Hi all,
When the model files on disk are changed outside the HTML Architect - a branch switch, a pull, a rebase, or an edit made in another tool - the running Architect keeps showing the model it loaded at startup. The only reliable way to get back in sync is to restart the server, which is disruptive during everyday work and especially painful when switching branches often.
The backend already watches the model directory for changes, so the information that something changed on disk exists somewhere in the system. What seems to be missing is a path from there to the user: the web UI has no way of learning about it and no way of acting on it.
I’d like to look into this and see what is actually feasible. Two directions come to mind, and the right one probably depends on what the existing machinery can support:
Notify and ask - the Architect tells the user that the model on disk has changed and offers to reload it, leaving the decision (and any unsaved work) in the user’s hands.
Reload automatically - the Architect refreshes itself when it is safe to do so, and only falls back to asking when there are unsaved changes that would be lost.
A simple manual “Reload model” action might also be worth having on its own, independently of any detection, as a predictable escape hatch.
Before going further I’d like to know whether this direction fits how the maintainers see the Architect evolving. Is there a reason external changes were deliberately left out of scope - unsaved-state safety, multi-user concerns, or something else I’m not seeing?
Thanks!