Download it now!
Bug 31231 - Documents with charts open with the document set as already modified
Summary: Documents with charts open with the document set as already modified
Status: NEW
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Calc (show other bugs)
(earliest affected)
Inherited From OOo
Hardware: All All
: low minor
Assignee: Not Assigned
Whiteboard: target:6.0.0
Keywords: bibisected, bisected, regression
: 113689 114121 114547 (view as bug list)
Depends on:
Blocks: Chart Save ModifiedStatus
  Show dependency treegraph
Reported: 2010-10-30 03:30 UTC by Michel Rudelle
Modified: 2019-04-20 16:09 UTC (History)
12 users (show)

See Also:
Crash report or crash signature:

Example of a defective file (50.15 KB, application/vnd.oasis.opendocument.spreadsheet)
2010-10-30 03:45 UTC, Michel Rudelle
Example of a correct file (37.03 KB, application/vnd.oasis.opendocument.spreadsheet)
2010-10-30 03:46 UTC, Michel Rudelle
Various case showing the relation of the function INDIRECT with the issue (42.33 KB, application/zip)
2010-11-01 10:05 UTC, Michel Rudelle

Note You need to log in before you can comment on or make changes to this bug.
Description Michel Rudelle 2010-10-30 03:30:18 UTC
For some Calc files (always the same), the save button is active without having made any modification.
So, if I just open a file to read it, I can't close it without obtaining the message saying that the document was modified.
I join two documents examples (with and without this default) with a slight difference: the choice of different data for one of the curves of the last sheet

This problem is reproducible with the same files as well with OOo 3.2 OOo 3.3 and LibO, and on two OS (Vista and 7).
Comment 1 Michel Rudelle 2010-10-30 03:45:01 UTC
Created attachment 39898 [details]
Example of a defective file

Just open it and look at the save button (bad)
Comment 2 Michel Rudelle 2010-10-30 03:46:54 UTC
Created attachment 39899 [details]
Example of a correct file

Just open it and look at the save button (good)
Comment 3 Michel Rudelle 2010-11-01 10:05:58 UTC
Created attachment 39960 [details]
Various case showing the relation of the function INDIRECT with the issue
Comment 4 Michel Rudelle 2010-11-01 10:08:01 UTC
Here are observations which can help:

The bug comes from the use of the function INDIRECT together with the use of the results in a graph (joined example : “modif-bad”) - a similar file without the function INDIRECT does not raise problem (joined example “modif-ok1”)

The use of results of the function INDIRECT in calculations and others functions does not create the problem (joined example : “modif-ok”).
Comment 5 Kohei Yoshida 2010-11-02 05:25:21 UTC
Let me look into this.
Comment 6 Kohei Yoshida 2010-11-02 06:48:36 UTC
Unfortunately there are gazillion places that set a document modified flag during import.  I've only taken a cursory look at it, but the prospect doesn't look too good.
Comment 7 Kohei Yoshida 2010-11-02 08:33:53 UTC
Also reported in Oracle's issue tracker
Comment 8 tester8 2011-06-12 13:19:50 UTC
NOT reproduced with
Ubuntu 10.04.2 x86
LO 3.4
Comment 9 Michel Rudelle 2011-06-12 13:55:18 UTC
(In reply to comment #8)
> NOT reproduced with
> Ubuntu 10.04.2 x86
> LO 3.4

Indeed, it seems that this issue is fixed with Libo 3.4
Comment 10 Michel Rudelle 2012-03-11 08:02:47 UTC
Unfortunately, this bug which was fixed in version 3.4, comes back in version 3.5!
The behaviour is exactly the same with files previously joined.
I hope really that this can help you, because it is very confusing.
[LibO 3.5.1 rc2]
Comment 11 Roman Eisele 2012-05-03 05:10:22 UTC
Changed 'Component' to 'Spreadsheet', therefore 'Summary' abbreviated.
Comment 12 Joel Madero 2013-07-03 01:22:54 UTC
Seems fixed in version 4.0.4 so marking as WFM - if this is still an issue for you in version 4 mark as NEW again. Thanks!
Comment 13 Michel Rudelle 2013-07-05 09:46:29 UTC
Thanks to follow this issue
You are right, It's Ok with
but still wrong with:
Version: Build ID: 1b3956717a60d6ac35b133d7b0a0f5eb55e9155
Version: Build ID: 5b9bad7482a98f2d0d37c4b75a13292abe653ea3
[All tests performed with attachments on Vista-32b]
Therefore I reopen the bug
Comment 14 Jean-Baptiste Faure 2014-05-01 21:03:13 UTC
Works as expected in LibO but not in LO 4.1.6 (generic Linux version 64 bits) and in the master (Build ID: d1ffea3a272d61f1343047e86b4eeabe032aa352)

Best regards. JBF
Comment 15 Joel Madero 2014-11-03 03:23:58 UTC
As there is confusion as to if this bug is still reproducible, moving back to UNCONFIRMED to let QA team try to get a handle on what's going on. Thanks
Comment 16 Buovjaga 2014-11-15 17:19:23 UTC
Tested attachment 39898 [details].
On Windows and 4.4 alpha, the problem doesn't exist.
On Ubuntu 4.4 alpha, and Windows, the save button is active and it prompts to save on close.

Win 7 64-bit and Version:
Build ID: b021b5983c62e266b82d9f0c5c6d8d8900553827
TinderBox: Win-x86@39, Branch:master, Time: 2014-11-12_01:10:08

Ubuntu 14.10 64-bit Version:
Build ID: 5bff4b016c4b44f4123e0e6a4fd4c0c4dc0cfa2d
TinderBox: Linux-rpm_deb-x86_64@46-TDF-dbg, Branch:master, Time: 2014-11-13_00:14:29


Build ID: 430m0(Build:2)
Comment 17 Jean-Baptiste Faure 2014-11-15 18:21:55 UTC
Not reproducible with my own build of master (Version:
Build ID: 6b30907a926890f835c094a5afdf4c0e6d8a1d19) under Ubuntu 14.10 x86-64

But reproducible with my own build of (Build ID: 547d613e0868726e602ccb00fca0e7518a6c4bee) under Ubuntu 14.10 x86-64

Best regards. JBF
Comment 18 Matthew Francis 2015-04-06 05:28:44 UTC
Reproduced on Linux with OOo 3.3.0, LO 3.3.0 and current 4.5 master (5cfee95f4e611911a862788370402e861531e9c1)

-> Version: Inherited from OOo
-> Platform: All
Comment 19 tommy27 2016-04-16 07:23:26 UTC Comment hidden (obsolete)
Comment 20 Jean-Baptiste Faure 2016-04-17 06:35:41 UTC
In LO, both testfiles are tagged modified as soon as they are opened, no matter the answer I give to the question if all formula must be recalculated because the file has been saved by another application than LibreOffice.

Best regards. JBF
Comment 21 QA Administrators 2017-05-22 13:23:23 UTC Comment hidden (obsolete)
Comment 22 Jean-Baptiste Faure 2017-06-10 20:27:17 UTC
Both files open as modified in LO 5.3.3 from Ubuntu PPA, LO built at home and LO built at home.
Ubuntu 16.04 x86-64

Best regards. JBF
Comment 23 Yousuf Philips (jay) (retired) 2017-09-21 00:30:01 UTC
So this issue happens with any document that has a chart in it, likely because it needs to refresh the image rendering of the chart in the document, as can be tested with any of the below files.

(In reply to Michel Rudelle from comment #2)
> Created attachment 39899 [details]
> Example of a correct file
> Just open it and look at the save button (good)

Opening this file in master has the save button enabled (aka bad). It went from good to bad in 5.0.

Build ID: 7315f325ff7ada3d6bd85a471058fdaeaff8cdb0
CPU threads: 2; OS: Linux 4.4; UI render: default; VCL: gtk2; 
TinderBox: Linux-rpm_deb-x86_64@70-TDF, Branch:master, Time: 2017-09-17_06:58:21
Locale: en-US (en_US.UTF-8); Calc: group

@Eike: Any thoughts?
Comment 24 Mike Kaganski 2017-11-07 05:57:29 UTC
Interesting how this bug, opened for pre-LO versions, then reported to be WFM twice, is now used for problem that had begun for 5.0 after being OK in 4.4...

Well, however, will post the bibisect here.

$ git bisect log
# bad: [b7988d11e5d3751a4b366b2bfc9048f7a30e8526] source sha:87ac0b1e75a880a68ecb748bd4b34ae5a3d2ae98
# good: [f449493ae11ac76cc7396bddeaa624a60c565936] source sha:57d6b92b69a31260dea0d84fcd1fc5866ada7adb
git bisect start 'master' 'oldest'
# bad: [66e2ae767eb4bb83444e3d03bcb90adcbe6d4991] source sha:5a308b1239a09417507b0d05090ff2d3418d5133
git bisect bad 66e2ae767eb4bb83444e3d03bcb90adcbe6d4991
# bad: [90c1dbb098a6d957f2293692716251ee5a6053ca] source sha:2813632238380e0bfe40c0e6404a07102cde1398
git bisect bad 90c1dbb098a6d957f2293692716251ee5a6053ca
# good: [5b8e174eb7b3d996d6c90862d7228e1b928a9787] source sha:4de09a9efdb62cf90ce18662852e556cf7148e14
git bisect good 5b8e174eb7b3d996d6c90862d7228e1b928a9787
# good: [5d530472220267ad6588480f7a67a8f28f83d070] source sha:ff2990fded4f1745c44f62cca53e3bff28e371fb
git bisect good 5d530472220267ad6588480f7a67a8f28f83d070
# bad: [1f0e00b205dd7bd5e9d2cd0ce1e61fcbb7da78ac] source sha:7b1261f6f956271ec2a545f635e11432a5e64fa1
git bisect bad 1f0e00b205dd7bd5e9d2cd0ce1e61fcbb7da78ac
# good: [95352698710d22573b9f43c92fddc73a48f5f64e] source sha:8d98f035cf6c410d4b29bd58d86cfe86127db68b
git bisect good 95352698710d22573b9f43c92fddc73a48f5f64e
# bad: [6d37964763973590a141432004ce89ed14fbdb02] source sha:6123d6a9fbb268f823224d054cb0fe215aa3015a
git bisect bad 6d37964763973590a141432004ce89ed14fbdb02
# good: [06e434742b9befe90f6c1f0cd5c1abacab00a868] source sha:de7a5eb44f942054c42157bf9243fb89a1d39368
git bisect good 06e434742b9befe90f6c1f0cd5c1abacab00a868
# bad: [a046e35d5a4e4e62978ee091be9c7fdd8d6eba68] source sha:1b5c8e4a031af17c47a2900da09c1db1df1242df
git bisect bad a046e35d5a4e4e62978ee091be9c7fdd8d6eba68
# bad: [ada1916a118c4aa6cc39722e39c885edd8120480] source sha:c29657e0d6bb707345584ac7a7f5ae5016f37297
git bisect bad ada1916a118c4aa6cc39722e39c885edd8120480
# good: [ba326ca77713f2080a89e95181afdbdc2e11039f] source sha:d1ce6effb62745e9f9c85badd44a1be0aa4a1d8f
git bisect good ba326ca77713f2080a89e95181afdbdc2e11039f
# bad: [75fba066c605374e1d4bb2a9f15b81e2eef192e9] source sha:dc28e90d200a839d4017d548217ee5ce8a23f848
git bisect bad 75fba066c605374e1d4bb2a9f15b81e2eef192e9
# good: [870b1afa9d74304494e209b0bd9f24dfe4127bef] source sha:465356ecc81e23016b4289ab16e99084f2a7b84e
git bisect good 870b1afa9d74304494e209b0bd9f24dfe4127bef
# first bad commit: [75fba066c605374e1d4bb2a9f15b81e2eef192e9] source sha:dc28e90d200a839d4017d548217ee5ce8a23f848

author	Noel Grandin <>	2014-12-25 13:17:55 (GMT)
committer	Noel Grandin <>	2015-01-06 08:59:41 (GMT)
commit	dc28e90d200a839d4017d548217ee5ce8a23f848
tree	a6ae872fb19a046292d96d280da286a20a397def
parent	465356ecc81e23016b4289ab16e99084f2a7b84e
fdo#84938: convert IMPORT_ constants to 'enum class'
Comment 25 Mike Kaganski 2017-11-07 07:19:52 UTC
Since it's unclear to me why could that commit cause this problem, I re-bibisected on Ubuntu:

~/bibisect-50max$ git bisect log
# bad: [dda106fd616b7c0b8dc2370f6f1184501b01a49e] source-hash-0db96caf0fcce09b87621c11b584a6d81cc7df86
# good: [5b9dd620df316345477f0b6e6c9ed8ada7b6c091] source-hash-2851ce5afd0f37764cbbc2c2a9a63c7adc844311
git bisect start 'latest' 'oldest'
# bad: [0c30a2c797b249d0cd804cb71554946e2276b557] source-hash-45aaec8206182c16025cbcb20651ddbdf558b95d
git bisect bad 0c30a2c797b249d0cd804cb71554946e2276b557
# bad: [770ff0d1a74d2450c2decb349b62c5087e12c46b] source-hash-549b7fad48bb9ddcba7dfa92daea6ce917853a03
git bisect bad 770ff0d1a74d2450c2decb349b62c5087e12c46b
# good: [227af65db5e34efcf8dcb0b53333efecd30f37f8] source-hash-193c7ba9be48f00b46f9e789f233db577e7b3303
git bisect good 227af65db5e34efcf8dcb0b53333efecd30f37f8
# good: [78b395d05689a5207f2ec4cc29ec296d64076a96] source-hash-a2e4be6ded508030a6c2a33919cbe8cb504382e0
git bisect good 78b395d05689a5207f2ec4cc29ec296d64076a96
# good: [8dd6442885c969ae43ae5ff9ddfc53c9f04a9c27] source-hash-d07f0997c54e9cef31d996ebeb2aabfb4b4e0265
git bisect good 8dd6442885c969ae43ae5ff9ddfc53c9f04a9c27
# good: [56e2ff1d44b7bcd4fff6ce86c93fd9b666808d0b] source-hash-d7794d2584cd5d476b011b5344c77ad59c179c58
git bisect good 56e2ff1d44b7bcd4fff6ce86c93fd9b666808d0b
# good: [4f5bf35bab35f426249aa06c92be46174890a379] source-hash-df5fa4082cfb17c5d5be6678995689485df6d429
git bisect good 4f5bf35bab35f426249aa06c92be46174890a379
# good: [086887d27fcae1b0e65a150051c179e27721a341] source-hash-2eb4bd3c8ce8d4ac76680e5179364b12a656ae94
git bisect good 086887d27fcae1b0e65a150051c179e27721a341
# good: [0889bdba9a9cc0a9aa801dc90fc3ec47004ba6a5] source-hash-d1ce6effb62745e9f9c85badd44a1be0aa4a1d8f
git bisect good 0889bdba9a9cc0a9aa801dc90fc3ec47004ba6a5
# bad: [6dca089c8fd025a2fe843d886635b7ba649d8450] source-hash-abe670157b69aa7fe4b478f1fd13757d7b7fcc4b
git bisect bad 6dca089c8fd025a2fe843d886635b7ba649d8450
# bad: [eb18d55b7b5df079783a35286333fcda73870115] source-hash-c29657e0d6bb707345584ac7a7f5ae5016f37297
git bisect bad eb18d55b7b5df079783a35286333fcda73870115
# bad: [0f0b8c55f1bd241bd2cc274907413ebec6304c39] source-hash-dc28e90d200a839d4017d548217ee5ce8a23f848
git bisect bad 0f0b8c55f1bd241bd2cc274907413ebec6304c39
# good: [851ab1111c5e731876d919e84f099b6aa293f4fb] source-hash-465356ecc81e23016b4289ab16e99084f2a7b84e
git bisect good 851ab1111c5e731876d919e84f099b6aa293f4fb
# first bad commit: [0f0b8c55f1bd241bd2cc274907413ebec6304c39] source-hash-dc28e90d200a839d4017d548217ee5ce8a23f848

So, the result is the same.
Comment 26 Mike Kaganski 2017-11-07 07:34:48 UTC
Bug 86321 comment 29 mentioned that this commit had fixed an issue with a diagram not updated on source data edit.
Comment 27 Mike Kaganski 2017-11-07 10:40:12 UTC
Well, on the second glance, it turned out that the commit did change logic in one place: sc/source/filter/xml/XMLTableShapeResizer.cxx. Given that SvXMLImportFlags::ALL is a bitmask 0xffff, the new code started to check if any bit is set, instead of checking for exact match.

A patch is This is for the regression in Calc. The Impress thing seems to have not working at the time of the regression, so should be fixed separately. Please create a proper issue for it. (This issue looks invalid anyway.)
Comment 28 Commit Notification 2017-11-07 13:27:42 UTC
Mike Kaganski committed a patch related to this issue.
It has been pushed to "master":

tdf#31231: properly check for SvXMLImportFlags::ALL

It will be available in 6.0.0.

The patch should be included in the daily builds available at in the next 24-48 hours. More
information about daily builds can be found at:

Affected users are encouraged to test the fix and report feedback.
Comment 29 Eike Rathke 2017-11-07 13:52:46 UTC
*** Bug 113689 has been marked as a duplicate of this bug. ***
Comment 30 Buovjaga 2017-11-10 11:17:12 UTC
For what it's worth, defaut.ods still opens displaying the modified dot on the Save button.

Arch Linux 64-bit, KDE Plasma 5
Build ID: 1aba1955f161cc112dab80b6b3e78ec7761616fc
CPU threads: 8; OS: Linux 4.13; UI render: default; VCL: kde4; 
Locale: fi-FI (fi_FI.UTF-8); Calc: group
Built on November 10th 2017
Comment 31 Jacques Guilleron 2017-11-29 11:21:14 UTC
*** Bug 114121 has been marked as a duplicate of this bug. ***
Comment 32 Jacques Guilleron 2017-12-19 16:35:21 UTC
*** Bug 114547 has been marked as a duplicate of this bug. ***
Comment 33 Xisco Faulí 2018-03-20 10:50:01 UTC
Dear Mike Kaganski,
This bug has been in ASSIGNED status for more than 3 months without any
activity. Resetting it to NEW.
Please assigned it back to yourself if you're still working on this.
Comment 34 Xisco Faulí 2018-06-05 19:30:04 UTC
Adding Cc: to Noel Grandin
Comment 35 robert 2019-04-20 08:06:26 UTC
Bug still present:

Version: (x64)
Build ID: aecc05fe267cc68dde00352a451aa867b3b546ac
CPU threads: 8; OS: Windows 6.1; UI render: default; VCL: win; 
Locale: en-GB (en_GB); UI-Language: en-US
Calc: threaded
Comment 36 Michel Rudelle 2019-04-20 16:09:46 UTC
I confirm, this bug is still present:

Build ID: aecc05fe267cc68dde00352a451aa867b3b546ac
Threads CPU : 4; OS : Linux 4.13; UI Render : par défaut; VCL: gtk3; 
Locale : fr-FR (fr_FR.UTF-8); Langue IHM : fr-FR
Calc: threaded

Ubuntu 18.04.2 LTS
Gnome 3.28.2