Created attachment 126506 [details]
1) Open attached doc
2) Notice all 4 shapes are visible and appear when you print them (print to file if you dont have a printer ;D)
Build ID: d2e4753c3f511cfc6b2932ce60d0bc2e09296f9f
CPU Threads: 2; OS Version: Linux 3.19; UI Render: default;
TinderBox: Linux-rpm_deb-x86_64@70-TDF, Branch:master, Time: 2016-07-26_17:32:37
Locale: en-US (en_US.UTF-8); Calc: group
So ODF specifies that the <draw:layer> tag can have a draw:display attribute with values that change how it appears on screen or outputted to the printer and these values are being ignored on import. We do support the functionality for the 'none' value, which is equivalent to making the layer invisible, but not sure about the 'printer' (only show it when printed) and 'screen' (only show it on the screen) values.
Created attachment 126509 [details]
Indeed, the attributes draw:display and draw:protected have never been implemented. Instead the settings config:name="VisibleLayers" and config:name="PrintableLayers" and config:name="LockedLayers" are used with binary data.
Bug 53864 "Export ignores layer options 'printable' and 'visible'" could be solved by another flag leading to 2^4 combinations, or the export is bind to one of the existing states. In this case I suggest to use printable.
We need no additional flag but only an export to the ODF attributes instead of writing it to binary(!) private config items.
The UI has checkboxes for visible, printable and locked. Checkbox 'visible' together with checkbox 'printable' results in the four attribute values 'always', 'none', 'printer' and 'screen' of attribute draw:display. And UI checkbox 'locked' corresponds directly to the values 'false' and 'true' of attribute draw:protected.
I'll try to find a solution.
Dear Regina Henschel,
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.
I have suggested it as topic for the Hackfest in Hamburg 7./8. Apr. 2018.
Ok, assigning it back to you then...
Yes, I'm still working on it. Armin reviews my work, but he is currently very busy with all the regressions, therefore progress is slow.
The solution for this issue needs large changes in core.