GNOME Bugzilla – Bug 724761
ods roundtrip of viewport position
Last modified: 2014-02-20 19:45:02 UTC
If I accidentally save one of the roundtrip test files with the cursor not at A1 I get a failure. Does the ods format handle this? t6502-styles.pl: Check style ods roundtrip. | Using exporter Gnumeric_XmlIO:sax --- style-tests.xml 2014-02-19 20:32:57.000000000 -0500 +++ style-tests-new.xml 2014-02-19 20:32:57.000000000 -0500 @@ -423,7 +423,7 @@ <gnm:Cell Row="31" Col="0" ValueType="60">Sans 24</gnm:Cell> <gnm:Cell Row="31" Col="1" ValueType="60">Text</gnm:Cell> </gnm:Cells> - <gnm:SheetLayout TopLeft="A11"/> + <gnm:SheetLayout TopLeft="A1"/> <gnm:Solver ModelType="0" ProblemType="0" MaxTime="60" MaxIter="1000" NonNeg="1" Discr="0" AutoScale="0" ProgramR="0"/> </gnm:Sheet> </gnm:Sheets>
I don't think (but am not sure) that there is a standard way of storing the cursor position. Of course we can include it in our own settings.
LO keeps the cursor position within its proprietary <config:config-item-set config:name="ooo:view-settings">
Looking at our code and trying out some examples, it looks to me like we correctly save and restore cursor position to and from ODF.
Morten, I can't replicate this while saving and opening files using Gnumeric. Can you?
This is strange. I hit that a dozen times yesterday on my home machine, but I don't see it here at work.
Got it. It's not the cursor position, but the viewport position, i.e., what cell is the first one visible. This hits my laptop much more because it's old and has a small screen showing fewer rows than my work machine.
We are now writing the viewport position (using LO's mechanismm). We are not reading it yet, but the writing can be tested by opening the file with LO.
This problem has been fixed in the development version. The fix will be available in the next major software release. Thank you for your bug report.