Feature requests (v2)

[Ni55aN]

By collecting feature requests and use case information directly from our users, we can gain insights into their pain points, identify common patterns and focus on developing features that will have the greatest impact.

To ensure effective communication and gather comprehensive information, we kindly request users to provide the following details in their feature requests (feature per comment):

1. Feature Description: Please provide a clear and concise description of the desired feature (2-3 sentences). This should include what the feature aims to accomplish and how it would enhance the user experience.
2. Use Case Context: Describe the specific context in which the feature would be used. Help us understand the problem you're trying to solve or the task you're trying to achieve. Providing real-world examples or scenarios where the feature would be beneficial will greatly assist us in assessing its potential value.
3. Existing Solutions: If you are aware of any existing solutions or workarounds that partially address your feature request, please share them with us. This will help us identify gaps in the current functionality and potentially leverage existing solutions as a starting point.

In return for submitting a feature request, you can expect one of the following from us, based on the nature of the feature:

• General-Purpose Plugin: If the requested feature aligns with the project's overall goals and do not require excessive configuring for many use cases, we will consider developing it as a plugin. This way, it can be easily integrated into the existing application without affecting its core functionality
• Example Implementation: In cases where multiple solutions are possible, and the feature heavily depends on specific use cases or applications, we will provide an example. This will demonstrate how the feature can be implemented and customized based on your specific requirements.

⚠ If you would rather provide your feedback in a more informal manner or if you are unsure whether a similar functionality or workaround already exists, you can join our Discord server and share your feature request there.

💡 Please use the "thumbs-up" 👍 reaction to show support for existing feature requests rather than duplicating them in separate comments.

v1

You may be curious about the progress made on feature requests from the previous version. Most of them had been actively reviewed based on your feedback

Sponsorship and contribution

If you are interested in supporting the development of a specific feature, you can consider sponsoring its implementation or providing drafts or prototypes.

[Ni55aN]

Pan the viewport when dragging a node

1. Enable panning when dragging a node near viewport edge. When the dragged node reaches the boundaries of the viewport, the area should start translating.
2. It would be helpful if users could drag a node beyond the boundaries of the viewport without having to release the node and separately pan the viewport
3. Unreal Engine Blueprint

[Ni55aN]

Implemented in the Panning boundary example

[ShaitanLyss]

Macro nodes

1. Enable the creation of macro nodes, a macro node is a node that represents a collapsed node graph.
2. It allows users to create their own custom nodes for reusability and sharing
3. This could be implemented by creating a node editor for each macro node but I assume it would be slightly more optimal if they would all be part of the same node editor (maybe not) or much better if they could be compiled in some way (codegen ?). This feature request is akin to unreal blueprint macros.

[stallboy]

Keep the node position and render it sync

1. store the node position with id, when called createEditor again, I can pass the position info in parameters and render it sync so the UI is not flickering.
2. the createEditor function transforms parameter to node and connection. that's simplest programming model when use rete, but it surfers from the flickering problem when I update the parameter the node position change and the node created async .
3. one can update the node and connection incremental, but the code is really complex, the mind model turns from pure function model to state-machine model. the other problem is the how to auto layout it incremental..