Re: [Cdk-jchempaint] highlighting, blurring, etc

[Egon W. <ego...@gm...>]

On Thu, Jan 9, 2014 at 3:29 PM, John May <jo...@eb...> wrote:
> As you say a toolkit independent abstraction - however if I were to plug in the
> CDK-JChemPaint IUndoRedoable edits into JChemPaint they would not work.

Yeah, it does depend on bridges to the various toolkits. Arvid has
been doing this for SWT.

I do not know what Stefan and Gilleain did for Swing in the JChemPaint
applet... I have no insight in what they did in that respect after the
fork.

> Part of the problem I see is that the granularity ‘ChangeCoordsEdit.java’
> etc isn’t needed - for AtomContainers a more robust approach is to simply
> clone the container an push it onto the Undo/Redo stack.

That was the original JChemPaint implementation, but was too slow.

I would need to dig up more history to see if that was the only reason
or not... I do not remember at this moment.

Arvid, do you happen to remember the full argumentation for the edit design?

> With fewer edit classes it’s now trivial to write the undo/redo manager interaction.

That I agree with.

> What I really worry about is -
> https://github.com/egonw/cdk/tree/13-unsorted-patches/src/main/org/openscience/cdk/controller
> and
> https://github.com/egonw/cdk/tree/13-unsorted-patches/src/main/org/openscience/cdk/controller/undoredo
>
> That’s a load much more code to add to an already strained code base. Worst
> of all - everything public! and adds to the JavaDoc. I’m really trying to
> trim down the bloat and make things more streamlined, efficient and easier
> to use.

Got it.

Maybe with the current CDK just a stack of IChemModel's is now fast
enough and the way to go...

Egon

-- 
Dr E.L. Willighagen
Postdoctoral Researcher
Department of Bioinformatics - BiGCaT
Maastricht University (http://www.bigcat.unimaas.nl/)
Homepage: http://egonw.github.com/
LinkedIn: http://se.linkedin.com/in/egonw
Blog: http://chem-bla-ics.blogspot.com/
PubList: http://www.citeulike.org/user/egonw/tag/papers
ORCID: 0000-0001-7542-0286