[Marble-devel] [Patch] Introspection/ Debug view for the GeoData

Patrick Spendrin ps_ml at gmx.de
Tue Aug 19 23:15:10 CEST 2008

tackat at t-online.de schrieb:
> Hi Simon,
> Looks nice. I still need to look through the whole patch but here are a
> few first thoughts:
> - The Debug plugin should never appear for the end-user. Hence it's
> important that the plugin will only get compiled if Marble gets compiled
> if the plugin is explicitely included into the compile process.
> Therefore I suggest to change the cmake script in a way that the
> debug-plugin will only get compiled if the BUILD_MARBLE_TESTS cmake
> option is enabled. An additional requirement for me is that the plugin
> is disabled by default and needs to get enabled in the plugin settings
> dialog.
> - Now that I see the actual GeoCoordinates constructor I'm not so sure
> anymore whether we shouldn't have the parent as the last argument.
> Patrick, Inge, Jens-Michael: what do you think?
> Regards,
> Torsten 
> -----Original Message-----
>> Date: Mon, 18 Aug 2008 00:47:44 +0200
>> Subject: Re: [Marble-devel] [Patch] Introspection/ Debug view for the
>> GeoData
>> From: Simon Schmeisser <mail_to_wrt at gmx.de>
>> To: Patrick Spendrin <ps_ml at gmx.de style="margin:0px;">
>> On Monday 18 August 2008 00:12:41 Patrick Spendrin wrote:
>>> Simon Schmeisser schrieb:
>>>> Hi Guys,
>>>> as already discussed briefly on the mailing list I have developed
>>>> a small feature that provides introspection for the GeoData. By
>>>> GeoData I mean the structure where the content of loaded KML files
>>>> is stored (currently only used by two plugins: geodata and
>>>> placemarks).
>>>> It works by creating a QAbstractTreeModel based model, that
>>>> represents the data structure of the main GeoDataDocument (or
>>>> multiple GeoDataDocuments in future maybe ). This Model is than
>>>> displayed by a standard QTreeView.
>>>> Necessary changes: - Every GeoDataObject based class needs to know
>>>> about it's parent node in the GeoData tree.
>>>> - GeoDataObject derifed classes that have sup-structures need to
>>>> override "child( int pos )" and "childCount". This has been done
>>>> for GeoDataContainer.
>>>> Future Improvements:
>>>> - Add a title ;)
>>>> - improve plugin system for things like this (doesn't render
>>>> anything but should maybe get an menu entry)
>>>> - implement GeoDataPlacemark::child and childCount to return the
>>>> geometry it contains.
>>>> - display attributes like name, id, visibility, style as childs or
>>>> table columns
>>>> - direct editing?
>>>> Please review the attached diff and contact me to fix any
>>>> outstanding issues. I'll update the patch if necessary after the
>>>> double -> qReal transition.
>>> The Copy Ctors should not get another argument - the parent
>>> information has to be set from outside. This means as well that
>>> after a copy the parent informations are out of date and need to be
>>> resync'ed.
>> fixed, new patch is attached
>>>> cu Simon
>>> regards,
>>> Patrick
After rereading your patch, my last question disappeared:
GeoDataCoordinates (which is no GeoDataObject as you said above) has not 
been included in this patch (and shouldn't since it would slow down 
marble probably a bit).
Please apply the second patch.
I am not sure what needs to be changed to make the Model editable - this 
is needed to interactively change the maps later. This can be discussed 
maybe later.

> ------------------------------------------------------------------------
>>>> _______________________________________________
>>>> Marble-devel mailing list
>>>> Marble-devel at kde.org
>>>> https://mail.kde.org/mailman/listinfo/marble-devel
>> I have also already commited the jakobsweg.kml sample file, thus
>> reducing the patch quite a bit.
>> regards Simon

web:                 http://windows.kde.org
mailing list:        kde-windows at kde.org
irc:                 #kde-windows (irc.freenode.net)

More information about the Marble-devel mailing list