Created attachment 122021 [details] Test sheet with the sum error We have a spreadsheet which is summing up for us. But now in version 5.1.0.1 on one of our spreadsheet id does not work any longer. It works correctly in version 5.0.4.2. What is happening: In column I we have a sum function. It shows no summing. Expected: Column I should be automatically be summed up. Note: When marking a row from C -H the sum is shown on the reporting line in the bottom of the sheet. Please find attached a sheet with the data.
I can confirm this on 5.1.0.1 Ubuntu 15.04. The sum function works if you narrow it down to 5 from =SUM(C88:H88) to =SUM(D88:H88) I've tested the same file on LO: 4.4.6.3 and I got hidden results for sum function that I can reshow it by modifying any cell in the sum range
Perhaps you have a different set up for: Menu/Tools/Options/LibreOffice calc/Formula - Recalculation on file load. on 5.0 than on 5.1. With always recalculate works for me.
(In reply to m.a.riosv from comment #2) > Perhaps you have a different set up for: > Menu/Tools/Options/LibreOffice calc/Formula - Recalculation on file load. > on 5.0 than on 5.1. With always recalculate works for me. It doesn't make any difference. F9 does not yield calculation as well as CTRL+SHIFT+F9.
I can reproduce with 5.2.0 alpha0. Seems to be related to frozen columns: If I open the file with 5.0.4 and unfreeze rows and columns (I also made a simple modification, like writting 1 in D1 to be sure the Windows > Freeze is retained after saving), then at opening with 5.2.0 result are correctly showned.
bibisected : open file, ctrl+shift+f9 BAD - no results in column I GOOD - results in column I This seems to have begun at the below commit. Adding Cc: to Noel Grandin; Could you possibly take a look at this one? Thanks 3567c3e4e3fd549d1e71b0db623e6b64632b0f02 is the first bad commit commit 3567c3e4e3fd549d1e71b0db623e6b64632b0f02 Author: Norbert Thiebaud <nthiebaud@gmail.com> Date: Mon Dec 7 04:08:14 2015 -0800 source 7e9bdfe60784d49ab7d6ffe8df256834d39e1234 source 7e9bdfe60784d49ab7d6ffe8df256834d39e1234 :040000 040000 a33489d12653fc8a8f438b3d70ca83aa528dc1d3 11a5f8cac4a70bb02a671124b12a757dd76c54d5 M instdir author Noel Grandin <noel@peralex.com> 2015-12-01 12:55:35 (GMT) committer Noel Grandin <noel@peralex.com> 2015-12-02 06:47:30 (GMT) commit 7e9bdfe60784d49ab7d6ffe8df256834d39e1234 (patch)
That commit was reverted shortly thereafter in a38110216c895e876ab1b4697da74f559d06a3cb
Bug NOT present in Version: 5.0.5.1 Build ID: 7609023f63524a6c8326f6c82e7e23f55a5b7bb5 Locale: en-US (en_US.UTF-8)
I can still repro Version: 5.2.0.0.alpha0+ Build ID: e1367aaf7d9cc93bbbe3ba64015fde3469c61c6e CPU Threads: 4; OS Version: Linux 4.2; UI Render: default; TinderBox: Linux-rpm_deb-x86_64@70-TDF, Branch:master, Time: 2016-01-20_23:32:01
@raal: This is "bibisected", but not "bisected" yet. Am I right?
(In reply to Kevin Suo from comment #9) > @raal: This is "bibisected", but not "bisected" yet. Am I right? bibisected = range of commits bisected = we know one commit but in this case the culprit commit was reverted, so I'm deleting the keywords.
This seems to have begun at the below commit. Adding Cc: to Jan Holesovsky ; Could you possibly take a look at this one? Thanks bibisect-win32-5.2 - bibisected from reverted commit a38110216c895e876ab1b4697da74f559d06a3cb git bisect start 4d670431854eba60408c58fc2cdd700aeec8ec14 061e3009569ef8cf4720098760f3564019ef4196 0e1c7b8968a73e15cd6b03dc5a99b31aa9c86f9a is the first bad commit commit 0e1c7b8968a73e15cd6b03dc5a99b31aa9c86f9a Author: Norbert Thiebaud <nthiebaud@gmail.com> Date: Wed Dec 9 00:28:54 2015 -0800 source 0bf39de7e54f041ade0f659db47cabcc6589ada7 author Jan Holesovsky <kendy@collabora.com> 2015-11-30 11:32:16 (GMT) committer Jan Holesovsky <kendy@collabora.com> 2015-12-03 15:14:03 (GMT) commit 0bf39de7e54f041ade0f659db47cabcc6589ada7 (patch)
Same phenomenon here as in https://bugs.documentfoundation.org/show_bug.cgi?id=97369#c30 , inserting and deleting a row in a "suitable" place, and then forcing a recalc, helps. Just for developer information, not intended as a useful workaround for end-users.
Bugs seems not to be present in Version: 5.1.1.1 Build ID: c43cb650e9c145b181321ea547d38296db70f36e CPU Threads: 2; OS Version: Linux 4.2; UI Render: default; Locale: en-US (en_US.UTF-8) Please verify
Same symptoms as in bug 97369 and the fix for bug 97369 cures the problem here. So duplicate. Best regards. JBF *** This bug has been marked as a duplicate of bug 97369 ***