Hi all just a quick reaction from my side (sorry for being so brief): - it is quite unfair with Guiseppe if you people start proposing how things should be done _now_ - IMO there was plenty of time to do such decisions before - I think it is quite irrelevant whether the implementation is in C++ or in Python: in case the interfaces (such as QgsDataItem) are implemented correctly there is no real difference (if the code is shipped with QGIS). - it has been decided in the proposal that support for saving vector layers to a particular data providers will be implemented to get generic vector import/export, the rest of the code is not going to be that reusable anyway. - the Qbrowser API is IMHO not completely finished and still might see some changes. This might cause additional complications for Giuseppe.
So in general I would suggest proceeding with the original plan and to try to integrate some parts into QBrowser - e.g. implement QgsDataItem support for PostGIS and SpatiaLite, together with the paramWidget() that might reuse a lot of work from the db manager plugin itself. Regards Martin On 5/25/11, Giuseppe Sucameli <[email protected]> wrote: > On Wed, May 25, 2011 at 1:21 PM, Radim Blazek <[email protected]>wrote: > >> On Wed, May 25, 2011 at 10:39 AM, Giuseppe Sucameli >> <[email protected]> >> wrote: >> >> > However we should consider to integrate useful python plugins in qgis >> code, porting >> > their base functions to C++ if necessary. >> >> So why to write another important plugin in Python, with perspective >> that it has to be ported later to C++? >> > > Again, I would develop in python the GUI only, not the base functions ;) > > Tell me what in the QBrowser class cannot be done in python, it's just a > container. It uses the QgsBrowserModel and some other Qt based classes > I can easily handle in python without any problem. > > I don't think the QBrowser code makes particular GUI widgets/functions we > should insert in app/core/gui. Different would be if you had developed a > QgsBrowser tree widget implementing also the QgsLegend interface, as > everybody could reuse that. > > However, discussing it we're also going to find what the current > implementation > misses and what should be changed, doesn't matter if at the end it will be > done in C++ or python. It will be Martin who has to decide. > > No other devs want to partecipate to the discussion? ;) > I think that a lot of interesting things are coming out. > > Regards. > > -- > Giuseppe Sucameli > _______________________________________________ Qgis-developer mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/qgis-developer
