While mypy can understand nested references like ConfigBool.Key.COLLAPSE_RECENT,
PyCharm doesn't understand the metaclass syntax, and shows the definitions
as invalid.
- anki._backend stores the protobuf files and rsbackend.py code
- pylib modules import protobuf messages directly from the
_pb2 files, and explicitly export any will be returned or consumed
by public pylib functions, so that calling code can import from pylib
- the "rsbackend" no longer imports and re-exports protobuf messages
- pylib can just consume them directly.
- move errors to errors.py
Still todo:
- rsbridge
- finishing the work on rsbackend, and check what we need to add
back to the original file location to avoid breaking add-ons
This was the behaviour before, but got lost when ._reset_counts() was
added. Also added the check back to schedv2:fillRev(), which also
appears to have gotten lost in the move to ._reset_counts()
Just a stop-gap fix until this code can get a proper rewrite.
We want to avoid unburied != self.today, because the unburied time
is synchronized as part of the collection at the moment, and we don't
want a client with an older unburied time that was modified more
recently to cause cards to be unburied twice - so we only unbury
if today is more than 7 days earlier than the last unbury time.
Unsure what caused these users to end up in that state in the first
place:
https://anki.tenderapp.com/discussions/ankidesktop/41335-cards-remain-buried-forever-unless-manually-unburied
- wins back the performance lost by the decks and dconf not being
in memory, and the overhead of serializing data for DB calls
- card counts are no longer capped to 1000
- learn counts are currently still calculated separately - can't merge
v1 counts without changing the existing behaviour
- partially rendering the tree may yield more savings
- the old "mutate but don't save" approach to resetting the "done today"
counts no longer works, and was inefficient anyway - now we just check
the day when returning the count
- remove separate implementation for v1 scheduler
This is a stop-gap solution - a bigger refactor will need to wait
until the deck/note type changes have stabilized.
