Replies: 7 comments 9 replies
-
This is already exists and is a simple as it gets: you just need to share the folder with your nodes. For instance, I made the example nodes for working with Pillow available here. Here are the steps to use the nodes:
In other words, it is pretty straightforward: just download the nodes and enable them inside the app. |
Beta Was this translation helpful? Give feedback.
-
Sharing the nodes will be exactly like I just described: just upload them to your github and share the link with people on the internet, but there will be an additional tool: As we briefly discussed on discord, I'm thinking of creating a website and put it on https://gallery.nodezator.com that works like a google for nodes and a node gallery, where people will be able to:
|
Beta Was this translation helpful? Give feedback.
-
I think I'll add all those steps/instructions in the manual, including inside the app, so people know how to use it. Do you think all the steps are ok? Are they simple enough? I think having people share their nodes as repos on github is the best solution, because it is something which is straightforward and decentralized. If I were to create a centralized website or database, people would be dependent on the online service be always the case. I've seen projects shutdown before and leave people unable to access essential resources. CTAN for instance had to shutdown new user registrations, see what they say on their website:
For Nodezator, I'd like node sharing/distribution to rely on github, this way all nodes will always be available no matter what happens to the project (though don't worry, the project is strong and came to stay, I'm just doing what I think is best). And we'll still have https://gallery.nodezator.com to share useful nodes node packs and help people find them, pointing people to the github repos where they can be downloaded. |
Beta Was this translation helpful? Give feedback.
-
Yes, I already intend to do so. The source code for some of the nodes, for instance, even have a right-click event listener, I just didn't implemented the context menu yet. The only thing preventing me from implementing them is to know what to put on them. Some objects even have context menu's already. For instance, input sockets created for subparameters (sockets created under a variable parameter like *args or **kwargs) already have a context menu, you can see it in actions in this part of the video, because I already had some idea of which commands to put there. In other words, context menus will sure be there. I'll create 02 new discussions later today:
This way we'll be able to keep track of things to be done and when they will be arriving, etc.
This sounds good, I too think right-clicking should display the option to do so in a context menu. I'll add it to the to-do list for the next version which I mentioned above as the New discussion number 1.
I'm thinking of adding the double-clicking to edit text blocks to the New discussion number 1 later today as well. I'm not sure it would be good interface for breaking connections though. It could be enabled though, but let users opt-out of it in the user preferences, I'm not sure. Let me ponder about it a bit more. |
Beta Was this translation helpful? Give feedback.
-
Resizable main window a consideration? |
Beta Was this translation helpful? Give feedback.
-
Some other thoughts:
|
Beta Was this translation helpful? Give feedback.
-
Another feature idea --- would it be desirable to add an option for a graphical icon for a node? The GraphSCAD folks do this of course, also the Nodebox folks, so it's probably more useful for the 3D/graphical work I'm trying to get done, but it might be useful for other things --- though it raises the question, "What does an algorithm look like?" |
Beta Was this translation helpful? Give feedback.
-
There's already been a bit of discussion at: https://old.reddit.com/r/Python/comments/vf6h1z/nodezator_new_python_node_editor_released_to/
Here are some things to begin:
Beta Was this translation helpful? Give feedback.
All reactions