Description: under <7.2 the Testfile is working as espected. The conditional formatting based on a formula is used to highlight duplicate entries in color. This is used by me to prevent that e.g. orders are entered twice. Since 7.2xx this function doesn't works as before. █ last tested under:█ Version: 7.2.0.4 (x64) / LibreOffice Community Build ID: 9a9c6381e3f7a62afc1329bd359cc48accb6435b CPU threads: 12; OS: Windows 10.0 Build 22000; UI render: Skia/Vulkan; VCL: win Locale: de-DE (de_DE); UI: de-DE Calc: CL Steps to Reproduce: ► see GIF "LO7.1.5.2_working.gif" and "LO7.2.0.4_failure.gif" 1. open the testfile with LO7.1xx and it is working as before Actual Results: 1. open the testfile with >LO7.2 and see the difference Expected Results: conditional formatting should work as under <7.2 Versions Reproducible: Always User Profile Reset: Yes OpenGL enabled: Yes Additional Info: Also tested under Win10 21H1 19043.1165 and different 7.2.0 RC Versions
Created attachment 174400 [details] LO7.2.0.4_failure under 7.2RCx
Created attachment 174401 [details] conditional formatting under 7.1.5.2 was working fine
Created attachment 174402 [details] TESTFILE to play arround for better understanding of the problem
Confirm with Version: 7.2.0.4 (x64) / LibreOffice Community Build ID: 9a9c6381e3f7a62afc1329bd359cc48accb6435b CPU threads: 4; OS: Windows 10.0 Build 19043; UI render: Skia/Raster; VCL: win Locale: ro-RO (ro_RO); UI: en-US Calc: threaded
Created attachment 174420 [details] Sample file The issue it's not related with the conditional format, seems the problem it's that COUNTIF() taking care of the cell format. In the attached file, cells Sheet1.A16:B25 has a format "BA "Estandar, on the second sheet the format is Estandar, with LO 7.2 gives different results in E2:F11 and C15:C25 between the two sheets, and the only difference it's the mentioned format. Do a hard recalc on 7.2 to see the issue.
I don't think it is a new default behavior. It looks like a critical issue.
(In reply to m.a.riosv from comment #5) > Created attachment 174420 [details] > Sample file > > The issue it's not related with the conditional format, seems the problem > it's that COUNTIF() taking care of the cell format. > > In the attached file, cells Sheet1.A16:B25 has a format "BA "Estandar, on > the second sheet the format is Estandar, with LO 7.2 gives different results > in E2:F11 and C15:C25 between the two sheets, and the only difference it's > the mentioned format. > Do a hard recalc on 7.2 to see the issue. Hi (Matrisov)?, yeah, you are right! When I change the cell formating to "normal" decimal the conditionaL formating the conditional formating is workiung again. But I woul like to keep the asigned "BA " in front of our "BA" → is our "Job Number"... The Solution for me is in the moment to switch the format-code to: ["BA "0] instead of ["BA "Standard] → so it is possible for me to work as before but with the newer Version (I love the Autofilter update ☺ ) Thanks again for your advice
Let's change to critical, l think it's unacceptable calculations being affected by the cell format.
On pc Debian x86-64 with master sources updated today, I confirm this. (I did a hard recalc on Mario's file and have different results between both sheets).
Bibisected with linux-64-7.2 to https://git.libreoffice.org/core/commit/3069df790cca2917e5aedd87bac1af65f9605d51 tdf#142910 sc filter: fix "greater than" or "smaller than" etc Adding Cc: to Balazs Varga
Created attachment 174644 [details] attachment 174420 [details] in 7.1.0 and current master after a hard recalculate Several values indeed change after a hard recalculate in: Version: 7.3.0.0.alpha0+ (x64) / LibreOffice Community Build ID: d350a1364a1c34b96d00f2f716c44882b7b57fe9 CPU threads: 4; OS: Windows 10.0 Build 18363; UI render: Skia/Raster; VCL: win Locale: en-US (hu_HU); UI: en-US Calc: CL
Seeing that change mentioned in comment 10 uses ScDocument::RoundValueAsShown() this is fixed with the commit for bug 144209 https://git.libreoffice.org/core/commit/71b003a12f8afdff42a25786ad0a12ddd6609d59 I tried the "Sample file" https://bugs.documentfoundation.org/attachment.cgi?id=174420 and results after Shift+Ctrl+F9 hard recalc are the same as in 7.1.5. Note that E2 =COUNTIFS($A$2:$B$30;A2)>1 loads as (however) stored TRUE in both versions and is recalculated to FALSE, which is correct, the value 1 occurs only one time.
Note the pending reviews https://gerrit.libreoffice.org/c/core/+/121459 for 7-2 https://gerrit.libreoffice.org/c/core/+/121460 for 7-2-1