Why are the .grist backups so big?

We have this doc that we backup from time to time.
It went from 84mb to 900mb to 1200mb in a very short time.
The few images and tables added should not be that big, so I’m wondering why?
Is the .grist file also saving doc history?

Yes, at the moment, Grist files include history of changes. So if, for example, you imported a bunch of data, and even if you then cancelled the import, or deleted the data, the backup file will include that history. The history isn’t indefinite – it gets pruned to about 1000 actions or ~1G of data, but that could explain why the backup is so large.

Would it be always desirable to drop history when saving a backup? I think that would make sense to me.

I suppose history could be useful in some scenarios, perhaps one could choose to download with or without document history. The download dialogue could have a checkbox saying: Download with document history (considerably larger file)
The Checkbox could be un-checked by default.
It would also save your servers some bandwidth/cost.


There was a similar request here: