-
Notifications
You must be signed in to change notification settings - Fork 3
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Small feature requests #8
Comments
[REMOVED: just directly edited the head comment instead of commenting separately.] |
I'd like to co-opt this thread for all smaller feature requests. Here are a couple more:
|
I played a bit with the latest version (branch: multiple_scripts). I realy like the new look and feel! A few small suggestions:
|
Yikes! TBH, I didn't get the the nglview widget set up in lab yet (until recently I was a notebook user), so I didn't even notice! Definitely an undesirable bug 🐛 |
Having learned more about the libraries we depend on, I don't think this is going to be very practical. IMO the crux of the issue is that ipycanvas doesn't allow for easy rescaling of objects, so having something like a zoom function in the main canvas area would be absolute hell to implement. As a result, we'll want to keep the actual nodes fairly compact. I do think we can get away with one of the other request about resizing nodes, at least so far as they have "expanded" and "compact" views... but I think trying to squeeze in an entire array or plot is overkill, even for the "expanded" view. Hopefully this request is largely moot anyhow. As of #33, the default behaviour for pyiron nodes is to have a "display" button that takes over one of the other GUI windows with a much larger representation of the object (e.g. an NGLView plot for structures, a string for arrays, etc.) So this data is (or can be) available without having to add extra display nodes now. |
A few ideas about features to enhance pyironflow and improve user friendliness:
use different colors for the exec and data connectors (both for input and output)More colour control #13make dot on exec input clickable (i.e., rather than connecting it with a click button one can directly activate the node/workflow by clicking on the exec input)Plot3d
and similar. For actual jobs that run, you still need the click button, but I think job management needs more thought anyhow. Ok, I thought more and I want to "uncomplete" this request. Push-button execution should be reasonable for "exec"y ports as well as "display"y ports. Will do it later.a click on the node shows its content (gui) but makes it also moveable. It would be good to have to separate events, e.g. clicking enter to show the nodes gui when the mouse pointer is over itMostly in Just one usage mode #15, which changes the feel quite a bit overall. Keyboard is totally ignored now, so no using "enter" with a hovering mouse.sort node list alphabeticallySort selection keys #11make nodes resizable (by using mouse pointer or based on content)Free-form resizing is pretty tough, but nodes now automatically size their height based on how many IO ports they have, and are collapse-/expandable to compress/show this IO over in Collapsable input #41put static output (e.g. matplotlib, pandas tables, etc. ) directly into nodeCurrently declined, see comment belowallow also touch devices (e.g. IPad), i.e., next to mouse events also the equivalent touch events should be supportedThis shouldn't be too hard, but for practical reasons it will take a while to get to, so I made a separate issue for it: Touch support #47use Tab widget from ipywidgets to manage multiple scripts (could be similar in look and feel to browser tabs)Multiple scripts #26; I'm not totally happy with all the button icons, but I'm having trouble getting the ones I want to render...these are good enough for now though. The "+" (i.e. new script) tab is wider than it should be, but this is a known ipywidgets issue and doesn't seem to be fixable from the python side.include node editor (allows users to interactively develop and test new nodes within ironflow, without forcing them to switch environment and use an external code editor)Notebook nodes #14, at least to the extent that all the tools are now available to write new node classes wherever the gui is being called from (e.g. the jupyter notebook) and thengui.register_user_node(MyUserNodeClass)
makes them immediately available in the gui.The text was updated successfully, but these errors were encountered: