kopia lustrzana https://dev.funkwhale.audio/funkwhale/funkwhale
52 wiersze
1.8 KiB
Python
52 wiersze
1.8 KiB
Python
from django.db import models
|
|
|
|
from funkwhale_api.common import fields
|
|
from funkwhale_api.favorites.models import TrackFavorite
|
|
from funkwhale_api.history.models import Listening
|
|
|
|
|
|
def combined_recent(limit, **kwargs):
|
|
datetime_field = kwargs.pop("datetime_field", "creation_date")
|
|
source_querysets = {qs.model._meta.label: qs for qs in kwargs.pop("querysets")}
|
|
querysets = {
|
|
k: qs.annotate(
|
|
__type=models.Value(qs.model._meta.label, output_field=models.CharField())
|
|
).values("pk", datetime_field, "__type")
|
|
for k, qs in source_querysets.items()
|
|
}
|
|
_qs_list = list(querysets.values())
|
|
union_qs = _qs_list[0].union(*_qs_list[1:])
|
|
records = []
|
|
for row in union_qs.order_by("-{}".format(datetime_field))[:limit]:
|
|
records.append(
|
|
{"type": row["__type"], "when": row[datetime_field], "pk": row["pk"]}
|
|
)
|
|
# Now we bulk-load each object type in turn
|
|
to_load = {}
|
|
for record in records:
|
|
to_load.setdefault(record["type"], []).append(record["pk"])
|
|
fetched = {}
|
|
|
|
for key, pks in to_load.items():
|
|
for item in source_querysets[key].filter(pk__in=pks):
|
|
fetched[(key, item.pk)] = item
|
|
|
|
# Annotate 'records' with loaded objects
|
|
for record in records:
|
|
record["object"] = fetched[(record["type"], record["pk"])]
|
|
return records
|
|
|
|
|
|
def get_activity(user, limit=20):
|
|
query = fields.privacy_level_query(user, lookup_field="user__privacy_level")
|
|
querysets = [
|
|
Listening.objects.filter(query).select_related(
|
|
"track", "user", "track__artist", "track__album__artist"
|
|
),
|
|
TrackFavorite.objects.filter(query).select_related(
|
|
"track", "user", "track__artist", "track__album__artist"
|
|
),
|
|
]
|
|
records = combined_recent(limit=limit, querysets=querysets)
|
|
return [r["object"] for r in records]
|