QuivrHQ/quivr
1
1
Fork 0
mirror of https://github.com/QuivrHQ/quivr.git synced 2024-08-17 02:10:31 +03:00
Code Issues Packages Projects Releases Wiki Activity

feat(analytics): rely on sql rather that python loop for brains

Browse Source
This commit is contained in:
Stan Girard 2024-04-17 16:16:13 +02:00
parent b62297341f
commit f0ad664673
1 changed files with 26 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
backend/modules/analytics/repository/analytics.py
View File

@ -2,46 +2,55 @@ from collections import defaultdict
from datetime import datetime, timedelta
from typing import Optional
from uuid import UUID
from models.settings import get_supabase_client
from modules.analytics.entity.analytics import BrainsUsages, Range, Usage
from modules.brain.service.brain_user_service import BrainUserService
brain_user_service = BrainUserService()
class Analytics:
def __init__(self):
supabase_client = get_supabase_client()
self.db = supabase_client
def get_brains_usages(self, user_id: UUID, graph_range: Range, brain_id: Optional[UUID] = None) -> BrainsUsages:
def get_brains_usages(
self, user_id: UUID, graph_range: Range, brain_id: Optional[UUID] = None
) -> BrainsUsages:
user_brains = brain_user_service.get_user_brains(user_id)
if brain_id is not None:
user_brains = [brain for brain in user_brains if brain.id == brain_id]
usage_per_day = defaultdict(int)
usage_per_day = defaultdict(int)
for brain in user_brains:
chat_history = (
self.db.from_("chat_history")
.select("*")
.filter("brain_id", "eq", str(brain.id))
.execute()
).data
brain_ids = [brain.id for brain in user_brains]
chat_history = (
self.db.from_("chat_history")
.select("*")
.in_("brain_id", brain_ids)
.execute()
).data
for chat in chat_history:
message_time = datetime.strptime(chat['message_time'], "%Y-%m-%dT%H:%M:%S.%f")
usage_per_day[message_time.date()] += 1
for chat in chat_history:
message_time = datetime.strptime(
chat["message_time"], "%Y-%m-%dT%H:%M:%S.%f"
)
usage_per_day[message_time.date()] += 1
start_date = datetime.now().date() - timedelta(days=graph_range)
all_dates = [start_date + timedelta(days=i) for i in range(graph_range)]
for date in all_dates:
usage_per_day[date] += 0
usage_per_day[date] += 0
usages = sorted(
[Usage(date=date, usage_count=count) for date, count in usage_per_day.items() if start_date <= date <= datetime.now().date()],
key=lambda usage: usage.date
[
Usage(date=date, usage_count=count)
for date, count in usage_per_day.items()
if start_date <= date <= datetime.now().date()
],
key=lambda usage: usage.date,
)
return BrainsUsages(usages=usages)