Bug 129541 - calc CALCULATING: autocalculate with 'shared formulas' still broken if a shared formula has #REF! problem
Summary: calc CALCULATING: autocalculate with 'shared formulas' still broken if a shar...
Status: RESOLVED DUPLICATE of bug 128975
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Calc (show other bugs)
Version:
(earliest affected)
4.2 all versions
Hardware: All All
: medium normal
Assignee: Not Assigned
URL:
Whiteboard:
Keywords: bibisected, bisected, regression
Depends on:
Blocks:
 
Reported: 2019-12-21 17:10 UTC by b.
Modified: 2020-02-18 02:34 UTC (History)
1 user (show)

See Also:
Crash report or crash signature:


Attachments
a file with the failing formula and a macro to copy it in the failing place (13.68 KB, application/vnd.oasis.opendocument.spreadsheet)
2019-12-21 17:17 UTC, b.
Details

Note You need to log in before you can comment on or make changes to this bug.
Description b. 2019-12-21 17:10:53 UTC
Description:
shortest: 

load attachment from next comment, 
click button to provoke error, 
calculate cells B2 and B3 manualy, 
displayed values are wrong, 

short: 

A3: 3              '(any number will fit), 
B3: =SUM(A2:A4)    'similar formulas will fail too, 
A3:B3: copy 
A1:B3: paste

observe B1: '#REF!' - correct, 

play with values in A1:A3, B2:B3 cut off from autocalculation - not correct,  

alternative: 

A1:B1 must not! be filled with the same formula, 
A3:B3: copy        '    
A2:B4: paste

play with values in A2:A4, B2:B4 should react correctly, 

long: 

there have been several cases where autocalculate was broken for 'shared formulas', delete or insert in referenced ranges, changing formulae, inserting columns and so on, 

some have been patched in the nearby past, this one remains, 

i think it's already adressed in another case treating nested formulas, 

may be this is the essence causing the errors there, 

have fun, 

merry x-mas

b. 

P.S: 

call me crazy ... i only want this program to become reliable again - if possible, 

bug flooding? - imho the original bug is only one, something 'bigger', more fundamental, in the past (4.1 -> 4.2), as of now and as of how bugs are handeled there's nothing i can do than observe and report, 

forum trolls: pls. do NOT! ask if i have / had <data-calculate-autocalculate> enabled, also resetting user profile won't help, 

will provide testsheet with macro to produce the error in next comment, 

Steps to Reproduce:
see above

Actual Results:
cells cut off from auto recalculation

Expected Results:
all cell should be recalculated correctly in any case, 


Reproducible: Always


User Profile Reset: No



Additional Info:
Version: 6.5.0.0.alpha0+ (x64)
Build ID: 209fc9fd7fa433947af0bf86e210d73fa7f5a045
CPU threads: 8; OS: Windows 6.1 Service Pack 1 Build 7601; UI render: default; VCL: win; 
Locale: de-DE (de_DE); UI-Language: en-US
Calc: 

last verson i've seen without this bug: 4.1.6.2
Comment 1 b. 2019-12-21 17:17:47 UTC
Created attachment 156723 [details]
a file with the failing formula and a macro to copy it in the failing place

be careful with macros from others, 

either perform the steps manually - copy A3:B3, paste in A1:B3, delete value in A2, check B2 and B3 displaying wrong results, 

or have a look in the code only doing simple things inside the sheet and then fire the macro by pressing the button,
Comment 2 b. 2019-12-21 17:23:34 UTC
if the button - macro doesn't work it may be: 

document opened read only, 

<view-toolbars-form controls> stuck in edit mode ...
Comment 3 b. 2019-12-21 18:52:28 UTC
"i think it's already adressed in another case treating nested formulas," 

could be that: 

https://bugs.documentfoundation.org/show_bug.cgi?id=128975

already boiled down to simple case there ...
Comment 4 Xisco Faulí 2020-02-17 13:57:34 UTC
Regression introduced by https://cgit.freedesktop.org/libreoffice/core/commit/?id=6b4e767cf90e4f384a6b3dd43aa51c09b500b97b, which is the same commit as in bug 128975. Closing as RESOLVED DUPLICATED

*** This bug has been marked as a duplicate of bug 128975 ***
Comment 5 b. 2020-02-17 20:04:44 UTC
@xisco: 

i'd suggest to change bug / duplicate between 128975 and 129541, 129541 is the simpler shorter sample which will make it easier for dev's to deal with it ... 

reg. 

b.