[upstream] calc: move cell does not update row sum

[SatPhil]

See attached test case. Sum two rows: A1 + B1 = C1; A2 + B2 = C2 where B1 is blank. Cut B2 and paste into B1. Result C2 is decremented correctly by old B2 value; but result C1 is NOT incremented by new B1 value. Formula of C1 remains correct =SUM($Sheet1.A1:B1) as is C2's formula but C1 result is wrong. Saving file, close and re-open does not fix incorrect C1 result. Undo undo to return to initial state results in C2 now having incorrect result. Bug occurs whether drag and drop or control-x/control-v used. This is a data integrity issue as the wrong answer is being displayed.

ProblemType: Bug
DistroRelease: Ubuntu 16.04
Package: libreoffice-calc 1:5.1.6~rc2-0ubuntu1~xenial2
ProcVersionSignature: Ubuntu 4.4.0-83.106-generic 4.4.70
Uname: Linux 4.4.0-83-generic x86_64
NonfreeKernelModules: wl
ApportVersion: 2.20.1-0ubuntu2.9
Architecture: amd64
CurrentDesktop: Unity
Date: Mon Jul 17 17:33:42 2017
EcryptfsInUse: Yes
InstallationDate: Installed on 2016-05-08 (435 days ago)
InstallationMedia: Ubuntu 16.04 LTS "Xenial Xerus" - Release amd64 (20160420.1)
SourcePackage: libreoffice
UpgradeStatus: No upgrade log present (probably fresh install)

[Aron Budea (baron-z)]

Created attachment 146149
Sample ODS

- Open attached spreadsheet.
- Cut A1.
- Paste into A2 (overwrite existing data).

=> F2, which is the sum of A2:D2, isn't updated accordingly (should become 22 instead of 26).

Observed using LO 6.2.0.0.alpha1+ (313392119522c21a6ecd14403d6f92c948149df7) & 4.4.0.3 / Ubuntu 18.04.
No issue in LO 4.3.0.4.
=> regression

Bibisected to the following commit using repo bibisect-44max.

https://cgit.freedesktop.org/libreoffice/core/commit/?id=5ef856e975d7c0396984d588a43fd1a7c7085c55
author Kohei Yoshida <email address hidden> 2014-11-18 21:03:05 -0500
committer Kohei Yoshida <email address hidden> 2014-11-18 21:03:50 -0500

Use group area listeners during ODS import.

[Aron Budea (baron-z)]

There are a couple of other, already fixed bugs starting with the same commit, eg. bug 86615.

[Raal (raal)]

Confirm with Version: 6.2.0.0.alpha1+
Build ID: a50f845957fff66ca7f459da4878be4597e0bae3
CPU threads: 4; OS: Linux 4.4; UI render: default; VCL: x11;

[Olivier Tilloy (osomon)]

I can reliably reproduce the issue with the test case attached, on Ubuntu 16.04 and 19.04 (libreoffice 6.1.3.2).
I can also reproduce the issue with the upstream build of libreoffice (6.1.3.2).

Would you mind filing an upstream bug report at https://bugs.documentfoundation.org/enter_bug.cgi?product=LibreOffice&format=guided and sharing the link to that bug here?

Thanks!

[SatPhil (satphil)]

Description:
See attached test case. Sum two rows: A1 + B1 = C1; A2 + B2 = C2 where B1 is blank. Cut B2 and paste into B1. Result C2 is decremented correctly by old B2 value; but result C1 is NOT incremented by new B1 value. Formula of C1 remains correct =SUM($Sheet1.A1:B1) as is C2's formula but C1 result is wrong. Saving file, close and re-open does not fix incorrect C1 result. Undo undo to return to initial state results in C2 now having incorrect result. Bug occurs whether drag and drop or control-x/control-v used. This is a data integrity issue as the wrong answer is being displayed.

Steps to Reproduce:
1. Sum two rows: A1 + B1 = C1; A2 + B2 = C2 where B1 is blank.
2. Cut B2 and paste into B1. Result C2 is decremented correctly by old B2 value; but result C1 is NOT incremented by new B1 value.
3. Formula of C1 remains correct =SUM($Sheet1.A1:B1) as is C2's formula but C1 result is wrong.
4. Saving file, close and re-open does not fix incorrect C1 result.
5. Undo undo to return to initial state results in C2 now having incorrect result.
6. Bug occurs whether drag and drop or control-x/control-v used.

Actual Results:
C1 not incremented by new B1 value; undo does not return to initial state

Expected Results:
C1 incremented by new B1 value; undo does return to initial state

Reproducible: Always

User Profile Reset: No

Additional Info:
Reproduced reliably by <email address hidden> on Ubuntu 16.04 and 19.04 (libreoffice 6.1.3.2).
He can also reproduce the issue with the upstream build of libreoffice (6.1.3.2)

[SatPhil (satphil)]

Created attachment 146856
"Before" state of test case.

[SatPhil (satphil)]

Thanks for looking at this. Filed upstream on Bugzilla: bug 121568

https://bugs.documentfoundation.org/show_bug.cgi?id=121568

[Xiscofauli (xiscofauli)]

Regression introduced by https://cgit.freedesktop.org/libreoffice/core/commit/?id=5ef856e975d7c0396984d588a43fd1a7c7085c55, which is the same as in bug 121043.
Closing as RESOLVED DUPLICATED

*** This bug has been marked as a duplicate of bug 121043 ***

[Xiscofauli (xiscofauli)]

*** Bug 121568 has been marked as a duplicate of this bug. ***

[Xiscofauli (xiscofauli)]

Hi Eike, I've just found this issue as being the same root cause of bug 121568, I thought you might be interested in this issue as it's affecting formulas...

[Oliver-brinzing (oliver-brinzing)]

this issue is no longer reproducible with:

Version: 6.3.0.0.alpha0+ (x64)
Build ID: 3140194a85fe4a6ac69c8cddc4d3b019430cd6e8
CPU threads: 4; OS: Windows 10.0; UI render: default; VCL: win;
Locale: de-DE (de_DE); UI-Language: en-US
Calc: threaded

-> F2 is 22 now after pasting

[Xiscofauli (xiscofauli)]

Still reproducible in

Version: 6.3.0.0.alpha0+
Build ID: a11a2d84b09f85d2020c47f3ce42cd9efbff818a
CPU threads: 4; OS: Linux 4.15; UI render: default; VCL: gtk3;
Locale: ca-ES (ca_ES.UTF-8); UI-Language: en-US
Calc: threaded

[Oliver-brinzing (oliver-brinzing)]

(In reply to Xisco Faulí from comment #6)
> Still reproducible in
> Version: 6.3.0.0.alpha0+
> Build ID: a11a2d84b09f85d2020c47f3ce42cd9efbff818a

are you sure ? i can not reproduce again in my new dev build:

Version: 6.3.0.0.alpha0+ (x64)
Build ID: a11a2d84b09f85d2020c47f3ce42cd9efbff818a
CPU threads: 4; OS: Windows 10.0; UI render: default; VCL: win;
Locale: de-DE (de_DE); UI-Language: en-US
Calc: threaded

[Xiscofauli (xiscofauli)]

You're right!!
This is fixed by https://cgit.freedesktop.org/libreoffice/core/commit/?id=7fdc5df36f5b50e0629405a47ff3d5765fcfeb93

[Xiscofauli (xiscofauli)]

No need a test for this, this is already covered by https://gerrit.libreoffice.org/plugins/gitiles/core/+/6280b5c1c62ad40b5b9780a93c7cbee9ca0260f8%5E%21

[Xiscofauli (xiscofauli)]

Setting to VERIFIED FIXED

[Marcus Tomlinson (marcustomlinson)]

Upstream fix is scheduled for 6.3.0 (Week 32)