Adding Arnaud to the loop, being the author of the PR which added constrained numeric fields in PostgreSQL provider [1]
[1] https://github.com/qgis/qgis/commit/92f71b696ca93c792ae5602ed82863fcef0e5006 Arnaud: what was the reason for constraining that field ? The semantic of QgsField's length and precision attributes aren't fully defined and they seem not to match PostgreSQL numeric relatives "precision" and "scale" [2] [2] https://www.postgresql.org/docs/9.5/static/datatype-numeric.html#DATATYPE-NUMERIC-DECIMAL --strk; On Tue, Oct 11, 2016 at 10:51:34AM +0200, Sandro Santilli wrote: > On Tue, Oct 11, 2016 at 10:47:50AM +0200, Paolo Cavallini wrote: > > Il 11 ottobre 2016 10:45:18 CEST, Sandro Santilli <[email protected]> ha scritto: > > > >Does anyone have an idea about how should QgsField::length be > > >interpreted ? > > > > Isn't it related to the comma being counted in real fields? > > The bug ? Probably. QgsOgrProvider::convertField adds one byte > for the comma while QgsPostgresProvider::convertField does not > > But what I'd like to know is whether or not QgsField::length > is supposed to count or not the comma (or what else is meant > to be expressing) for a QVariant::Double and for other types. > > --strk; > _______________________________________________ > Qgis-developer mailing list > [email protected] > List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer > Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer _______________________________________________ Qgis-developer mailing list [email protected] List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer
