gnumeric issueshttps://gitlab.gnome.org/GNOME/gnumeric/-/issues2018-05-22T13:58:12Zhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/217field labels in Preferences dialog should use sentence capitalization2018-05-22T13:58:12ZBugzillafield labels in Preferences dialog should use sentence capitalization## Submitted by Adam Dingle
**[Link to original bug (#698672)](https://bugzilla.gnome.org/show_bug.cgi?id=698672)**
## Description
According to the GNOME interface guidelines, checkbox and field labels should use sentence capitaliza...## Submitted by Adam Dingle
**[Link to original bug (#698672)](https://bugzilla.gnome.org/show_bug.cgi?id=698672)**
## Description
According to the GNOME interface guidelines, checkbox and field labels should use sentence capitalization, i.e. only the first word is capitalized:
https://developer.gnome.org/hig-book/3.8/design-text-labels.html.en#layout-capitalization
In Gnumeric's preferences dialog many labels use header capitalization (e.g. "Default Compression Level For Gnumeric Files"). We should fix these.
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/216multiple bugs in the fourier transform function; aka FFT; involves complex n...2018-05-22T13:57:55ZBugzillamultiple bugs in the fourier transform function; aka FFT; involves complex numbers## Submitted by John Denker
**[Link to original bug (#696418)](https://bugzilla.gnome.org/show_bug.cgi?id=696418)**
## Description
Created attachment 239575
demonstrate multiple related bugs in the fourier(...) function
In the atta...## Submitted by John Denker
**[Link to original bug (#696418)](https://bugzilla.gnome.org/show_bug.cgi?id=696418)**
## Description
Created attachment 239575
demonstrate multiple related bugs in the fourier(...) function
In the attached example, one sheet contains live formulas, while the
other sheet shows the results I obtain chez moi, pasted by value.
Experience suggests this may help clarify the discussion, insofar as
it makes it 100% clear what results I am obtaining, even if different
versions behave differently. I am using gnumeric version 1.10.17.
Red background indicates some of the the /primary/ errors; errors
that are obvious consequences of earlier errors are not highlighted.
In cells G7:H10 we see that the inverse FT of {1;0;0;0} is {.25;.25;.25;.25}
This is as expected.
In cells D7:E10 we see that the forward FT of {1;1;1;1} is {1;0;0;0}
which is almost as expected. The slightly weird thing is that these
numbers are left-justified in the cells. This suggests that the
spreadsheet thinks these are complex numbers. This "should" be
harmless, however ....
In cells E7:F10 we see that the inverse FT of the results of the previous
computation gets completely the wrong answer. Even though the numbers in
cells E7:E10 look numerically correct, and and even though they have exactly
zero imaginary part, they are somehow damaged. They are different from the
ordinary numbers {1;0;0;0} and the inverse FT cannot cope with them.
There is no error message and no warning message; just grossly incorrect
numerical results.
In cells G16:19 we find that adding zero to these numbers -- even using
complex arithmetic -- undoes the damage. The new numbers display right-
justified in the usual way, and the FT can cope with them. The fact
that G16:G19 displays differently from E16:19 suggests there is something
weird in the core of the complex number system.
There's more: In cells D25:E28 we see that the FT can handle an input
with *one* complex number; so far so good. However, in cells D34:E37
we see that it cannot handle an input with *two* complex numbers.
It should go without saying that a FT that cannot handle arbitrary complex
numbers on the input is pretty much useless.
Suggestion: After the FT is fixed, it would be good to test it systematically.
One fairly sensitive test is to perform a transform/inverse-transform pair
and check that we get back what we started with. On the input, use random
data, or perhaps quenched pseudo-random data. That means complex data, not just real data. A useful qualitative check is to make sure that Parseval's
theorem is upheld; see row 13 in the attachment. It would be good to include
such tests in the release-validation/regression-testing suite.
**Attachment 239575**, "demonstrate multiple related bugs in the fourier(...) function":
[fourier-bug.gnumeric](/uploads/183eb06235cd463f57b6453b7929feb2/fourier-bug.gnumeric)
Version: 1.10.xhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/215implement split panes2018-05-22T13:57:40ZBugzillaimplement split panes## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#696297)](https://bugzilla.gnome.org/show_bug.cgi?id=696297)**
## Description
The current frozen panes feature is a bit limited since you cannot scroll the top o...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#696297)](https://bugzilla.gnome.org/show_bug.cgi?id=696297)**
## Description
The current frozen panes feature is a bit limited since you cannot scroll the top or left pane and since it is not possible to move the boundary between panes.
It would be valuable to implement the split panes feature where all this is possible.
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/214Minimum of Radial axis should be different from minimum displayed of radial a...2018-05-22T13:57:31ZBugzillaMinimum of Radial axis should be different from minimum displayed of radial axis in PolarPlot## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#695830)](https://bugzilla.gnome.org/show_bug.cgi?id=695830)**
## Description
Created attachment 238852
.png image to illustrate the problem
In PolarPlots, a pa...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#695830)](https://bugzilla.gnome.org/show_bug.cgi?id=695830)**
## Description
Created attachment 238852
.png image to illustrate the problem
In PolarPlots, a particular thing happens which does not happen in XYPlots.
One needs to define the minimum of the Radial axis, which is the centre of the drawing circle. But this minimum could be different from the minimum displayed of the Radial axis.
In the attached example, which is a latitude-longitude plot around the South Pole, there are data between -80°S and -70°S, so that it would be appropriate to draw only this radial range. However, if one sets the minimum of the radial axis to -80, the graph is strongly distorted since -80 is placed at the centre of the circle.
**Attachment 238852**, ".png image to illustrate the problem":
![PolarPlotColor](/uploads/79ed9b5516e7b071b4d5e89bcb06e150/PolarPlotColor.png)
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/213Persist Gnumeric text import settings2018-05-22T13:57:14ZBugzillaPersist Gnumeric text import settings## Submitted by pmcgover
**[Link to original bug (#695588)](https://bugzilla.gnome.org/show_bug.cgi?id=695588)**
## Description
This is an enhancement request to have Gnumeric persist the settings of the last csv file import/export ...## Submitted by pmcgover
**[Link to original bug (#695588)](https://bugzilla.gnome.org/show_bug.cgi?id=695588)**
## Description
This is an enhancement request to have Gnumeric persist the settings of the last csv file import/export action. This would make Gnumeric much more productive for common tasks.
The settings should also include option to change (and persist) the default column data type. The current default column data type is "General" which will automatically convert date fields to a different date format. However, if the default data type was persisted to "Text" then the data will not be "auto-changed" when various csv files are opened and saved (exported).
Version: 1.10.xhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/212implement duplication --in the same open workbook-- of two or more sheets a...2018-05-22T13:57:05ZBugzillaimplement duplication --in the same open workbook-- of two or more sheets at once## Submitted by marc dunord
**[Link to original bug (#694748)](https://bugzilla.gnome.org/show_bug.cgi?id=694748)**
## Description
to ease work, one could implement duplication --in the same open
workbook-- of two or more sheets a...## Submitted by marc dunord
**[Link to original bug (#694748)](https://bugzilla.gnome.org/show_bug.cgi?id=694748)**
## Description
to ease work, one could implement duplication --in the same open
workbook-- of two or more sheets at once.
that would allow same-book across-sheet references to be duplicated at
once and become independent of the original worksheetshttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/211automatic scaling of Polar Plot sometimes does not work2018-05-22T13:56:44ZBugzillaautomatic scaling of Polar Plot sometimes does not work## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#693895)](https://bugzilla.gnome.org/show_bug.cgi?id=693895)**
## Description
Steps to reproduce:
- insert 120 in A1, -75 in A2, 130 in B1, -80 in B2
- select A1...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#693895)](https://bugzilla.gnome.org/show_bug.cgi?id=693895)**
## Description
Steps to reproduce:
- insert 120 in A1, -75 in A2, 130 in B1, -80 in B2
- select A1:B2
- insert a PolarPlot
=> the axes are not automatically scaled
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/210RFE: zooming of a chart with mouse scrolling2018-05-22T13:56:35ZBugzillaRFE: zooming of a chart with mouse scrolling## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#691926)](https://bugzilla.gnome.org/show_bug.cgi?id=691926)**
## Description
A chart preview could be zoomed (adjusting the axes ranges) with mouse scrolling.
W...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#691926)](https://bugzilla.gnome.org/show_bug.cgi?id=691926)**
## Description
A chart preview could be zoomed (adjusting the axes ranges) with mouse scrolling.
Would you consider adding such a feature?
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/209Sheet name incompatibility with cell reference2018-05-22T13:56:19ZBugzillaSheet name incompatibility with cell reference## Submitted by jea..@..hoo.fr
**[Link to original bug (#691845)](https://bugzilla.gnome.org/show_bug.cgi?id=691845)**
## Description
Created attachment 233586
Book with a unic sheet named "A1"
When a sheet is named as a given cell...## Submitted by jea..@..hoo.fr
**[Link to original bug (#691845)](https://bugzilla.gnome.org/show_bug.cgi?id=691845)**
## Description
Created attachment 233586
Book with a unic sheet named "A1"
When a sheet is named as a given cell reference, calculations on this cell lead to warning "[sheet name] is not allowed as defined name"
See example on attached file (Book1.gnumeric), where the unic sheet is named "A1".
The message at the following address seems related to this bug:
https://mail.gnome.org/archives/commits-list/2009-April/msg05683.html
... but I don't know how to fix it.
Thanks for your help!
**Attachment 233586**, "Book with a unic sheet named "A1"":
[Book1.gnumeric](/uploads/5f8c58b1b2014157e292d0fd420a2320/Book1.gnumeric)
Version: 1.12.xhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/208failure to account for rotation/translation in custom shape import2018-05-22T13:56:03ZBugzillafailure to account for rotation/translation in custom shape import## Submitted by an unknown user
**[Link to original bug (#690875)](https://bugzilla.gnome.org/show_bug.cgi?id=690875)**
## Description
Created attachment 232382
sample file
In the attached sample file we have for the diamond:
<dr...## Submitted by an unknown user
**[Link to original bug (#690875)](https://bugzilla.gnome.org/show_bug.cgi?id=690875)**
## Description
Created attachment 232382
sample file
In the attached sample file we have for the diamond:
<draw:custom-shape draw:z-index="0" draw:style-name="gr1" svg:width="7.241cm" svg:height="6.986cm" draw:transform="rotate (0.523598775598299) translate (8.593cm 4.352cm)">
<text:p/>
<draw:enhanced-geometry svg:viewBox="0 0 21600 21600" draw:glue-points="10800 0 0 10800 10800 21600 21600 10800" draw:text-areas="5400 5400 16200 16200" draw:type="diamond" draw:enhanced-path="M 10800 0 L 21600 10800 10800 21600 0 10800 10800 0 Z N"/>
</draw:custom-shape>
On import we ignore draw:transform="rotate (0.523598775598299) translate (8.593cm 4.352cm)", so the diamond is not correctly rotated nor translated.
Ideally, for the rotation we should be able to rotate the whole sheet object.
**Attachment 232382**, "sample file":
[U.ods](/uploads/38f0bd031c1094777f8a36641001a434/U.ods)
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/207Python function help text leak2018-05-22T13:55:52ZBugzillaPython function help text leak## Submitted by an unknown user
**[Link to original bug (#690753)](https://bugzilla.gnome.org/show_bug.cgi?id=690753)**
## Description
The function help strings created in python_function_get_gnumeric_help are clearly leaking.
It a...## Submitted by an unknown user
**[Link to original bug (#690753)](https://bugzilla.gnome.org/show_bug.cgi?id=690753)**
## Description
The function help strings created in python_function_get_gnumeric_help are clearly leaking.
It also seems that the GnmFuncHelp structures created in python_function_get_gnumeric_help as well as the perl loader functions default_gnm_help and make_gnm_help are also leaking.
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/206Implement least square fitting2018-05-22T13:55:25ZBugzillaImplement least square fitting## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#690045)](https://bugzilla.gnome.org/show_bug.cgi?id=690045)**
## Description
When applying least squares technics, one tries to minimize a least squares cost fu...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#690045)](https://bugzilla.gnome.org/show_bug.cgi?id=690045)**
## Description
When applying least squares technics, one tries to minimize a least squares cost function and gnumeric solver is able to do so.
However, usually, one needs not only the optimal values of the parameters, but also for the confidence intervals, one needs the covariance matrix, which is the inverse of the Hessian matrix of the cost function.
This Hessian matrix is internally computed by gnumeric's solver when performing Newton iterations and a simple option to output it somewhere in the workbook would be very valuable.
I wrote a little patch (below) to output the Hessian on the command line at every Newton iteration. It is dirty but useful.
root@LGGE-PC144:/usr/local/gnumeric# diff plugins/nlsolve/gnm-nlsolve.c.orig plugins/nlsolve/gnm-nlsolve.c
318a319,321
> /* Frédéric Parrenin - for Hessian outputing */
> int ifp,jfp;
> /* Frédéric Parrenin - end of for Hessian outputing */
321a325,332
> /* Frédéric Parrenin - Hessian outputing */
> printf("\n");
> for (ifp=0;ifp<n;ifp++)
> {
> for (jfp=0;jfp<n;jfp++) {printf("%g ",H[ifp][jfp]);}
> printf("\n");
> }
> /* Frédéric Parrenin - end of Hessian outputing */
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/205implement rich text also for serie titles2018-05-22T13:55:16ZBugzillaimplement rich text also for serie titles## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#689961)](https://bugzilla.gnome.org/show_bug.cgi?id=689961)**
## Description
Currently, it is possible to enter rich text with a markup langage in axis title, b...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#689961)](https://bugzilla.gnome.org/show_bug.cgi?id=689961)**
## Description
Currently, it is possible to enter rich text with a markup langage in axis title, but this functionnality is not available for serie title.
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/204Charts are deleted when the column above them is deleted.2018-05-22T13:55:06ZBugzillaCharts are deleted when the column above them is deleted.## Submitted by Jason
**[Link to original bug (#689683)](https://bugzilla.gnome.org/show_bug.cgi?id=689683)**
## Description
I believe this is a bug and not by design.
When a chart is created the assumption is its a floating object...## Submitted by Jason
**[Link to original bug (#689683)](https://bugzilla.gnome.org/show_bug.cgi?id=689683)**
## Description
I believe this is a bug and not by design.
When a chart is created the assumption is its a floating object, however if you delete the column that intersects the left edge of the chart, the chart is deleted...
I accidently deleted a load of charts that were off screen by doing this.
You can see this in the attachment if column B is deleted.
Version: 1.10.xhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/203incomplete gradient2018-05-22T13:54:56ZBugzillaincomplete gradient## Submitted by an unknown user
**[Link to original bug (#687799)](https://bugzilla.gnome.org/show_bug.cgi?id=687799)**
## Description
Created attachment 228304
sample gnumeric file
In the attached sheet the columns are supposed to...## Submitted by an unknown user
**[Link to original bug (#687799)](https://bugzilla.gnome.org/show_bug.cgi?id=687799)**
## Description
Created attachment 228304
sample gnumeric file
In the attached sheet the columns are supposed to be coloured using a bicolour gradient. Since the columns do not start at 0, only part of the gradient is shown.
PS: I know this is a horrible chart. I use it to show what not to do!
**Attachment 228304**, "sample gnumeric file":
[feeding.gnumeric](/uploads/22e17f2b8754e95360732f9c6f920818/feeding.gnumeric)https://gitlab.gnome.org/GNOME/gnumeric/-/issues/202[win32] gnumeric freezes. most be shut down.2018-05-22T13:54:45ZBugzilla[win32] gnumeric freezes. most be shut down.## Submitted by Ken Macdonald
**[Link to original bug (#686869)](https://bugzilla.gnome.org/show_bug.cgi?id=686869)**
## Description
copying from windows update screen(update history) and pasting into gnumeric.
sometimes, when I del...## Submitted by Ken Macdonald
**[Link to original bug (#686869)](https://bugzilla.gnome.org/show_bug.cgi?id=686869)**
## Description
copying from windows update screen(update history) and pasting into gnumeric.
sometimes, when I delete spaces between fields after importing (using the Text Import Configuration) using the Ctrl+- keys, gnumeric will freeze.
No or little movement is shown in task manager in terms of memory use, and after waiting an hour the first time, gave up and ended the gnumeric task and rebooted the computer.
this has happenned in Windows XP and Windows server 2003.
this has caused me to save quite often, and to avoid deleting anything imported this way!
Version: 1.10.xhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/201additions to plugins/fn-derivatives2018-05-22T13:54:32ZBugzillaadditions to plugins/fn-derivatives## Submitted by Tino
**[Link to original bug (#686670)](https://bugzilla.gnome.org/show_bug.cgi?id=686670)**
## Description
Created attachment 227030
contains all modified/added files in plugins/fn-derivatives
I'd like to add a few...## Submitted by Tino
**[Link to original bug (#686670)](https://bugzilla.gnome.org/show_bug.cgi?id=686670)**
## Description
Created attachment 227030
contains all modified/added files in plugins/fn-derivatives
I'd like to add a few functions to the plugin, mainly barrier options
which are currently missing.
All the files I've changed/added are attached. It compiles with the git
version but I had to test with 1.10.17 (as I'm having trouble with either
gtk3 or cairo and gnumeric causes an X BadValue error and dies). A test
sheet is included so if something doesn't work let me know and I'll try to
fix it.
Let me know what you think.
**Attachment 227030**, "contains all modified/added files in plugins/fn-derivatives":
[fn-derivatives.tar.bz2](/uploads/40fd825aa261ce59706a7786eec8ddc4/fn-derivatives.tar.bz2)
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/200ability to modify the properties of several series at the same time2018-05-22T13:54:16ZBugzillaability to modify the properties of several series at the same time## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#685526)](https://bugzilla.gnome.org/show_bug.cgi?id=685526)**
## Description
It would be nice to be able to modify the properties of several series at the same ...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#685526)](https://bugzilla.gnome.org/show_bug.cgi?id=685526)**
## Description
It would be nice to be able to modify the properties of several series at the same time, by selecting all of them in the tree.
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/199lines in XYcolor chart are monochrome2018-05-22T13:54:05ZBugzillalines in XYcolor chart are monochrome## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#685504)](https://bugzilla.gnome.org/show_bug.cgi?id=685504)**
## Description
Lines in XYcolor chart are monochrome.
It seems to me they should be colored accord...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#685504)](https://bugzilla.gnome.org/show_bug.cgi?id=685504)**
## Description
Lines in XYcolor chart are monochrome.
It seems to me they should be colored according to a linear interpolation of the color of the markers.
Version: git masterhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/198Remember width of sheet tab space when saving2018-05-22T13:53:51ZBugzillaRemember width of sheet tab space when saving## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#685408)](https://bugzilla.gnome.org/show_bug.cgi?id=685408)**
## Description
Created attachment 225690
.png file to illustrate the RFE
If you have multiple she...## Submitted by Frédéric Parrenin `@parrenin`
**[Link to original bug (#685408)](https://bugzilla.gnome.org/show_bug.cgi?id=685408)**
## Description
Created attachment 225690
.png file to illustrate the RFE
If you have multiple sheets in a workbook, you need a large width to show all sheets' tabs, so you need to move the little arrow on the right (see attached screenshot).
However, this position is not remembered when you close and re-open the sheet, which is quite annoying.
By the way, why gnumeric does not use the entire width of the window for displaying the sheets' tabs? It seems the place at the right of the arrow is not used for anything. Moreover, I don't see the interest of having the 'SUM' information on the most right rectangle, which is a function like another one.
**Attachment 225690**, ".png file to illustrate the RFE":
![GUI2](/uploads/1c66747c0833af77cb24bad4a994f43d/GUI2.png)
Version: git master