Join the discussion

Write your take first — we'll ask for email only when you're ready to publish.

  • Hacker News
  • I might be missing something but it seems to only be able to edit and work with existing mmd and svg files?

    If I do "touch flow.mmd" then run oxdraw to edit the file, it doesn't give me an option to initialize it as a new diagram.

  • hi this true! i personally typically write my diagram or add nodes in a text editor prior to customizing but I do realize that many would prefer to be able to do everything in the gui so I will start adding features to add nodes and initialize the diagram in the gui!
  • Thank you! Very cool.

    I don't see a button for it (on mobile currently, and will check thoroughly at my computer) -- is there a button to add a node?

    Another feature I've always craved for code diagramming is "collapse downstream nodes" -- though it might be outside of your scope (and mermaids?).

  • Hi as of now I haven't added that feature so the current way would be to edit the .mmd text. However I think you are right that would be a good feature to add. For collapsing downstream nodes I think it would make sense with the request the someone else made in this thread for animations. From what I see it should certainly be possible to implement at some point!
  • Hello Rohan. This is really great. If you are able to include parameters to expose the intermediate data as inputs and outputs, so that this can be run to a step in the process and output the data, or run from a step with pre-prepared data. It would mean that other people could build on what you've done to create other diagrams and renderings.
  • Hi thanks! I'm not sure I fully understand what you mean by intermediate data? Would this be the location data computed for the different components?
  • I really wish Mermaid would just ratify a layout spec. Make it optional. Use it. Great. Don’t use it. The layout engine does its thing.
  • Great project, I've been generating diagrams with llm for a while, and I often struggle to refine layout through the model.

    But now I can interactively make changes to the diagram with this, it's very productive.

  • Thank you! Happy to hear that!
  • This is awesome. I was looking for exactly this last week. A tool I could prompt AI to come up with an architecture and then be able to pick up manually, but visually not editing the code.

    Being able to express a workflow or diagram and then have AI implement would be awesome to have a tight loop.

  • Great to hear! Absolutely this is the goal.
  • I really wish PlantUML would just solve this jarring problem.
  • It's definitely much needed.

    I use PlantUML for most diagramming but for anything with more than about 5 components in it I'm spending 20-30% of my time desperately trying to tweak the layout with hints.

    It's an interesting approach to embed comments and then build that into the layout engine. I've always thought it would solve a lot of my issues if I could just lock the coordinates for certain components and then let the layout engine do the rest with those as hard constraints. This looks like something similar to that approach.

    I really want this because the alternative is to spill over to completely manually maintained diagrams using GUI tools which then can't be easily integrated with source control - I want the same commit that changes the code to also change the architecture diagram for that code. Then it is part of code review and integrates to the whole process well.

  • I use PlantUML because it renders in GitLab's markdown, including wikis, MD docs and even PR comments. However, I have to use Mermaid for projects hosted on GitHub.

    The hassle of tweaking the layout in puml, such as pairing elements with an invisible connections and groups, adding or removing dashes from the arrows in class diagrams... is gone because Mermaid is simply inferior in that sense.

    Mermaid always feels like it's in beta and I don't understand why GitHub ignores the request to support puml (1). It seems that adoption of diagrams as code is tied to what is supported by major vendors and they don't care enough. Or maybe it is because mermaidchart made an official vscode plugin, who knows.

    While I agree that improvements are needed, I'm not convinced that creating a third standard is the answer. What I would like is to be able to assign weights to my elements and let the renderer do the work (not set x and y coordinates like in oxdraw).

    [1] https://github.com/orgs/community/discussions/10111

  • Great job on the releasing the project; it definitely solves a need of being able to use declarative syntax for defining the relationships, and then customizing the layout which the regular layout generators can't do.

    Project's Cargo.toml file says code is licensed under MIT license, but there's no license file in the repository, so Github doesn't show what the project is licensed under. Please add the license file so that people see it without having to dig through the code/configuration to determine that.

  • Thanks for catching that! Just added the license file.
  • If you wish to increase the adoption the tool, do consider hosting it to make it easy for people to use it. I see that it's heavily dependent on server-side code, so the cheap/free static hosting wouldn't be an option.
  • This looks like a very promising project, I have been looking for exactly this.

    One feature I would love to see a declarative diagramming solution would support is a hover pop-up with more information or nested diagrams.

  • Thanks! I think that sounds interesting, to make sure I'm understanding your use case would these pop ups be for your own use or for other people? For example would you want to send a link to someone else on your team and then the link shows the diagram with popups and nesting? Or would it be sufficient to send over the .mmd files and then the other person can use the cli to open the web interface which supports the popups and nesting. I imagine I could add the latter one quickly but for the former I would either add an easy way for users to self serve like with ngrok or some cloud solution. Or alternatively I could add some way to export the diagram just as a standalone HTML file in which case that could be sent and support the popups and hovering without the person you're sending it to having to have the CLI installed.
  • I've made a private MacPorts port[1]; if I find that I use it frequently enough, I might contribute it to the main MacPorts port repo[2].

    One thing that's missing from my perspective (and this is probably true for Homebrew packaging as well, but I don't do that) is Git tags / GitHub releases associated with your Cargo releases.

    I can work around it for now by using an explicit release (`9ccd9bf53f9a309ccda42b5c17e9c1056493fb90` is what I'm assuming was your 0.1.0 release point).

    I've also assumed that npm10 is sufficient (which currently installs node22 on MacPorts).

    [1] https://github.com/halostatue/ports

    [2] https://github.com/macports/macports-ports

    [3] https://github.com/halostatue/ports/commit/e7331a7fcae362b0d...

  • Thankyou!
  • > npm10 is sufficient (which currently installs node22 on MacPorts)

    Wait, no, node22 comes with npm10, not the other way around.

  • Great project! You chose to base on Mermaid.js and that is indeed the most popular declarative diagramming library right now. A more versatile declarative diagramming approach is provided by the D2 Language [0], which may be interesting to check when it comes to supporting more complex diagram types.

    [0] https://d2lang.com/

  • D2 is better, but it's not supported by GitHub. Go where your users are.
  • Thanks for sharing. One of the very fun parts of posting this for me was seeing how much bigger this niche was than I thought haha. D2 seems great and super featureful!