You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The properties of a snapshot are currently of type Hashmap<String, json::Value>. This type can be bridged into pythojn but there is no direct mapping currently.
The properties of a snapshot are currently not stored with the metadata. This means they are not stored inside of children so if a children wants the ancestry including the properties of each commit, they must fetch the full snapshot for each snapshot in the history to do so. If we store the properties in the metadata` that has the side effect of potentially bloating the size of the snapshots because each snapshot holds the metadata of its ancestors
One extra point: even if the spec allows to keep only part of the history in each snapshot, the current implementation keeps all of them. So every snapshot gets larger as the history tree grows. This is very much something we can improve on, but it's where we are at today.
In rust, users can associate extract data with commits using the properties parameter. This should be exposed to the python session
The text was updated successfully, but these errors were encountered: