From kgraphviewer-devel Thu Sep 14 20:22:53 2017 From: kleag () free ! fr Date: Thu, 14 Sep 2017 20:22:53 +0000 To: kgraphviewer-devel Subject: Re: kgraphviewer and QML ? Message-Id: <279557661.133093084.1505420573291.JavaMail.root () zimbra61-e11 ! priv ! proxad ! net> X-MARC-Message: https://marc.info/?l=kgraphviewer-devel&m=150542059027193 Thanks Millian, I'll have a look to that. Ga=C3=ABl ----- Mail original ----- > De: "Milian Wolff" > =C3=80: kgraphviewer-devel@kde.org > Cc: kleag@free.fr, kfunk@kde.org > Envoy=C3=A9: Jeudi 14 Septembre 2017 11:33:11 > Objet: Re: kgraphviewer and QML ? >=20 > On Donnerstag, 14. September 2017 10:10:39 CEST kleag@free.fr wrote: > > Hi, > >=20 > > I'm working on a new GUI for the linguistic analyzer of my lab: > > https://github.com/aymara/lima/tree/master/lima_gui > >=20 > > The interface is currently purely QML based with text based > > displays. A > > standard way at looking results is through graphs like the one > > attached. In > > fact, I started kgraphviewer (kgv) to visualize these graphs=E2=80=A6 > >=20 > > So, I thinked about integrating the kgraphviewer kpart in the GUI > > and then > > discovered that it is not quite possible. Or at least complex and > > inefficient. > >=20 > > I'd like to get your advice on what would be the right solution. I > > thinked > > at several possibilities: - use the graphivz lib API to generate an > > image > > and show the image in the UI; - use the kgv API to load the graph > > without > > dispalying it in the widget and export it to an image like above; - > > use the > > kgv API to load the graph and reimplement in QML the kgv canvas > > elements > > (based on QAbstractGraphicsShapeItem). > >=20 > > The latter seems to be the best but also complicated with the > > current QML > > API. > >=20 > > What do you think ? Does one of these solutions seems reasonable ? > > Other > > ideas ? > >=20 > > Thanks for your attention ? >=20 > I think you should have a look at https://github.com/KDAB/ > kdstatemachineeditor. Afaik it contains a DOT visualization based on > QML. Talk > to Kevin (CCed) for more information. >=20 > Cheers > -- > Milian Wolff > mail@milianw.de > http://milianw.de