Summary: | "Expand references when new columns/rows are inserted" spoils SUM formula when row is inserted | ||
---|---|---|---|
Product: | LibreOffice | Reporter: | Tom <Tom583> |
Component: | Spreadsheet | Assignee: | Not Assigned <libreoffice-bugs> |
Status: | NEW --- | QA Contact: | |
Severity: | major | ||
Priority: | high | CC: | fdbugs, gerard.fargeot, jbfaure, jmadero.dev |
Version: | 4.2.0.0.beta1 | Keywords: | bisected, regression |
Hardware: | All | ||
OS: | Linux (All) | ||
Whiteboard: | bibisected | ||
i915 platform: | i915 features: | ||
Attachments: | LO Calc 4.2.5.2 example |
Reproduced with LO 4.3.0.3, 4.2.6.1, 4.2.0.0.beta1 on Ubuntu 12.04 x86 Not reproduced with LO 4.1.6.2 e02439a3d6297a1f5334fa558ddec5ef4212c574 is the first bad commit commit e02439a3d6297a1f5334fa558ddec5ef4212c574 Author: Bjoern Michaelsen <bjoern.michaelsen@canonical.com> Date: Thu Oct 17 10:15:34 2013 +0000 source-hash-6b8393474974d2af7a2cb3c47b3d5c081b550bdb commit 6b8393474974d2af7a2cb3c47b3d5c081b550bdb Author: Luboš Luňák <l.lunak@suse.cz> AuthorDate: Tue Jun 4 13:24:59 2013 +0200 Commit: Luboš Luňák <l.lunak@suse.cz> CommitDate: Wed Jun 5 16:01:43 2013 +0200 fix gcc inline assembler operands usage Apparently whoever did these didn't get the gcc docs and specified every operand only as input, and then added volatile, explicit initialization and what not until it worked. Specify output operands correctly instead. I couldn't verify all assembler variants, as I don't know them, but the ones I don't know had at least some proper usage of output operands, so I'll assume those are all correct. Change-Id: I2910308b5e00cce8db756496df50ed26cfe35bb6 :100644 100644 49c023e2d36621396be9608a0a64bb670ac44e56 0057ddf8fa7aac0cc36b2b4cc3d47cb3b222e3de M ccache.log :100644 100644 1f724a450a2ba6123dc2772c4779157757705897 072f6720e641bd0f70bebcb3c47cfc6a55159e50 M commitmsg :100644 100644 058e9f056380574e4073b082180e87b9646d4e8b ecff2407bcf7e1bbf59642b9e021df82e44f526c M dev-install.log :100644 100644 115a24f6a9e45b293b22377dc409001a05069581 2b2cbad4f516bc1d6abed9e776c86ea90e3e757f M make.log :040000 040000 2d7a96e6b40994d4cff6f226ff2d54947490dabc 4ca7969f1b45eb47375749fa2b77e239eeb9e77b M opt # bad: [423a84c4f7068853974887d98442bc2a2d0cc91b] source-hash-c15927f20d4727c3b8de68497b6949e72f9e6e9e # good: [65fd30f5cb4cdd37995a33420ed8273c0a29bf00] source-hash-d6cde02dbce8c28c6af836e2dc1120f8a6ef9932 git bisect start 'latest' 'oldest' # bad: [e02439a3d6297a1f5334fa558ddec5ef4212c574] source-hash-6b8393474974d2af7a2cb3c47b3d5c081b550bdb git bisect bad e02439a3d6297a1f5334fa558ddec5ef4212c574 # good: [8f4aeaad2f65d656328a451154142bb82efa4327] source-hash-1885266f274575327cdeee9852945a3e91f32f15 git bisect good 8f4aeaad2f65d656328a451154142bb82efa4327 # good: [9995fae0d8a24ce31bcb5e9cd0459b69cfbf7a02] source-hash-8600bc24bbc9029e92bea6102bff2921bc10b33e git bisect good 9995fae0d8a24ce31bcb5e9cd0459b69cfbf7a02 # good: [8ad82bc1416a07501651e8d96fe268e47d3931d3] source-hash-13821254f88d2c5488fba9fe6393dcf4ae810db4 git bisect good 8ad82bc1416a07501651e8d96fe268e47d3931d3 # good: [d084d250b04446535ca1d7c29cf2062e6bd042b3] source-hash-688f72e3a2c3ef923389bbd21f6aea3afe1114db git bisect good d084d250b04446535ca1d7c29cf2062e6bd042b3 # good: [c2069a369d738078124812312d51f21ea1ce2421] source-hash-f160e4935c474a5293b3d3c11b3d538efb4767a0 git bisect good c2069a369d738078124812312d51f21ea1ce2421 # good: [a0f20bc04a32a7791ba765d2de2f44f1b74033d1] source-hash-1de66ba440855050a794b3b2a8647c1b02c210b8 git bisect good a0f20bc04a32a7791ba765d2de2f44f1b74033d1 # good: [a48fbf799e4d4d555fe383b7233c804f573eca4e] source-hash-bb6ecd8b40313b7cc83d4e619029f4e001334a52 git bisect good a48fbf799e4d4d555fe383b7233c804f573eca4e # good: [e7cbc9b8764280fd4799e234ca32925e91547a82] source-hash-31b35ed6bb7fe77f3f276b00fefce112a620b6ac git bisect good e7cbc9b8764280fd4799e234ca32925e91547a82 # first bad commit: [e02439a3d6297a1f5334fa558ddec5ef4212c574] source-hash-6b8393474974d2af7a2cb3c47b3d5c081b550bdb There is obviously a change, but where is the bug or incorrect behavior? In the current version or in the previous? Best regards. JBF (In reply to comment #3) > There is obviously a change, but where is the bug or incorrect behavior? In > the current version or in the previous? > > Best regards. JBF When inserting new row with "Expand references ...." enabled, References in formulae must be expanded if: + new row is adjacent to references + original reference contains at least 2 rows. Example 1: Formula =SUM(A2:B3)and insert row before row 2 make formula becomes =SUM(A2:B4) (there is 2 rows in the SUM range) Example 2: Formula =SUM(A2:B2)and insert row before row 2 make formula stay =SUM(A2:B2) (there is 1 row in the SUM range) Plus : this bug change also references when row inserted is *not* adjacent to the range. Inserting a new row in row 2 with formula =SUM(A5:B5) in C5 change the formula in =SUM(A5:B6) Same behaviour for columns. This is indeed a major bug. still broken in all 4.2.x, 4.3.x and 4.4 master. This basic functionality forces me to stay in 4.1.6 I get a different result for the bibisect which ends here in 43all: # first bad commit: [ba096f438393091574da98fe7b8e6b05182a8971] source-hash-8499e78ca03c792f4fa2650e02b519094ba0baa8 It looks like the behaviour changed as of the below commit. commit 27d02ddc7fc4e11bc4429839729ea3d6f7322245 Author: Kohei Yoshida <kohei.yoshida@gmail.com> Date: Fri Jul 19 20:31:08 2013 -0400 Properly handle optional edge expansion of referenced ranges. Change-Id: I499189f4f76eee4b963f643364d1fad26cf69785 |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
Created attachment 103318 [details] LO Calc 4.2.5.2 example If an empty row is inserted in a consecutive row of formulas that use the built-in function "SUM" with a range argument while the option "Expand references when new columns/rows are inserted" is enabled, the formula in the rows below the inserted row are spoiled. Example attached: 1) 1st sheet = initial spreadsheet 2) 2nd sheet = result after adding row 6 while option "Expand ..." is ENabled The formulas in D7, D8, D9, D10 and D11 are wrong. E.g. "D7" should read "=SUM(A7:C7). The first argument of SUM() was obviously not incremented. 3) 3rd sheet = result after adding row 6 while option "Expand ..." is DISabled The formulas in D7, D8, D9, D10 and D11 are correct. LibreOffice Calc 4.2.4.2 has the same problem but LibreOffice Calc 4.1.6.2 does not have this problem. Regards Tom