mud!

Finally trying out overarch today between other tasks, thank you for creating it 🙏

Feedback on your experience is very welcome. Overarch is still in active development (as time permits) and I really love to have some external input. Also discussions with clojure minded users will be really helpful.

I’m taking some notes and will send them your way when I’m done. Right now nothing’s blocking me, so that’s cool 🙂

Perfect!

actually looking into the source, I've already implemented a no-legend option. 🤫 Add a {:spec {:no-legend true}} in the view definition.

I won’t tell anyone 😄

And to tell you a bit more about my use case: The diagrams are part of a pitch and the potential customer doesn’t like English words. It’s great being able to define some elements with English names and descriptions and then overriding them with German translations in other places. As I didn’t dig into how to change the legend, I figured I could just hide it. And it wasn’t blocking because cropping the resulting diagrams was quick enough (only 4 diagrams).

Clojure to the rescue. 😀 Overriding some information in the views is really useful. I use it for example to render parts of the internal system as external, when they are not the current focus of the view.

No need to say sorry, that's one of the things this channel is for (plus other stuff)

Might there be any other (initially inobvious) structural invariants that would make this task easier?

Does it work to stupid-string-split it?

Doesnt sound like it 🙂

@U07SQTAEM doesn't seem so. The page break elements will literally occur anywhere, including in the middle of words. The XML seems to have been created in a single pass. The page break elements correspond exactly to where the source texts are split, so they are inserted in the right places, it's just inconvenient to recreate as anything other than a single page.

Anyway, if I can create some kind of Hiccup/XML split-with function, then I'm sure it can be reused in future projects... Guess I need to get to work.

@U6T7M9DBR Nope, there is an undefined depth at any page break point, so structural information and styling needs to be preserved, e.g. if the split occurs in the middle of a word, in the middle of a sentence, in the middle of a paragraph, and so on, the entire tree up to the root needs to split and elements before and after copied into left-right.

I hope, you can make something better than garbage in, garbage out.