On Wed, Jun 15, 2016 at 3:44 PM, Geoffrey Hutchison <[email protected]> wrote: > > - Are there general or specific “loose threads” in the code? It seems like > there are some needs for batch undo/redo in RWMolecule, for example.
I would like to investigate grouping commands, and naming them. Batches/larger commands would be good for efficiency, and achieve much the same. I could see both being useful. > > - You mentioned implementing selections.. where should selections "live"? > Does the molecule track this? Does the widget? For example, should selections > be a special type of molecular fragment? I was working on making some additions to molecule, and think that is a natural place. I could see some argument for attaching them to widgets, but would prefer to keep it simple for now. > > - Implementing fragments and residues.. this seems like it needs to live in > both Molecule and RWMolecule with the current class scheme. How do changes to > fragments coalesce? For example, does there need to be a mechanism to > invalidate a residue name when an edit occurs? Nothing really lives in RWMolecule, it simply references data in Molecule, and provides undo/redo functionality. Anything in RWMolecule that is not in Molecule should be to facilitate undo/redo. Those are my thoughts at least, and by layering RWMolecule on Molecule it allows much of the cost for undo/redo to live in that class too. > > - Rendering options? Right now, there's no way to change atom sizes or bond > sizes.. where would this code live? I would say the view configuration, in much the same way it did in Avogadro 1. The ball and stick plugin features a configuration widget, and offers multiple bonds/hydrogen visibility as two options. I meant to go back and add bond/atom radius multipliers - it shouldn't be too hard. > > - Settings in general.. Do we need to go add QSettings read/write to plugins > like in Avo1? How should this work with the Python scripts like input > generators? QSettings is a good question - is that still supported as the Qt community was talking about deprecating the INI file stuff, and QSettings I thought. I would love to wrap more of this in JSON, and have a JSON settings file, along with JSON data files. This could then be passed to Python even, or combined. That is a good question on persistent settings and input generators - I hadn't looked into that. > > - How / when will volume rendering show up? (i.e., what's the barrier for > that to be possible?) VTK changed quite a bit, I would like to resurrect it. I know know the code to edit the color maps much better too, and so could add the user interface for it more easily than back then. This was the big missing piece. I will see if I can take a look at where that is on the weekend - it was crashing which is why I removed the option, but I really want to get it back in there. ------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports. http://pubads.g.doubleclick.net/gampad/clk?id=1444514421&iu=/41014381 _______________________________________________ Avogadro-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/avogadro-devel
