Martin Kleppmann

@schickm Good thinking, I like the idea of treating initial setup as schema, and handling it differently from the data. Food for thought.

This PR is now mostly complete and working: cursors can be created and stored, and indexes are automatically recomputed as needed. There is one known issue: `Automerge.load(Automerge.store(doc))` currently throws an...

Hi @Gozala, thank you for this suggestion. It's already the case that a single Automerge change bundles together any number of individual updates — whatever is made within the scope...

> Could those hashes be derived rather than encoded into the batch? Yes, and the compressed document format does exactly what you suggest: it doesn't store hashes of each individual...

I had in mind the following approach, which tests only the backend, not the frontend (I think that's good because it allows the same fuzz tests to be run against...

Thanks @viktorkalajo for the bug report. This should be possible, sorry to hear that it's not working at the moment. I will look into it, but in the meantime, as...

Yes, that is unfortunately an expected behaviour of the workaround, since it removes the objects from the old location and creates fresh objects (with the old contents) in the new...

@beorn Sorry, the deep cloning workaround is the only solution for now, until the proper move operation lands.

Closing this issue as there is nothing to be done for now. Support for a proper move operation is planned, which will address this issue.

Okay @hcientist, makes sense. Reopening the issue to serve as a kind of documentation.