Plans to modernize the GUI framework?

[Shadow775711]

Hello,

I’m a new Ghidra user and first of all I’d like to thank you for this amazing tool and all the work put into its development.

Overall, the user interface works well and is functional, but I feel it’s starting to look a bit dated. Projects like ImHex have shown that it’s possible to create a modern and aesthetically pleasing GUI, which could help attract new users as well.

I do understand that Ghidra is a huge project and rewriting the graphical backend would be a very complex and time-consuming task. That’s why I wanted to ask if there are any long-term plans to move away from Swing to a more modern framework (e.g., JavaFX) or to use hardware-accelerated rendering (e.g., OpenGL/Vulkan) at least for more complex components like graph visualization?

I’m curious whether such modernization is being considered at all and whether potential contributions in that direction would be welcome.

Thank you in advance for your reply!

[dragonmacher]

There are no plans for such a modernization. Swing is so deeply embedded in the project that a full rewrite would likely be needed to change the UI.

Not only do we use Swing, but we have created our own layer of widgets on top of Swing to make UI development easier (e.g., GTables and GTrees). Creating new complex UI components in Ghidra is super easy. Moving to a new framework would mean replacing some or all of these custom widgets.

I had looked into JavaFx years ago. At that time, it seemed like the project had stalled. My impression is that JavaFx was intended to be a replacement for Swing, but I could not find community response that suggested JavaFx was gaining traction. I imagine that having so many devs familiar with Swing keeps that API in use. I think you can build modern, attractive UIs in Swing, but perhaps that is not typically a priority for many projects.

I suppose it would be possible to update the UI, while keeping Swing, to create a modern aesthetic. But, that will likely never be high enough on our priority list to get done.

My hope is that eventually our outdated UI will become retro-chic.

[Shadow775711]

I understand that a full UI rewrite isn’t planned and that Swing is deeply embedded in the project. I’m curious, though, why frontend improvements aren’t considered a higher priority. I understand the backend is crucial, since it handles the debugger, analysis, and translation of code to C, but a better frontend could also significantly improve the analyst’s experience.

[dragonmacher]

As most large projects, we have more work than resources to do the work. We always focus on the bugs and new features. I think generally people don't worry about UI updates, since the UI is working. Any work we evaluate has to consider the amount of work vs the potential reward. The amount of work to update the UI would be staggering, as it would touch most of the code base. From that perspective, spending that many resources for aesthetics has never seemed worth the tradeoff. Also, changes at that scale would introduce a large number of regressions into the system, which could take years to find and fix.

If our system had a clean separation of UI and business logic and had full test coverage for all features available from the UI, then the work would be much easier and the risk of breakage would be lower. Unfortunately, we do not have cleanly separated code with full coverage tests.

We do place a big emphasis on workflow issues, which often involves adding or updating the UI to be more efficient for the user.

[ryanmkurtz]

I've always been surprised that the RE community has not asked us to make it look more 8-bitty.

[astrelsky]

It has it's place but I don't think it fits here. People don't really want to be using outdated tools everyday or when working.

I will only speak for myself though.