in the function FORECAST.ETS.PI.ADD
Status: RESOLVED NOTABUG
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Calc (show other bugs)
Version:
(earliest affected)
6.1.5.2 release
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Not Assigned
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: Calc-Function
  Show dependency treegraph
 
Reported: 2021-04-09 21:43 UTC by Willow
Modified: 2021-12-04 14:02 UTC (History)
3 users (show)

See Also:
Crash report or crash signature:


Attachments
LibreOffice crashed (20.26 KB, image/png)
2021-04-09 21:48 UTC, Willow
Details
formula (274.56 KB, application/vnd.oasis.opendocument.spreadsheet)
2021-04-11 13:13 UTC, Willow
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Willow 2021-04-09 21:43:42 UTC
Description:
After LibreOffice crashed and delete each change in the file 

Steps to Reproduce:
1.Open the LibreOffice Calc 
2.Introduce formula and values 
3.Get a result 
4.LibreOffice crashed 

Actual Results:
LibreOffice crashed 

Expected Results:
Would give me a result 


Reproducible: Always


User Profile Reset: Yes



Additional Info:
[Information automatically included from LibreOffice]
Locale: es
Module: SpreadsheetDocument
[Information guessed from browser]
OS: Linux (All)
OS is 64bit: yes
Comment 1 Willow 2021-04-09 21:48:35 UTC
Created attachment 171072 [details]
LibreOffice crashed

LibreOffice crashed
Comment 2 Julien Nabet 2021-04-10 10:19:36 UTC
On which Linux distrib + which version are you?
Could you give an update to a recent LO version?
Indeed, 6.1.5 is quite old, last stable ones are 7.0.5 or brand new LO 7.1.2.

If you still reproduce the crash, could you provide the values entered to reproduce the crash?
Comment 3 Willow 2021-04-10 16:43:46 UTC
Linux 4.19.0-16-amd64 #1 SMP Debian 4.19.181-1 (2021-03-19) x86_64 GNU/Linux

Versión: 6.1.5.2
Id. de compilación: 1:6.1.5-3+deb10u7
Subprocs. CPU: 4; SO: Linux 4.19; Repres. IU: predet.; VCL: x11; 
Configuración regional: es-NI (es_NI.UTF-8); Calc: group threaded

Thread 1 "soffice.bin" received signal SIGSEGV, Segmentation fault.
0x00007f984a901b57 in ?? () from /usr/lib/libreoffice/program/libsclo.so
Comment 4 Julien Nabet 2021-04-10 17:04:07 UTC
About Debian, I personnally use Debian testing (which has LO 7.0.4).
Stable is more for servers.

Could you indicate parameters you provided to the function?

About backtrace, when you got "??", type "c" (for "continue") until there's no "??".
Then you can type "bt" (for "backtrace")
Comment 5 Willow 2021-04-11 12:52:35 UTC
PRONOSTICO.ETS.PI.SUM(C40,D13:D40,C13:C40)

(gdb)bt
#0  0x00007f2d0651fb57 in  () at /usr/lib/libreoffice/program/libsclo.so
#1  0x00007f2d065240ac in  () at /usr/lib/libreoffice/program/libsclo.so
#2  0x00007f2d0650b114 in  () at /usr/lib/libreoffice/program/libsclo.so
#3  0x00007f2d063fd493 in  () at /usr/lib/libreoffice/program/libsclo.so
#4  0x00007f2d063fd689 in  () at /usr/lib/libreoffice/program/libsclo.so
#5  0x00007f2d0682bcd4 in  () at /usr/lib/libreoffice/program/libsclo.so
#6  0x00007f2d0c31240b in  () at /usr/lib/libreoffice/program/libforuilo.so
#7  0x00007f2d0c3134dc in  () at /usr/lib/libreoffice/program/libforuilo.so
#8  0x00007f2d0c31d667 in  () at /usr/lib/libreoffice/program/libforuilo.so
#9  0x00007f2d18075180 in Control::ImplCallEventListenersAndHandler(VclEventId, std::function<void ()> const&) ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#10 0x00007f2d18075291 in Control::EventNotify(NotifyEvent&) ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#11 0x00007f2d1804f53c in vcl::Window::GetFocus() ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#12 0x00007f2d1808161e in Edit::GetFocus() ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#13 0x00007f2d0c31ac59 in formula::RefEdit::GetFocus() ()
    at /usr/lib/libreoffice/program/libforuilo.so
#14 0x00007f2d18001391 in  () at /usr/lib/libreoffice/program/libmergedlo.so
#15 0x00007f2d1807c348 in Edit::MouseButtonDown(MouseEvent const&) ()
    at /usr/lib/libreoffice/program/libmergedlo.so
--Type <RET> for more, q to quit, c to continue without paging--
#16 0x00007f2d1805a53c in  () at /usr/lib/libreoffice/program/libmergedlo.so
#17 0x00007f2d1805c6a5 in  () at /usr/lib/libreoffice/program/libmergedlo.so
#18 0x00007f2d0de67d82 in X11SalFrame::HandleMouseEvent(_XEvent*) ()
    at /usr/lib/libreoffice/program/libvclplug_genlo.so
#19 0x00007f2d0de2dab8 in SalX11Display::Yield() ()
    at /usr/lib/libreoffice/program/libvclplug_genlo.so
#20 0x00007f2d0de2db69 in  ()
    at /usr/lib/libreoffice/program/libvclplug_genlo.so
#21 0x00007f2d0de2c580 in SalXLib::Yield(bool, bool) ()
    at /usr/lib/libreoffice/program/libvclplug_genlo.so
#22 0x00007f2d1827816e in Application::Yield() ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#23 0x00007f2d182798e5 in Application::Execute() ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#24 0x00007f2d17333239 in  () at /usr/lib/libreoffice/program/libmergedlo.so
#25 0x00007f2d1827e9e6 in  () at /usr/lib/libreoffice/program/libmergedlo.so
#26 0x00007f2d1827eae0 in SVMain() ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#27 0x00007f2d17350de1 in soffice_main ()
    at /usr/lib/libreoffice/program/libmergedlo.so
#28 0x000055b9aa15e07b in  ()
#29 0x00007f2d1520209b in __libc_start_main (main=
    0x55b9aa15e070, argc=2, argv=0x7ffdbd93f7e8, init=<optimized out>, fini=<opt--Type <RET> for more, q to quit, c to continue without paging--
imized out>, rtld_fini=<optimized out>, stack_end=0x7ffdbd93f7d8)
    at ../csu/libc-start.c:308
#30 0x000055b9aa15e0ba in  ()
Comment 6 Julien Nabet 2021-04-11 12:59:34 UTC
I've got error 508 but since I use this formula in an empty spreadsheet, perhaps it's expected.

Could you attach a minimum example file to reproduce this or do you reproduce this too with an empty spreadsheet?
Comment 7 Julien Nabet 2021-04-11 13:00:06 UTC
(English name of the function FORECAST.ETS.PI.ADD)
Comment 8 Willow 2021-04-11 13:13:22 UTC
Created attachment 171099 [details]
formula

add the following file
Comment 9 Julien Nabet 2021-04-11 13:21:04 UTC
I tried to put the formula in English or in Spanish in "engaño" sheet, I don't reproduce the crash, only error 508.
I don't have more questions so put it back to UNCONFIRMED.
I suppose other people will find a way to reproduce this.
Comment 10 Willow 2021-04-11 13:24:56 UTC
I think that not reproduce a error because still lack a parameter 

PRONOSTICO.ETS.PI.SUM(C40,D13:D40,C13:C40,1)
Comment 11 Julien Nabet 2021-04-11 18:14:04 UTC
=FORECAST.ETS.PI.ADD(C40;D13:D40;C13:C400;1)
=> Err:502
(with LO Debian 7.0.4 or with master sources updated today).
Comment 12 Xisco Faulí 2021-04-11 20:54:29 UTC
(In reply to Willow from comment #3)
> Linux 4.19.0-16-amd64 #1 SMP Debian 4.19.181-1 (2021-03-19) x86_64 GNU/Linux
> 
> Versión: 6.1.5.2
> Id. de compilación: 1:6.1.5-3+deb10u7
> Subprocs. CPU: 4; SO: Linux 4.19; Repres. IU: predet.; VCL: x11; 
> Configuración regional: es-NI (es_NI.UTF-8); Calc: group threaded
> 
> Thread 1 "soffice.bin" received signal SIGSEGV, Segmentation fault.
> 0x00007f984a901b57 in ?? () from /usr/lib/libreoffice/program/libsclo.so

Thank you for reporting the bug.
it seems you're using an old version of LibreOffice.
Could you please try to reproduce it with the latest version of LibreOffice from https://www.libreoffice.org/download/libreoffice-fresh/ ?
I have set the bug's status to 'NEEDINFO'. Please change it back to 'UNCONFIRMED' if the bug is still present in the latest version.
Comment 13 Winfried Donkers 2021-12-04 14:02:40 UTC
Using attachment 171099 [details] and formula =FORECAST.ETS.PI.ADD(C40;D13:D40;C13:C400;1) from comment#11 problem is clear:
the first given argument (target) is part of the data given to calculate the forecast from. That is impossible, you cannot forecast a value that is given.

When you add in C41 a date (say 2022-03-01) and change the first argument from C40 to C41, you get a result. In this case 0 because the data given do not show any variation, making the use of exponential triple smoothing (ETS) forecast functions useless. When there are no periodic influences, standard forecast functions will suffice, when there is no variation at all, forecast functions are not needed.

Note: the last given argument (confidence level 1) is not realistic. This is a forecast function and 100% reliability is not achievable, unless the data set is such that forecasting functions are not needed (as is the case in this attachment);