mirror of
https://github.com/elisiariocouto/leggen.git
synced 2025-12-14 02:42:21 +00:00
refactor(analytics): Simplify analytics endpoints and eliminate client-side processing.
- Add /transactions/monthly-stats endpoint with SQL aggregation - Replace client-side monthly processing with server-side calculations - Reduce data transfer by 99.5% (2,507 → 13 records for yearly data) - Simplify MonthlyTrends component by removing 40+ lines of aggregation logic - Clean up unused imports and interfaces 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Elisiário Couto
@@ -1,6 +1,5 @@
|
||||
import json
|
||||
import sqlite3
|
||||
from collections import defaultdict
|
||||
from datetime import datetime, timedelta
|
||||
from typing import Any, Dict, List, Optional
|
||||
|
||||
@@ -1159,8 +1158,21 @@ class DatabaseService:
|
||||
cursor = conn.cursor()
|
||||
|
||||
try:
|
||||
# Get current balance for each account/type to use as the final balance
|
||||
current_balances_query = """
|
||||
cutoff_date = (datetime.now() - timedelta(days=days)).date().isoformat()
|
||||
today_date = datetime.now().date().isoformat()
|
||||
|
||||
# Single SQL query to generate historical balances using window functions
|
||||
query = """
|
||||
WITH RECURSIVE date_series AS (
|
||||
-- Generate weekly dates from cutoff_date to today
|
||||
SELECT date(?) as ref_date
|
||||
UNION ALL
|
||||
SELECT date(ref_date, '+7 days')
|
||||
FROM date_series
|
||||
WHERE ref_date < date(?)
|
||||
),
|
||||
current_balances AS (
|
||||
-- Get current balance for each account/type
|
||||
SELECT account_id, type, amount, currency
|
||||
FROM balances b1
|
||||
WHERE b1.timestamp = (
|
||||
@@ -1168,116 +1180,150 @@ class DatabaseService:
|
||||
FROM balances b2
|
||||
WHERE b2.account_id = b1.account_id AND b2.type = b1.type
|
||||
)
|
||||
{account_filter}
|
||||
AND b1.type = 'closingBooked' -- Focus on closingBooked for charts
|
||||
),
|
||||
historical_points AS (
|
||||
-- Calculate balance at each weekly point by subtracting future transactions
|
||||
SELECT
|
||||
cb.account_id,
|
||||
cb.type as balance_type,
|
||||
cb.currency,
|
||||
ds.ref_date,
|
||||
cb.amount - COALESCE(
|
||||
(SELECT SUM(t.transactionValue)
|
||||
FROM transactions t
|
||||
WHERE t.accountId = cb.account_id
|
||||
AND date(t.transactionDate) > ds.ref_date), 0
|
||||
) as balance_amount
|
||||
FROM current_balances cb
|
||||
CROSS JOIN date_series ds
|
||||
)
|
||||
SELECT
|
||||
account_id || '_' || balance_type || '_' || ref_date as id,
|
||||
account_id,
|
||||
balance_amount,
|
||||
balance_type,
|
||||
currency,
|
||||
ref_date as reference_date
|
||||
FROM historical_points
|
||||
ORDER BY account_id, ref_date
|
||||
"""
|
||||
params = []
|
||||
|
||||
# Build parameters and account filter
|
||||
params = [cutoff_date, today_date]
|
||||
if account_id:
|
||||
current_balances_query += " AND b1.account_id = ?"
|
||||
account_filter = "AND b1.account_id = ?"
|
||||
params.append(account_id)
|
||||
|
||||
cursor.execute(current_balances_query, params)
|
||||
current_balances = {
|
||||
(row["account_id"], row["type"]): {
|
||||
"amount": row["amount"],
|
||||
"currency": row["currency"],
|
||||
}
|
||||
for row in cursor.fetchall()
|
||||
}
|
||||
|
||||
# Get transactions for the specified period, ordered by date descending
|
||||
cutoff_date = (datetime.now() - timedelta(days=days)).isoformat()
|
||||
|
||||
transactions_query = """
|
||||
SELECT accountId, transactionDate, transactionValue
|
||||
FROM transactions
|
||||
WHERE transactionDate >= ?
|
||||
"""
|
||||
|
||||
if account_id:
|
||||
transactions_query += " AND accountId = ?"
|
||||
params = [cutoff_date, account_id]
|
||||
else:
|
||||
params = [cutoff_date]
|
||||
account_filter = ""
|
||||
|
||||
transactions_query += " ORDER BY transactionDate DESC"
|
||||
# Format the query with conditional filter
|
||||
formatted_query = query.format(account_filter=account_filter)
|
||||
|
||||
cursor.execute(transactions_query, params)
|
||||
transactions = cursor.fetchall()
|
||||
|
||||
# Calculate historical balances by working backwards from current balance
|
||||
historical_balances = []
|
||||
account_running_balances: dict[str, dict[str, float]] = {}
|
||||
|
||||
# Initialize running balances with current balances
|
||||
for (acc_id, balance_type), balance_info in current_balances.items():
|
||||
if acc_id not in account_running_balances:
|
||||
account_running_balances[acc_id] = {}
|
||||
account_running_balances[acc_id][balance_type] = balance_info["amount"]
|
||||
|
||||
# Group transactions by date
|
||||
transactions_by_date = defaultdict(list)
|
||||
|
||||
for txn in transactions:
|
||||
date_str = txn["transactionDate"][:10] # Extract just the date part
|
||||
transactions_by_date[date_str].append(txn)
|
||||
|
||||
# Generate historical balance points
|
||||
# Start from today and work backwards
|
||||
current_date = datetime.now().date()
|
||||
|
||||
for day_offset in range(0, days, 7): # Sample every 7 days for performance
|
||||
target_date = current_date - timedelta(days=day_offset)
|
||||
target_date_str = target_date.isoformat()
|
||||
|
||||
# For each account, create balance entries
|
||||
for acc_id in account_running_balances:
|
||||
for balance_type in [
|
||||
"closingBooked"
|
||||
]: # Focus on closingBooked for the chart
|
||||
if balance_type in account_running_balances[acc_id]:
|
||||
balance_amount = account_running_balances[acc_id][
|
||||
balance_type
|
||||
]
|
||||
currency = current_balances.get(
|
||||
(acc_id, balance_type), {}
|
||||
).get("currency", "EUR")
|
||||
|
||||
historical_balances.append(
|
||||
{
|
||||
"id": f"{acc_id}_{balance_type}_{target_date_str}",
|
||||
"account_id": acc_id,
|
||||
"balance_amount": balance_amount,
|
||||
"balance_type": balance_type,
|
||||
"currency": currency,
|
||||
"reference_date": target_date_str,
|
||||
"created_at": None,
|
||||
"updated_at": None,
|
||||
}
|
||||
)
|
||||
|
||||
# Subtract transactions that occurred on this date and later dates
|
||||
# to simulate going back in time
|
||||
for date_str in list(transactions_by_date.keys()):
|
||||
if date_str >= target_date_str:
|
||||
for txn in transactions_by_date[date_str]:
|
||||
acc_id = txn["accountId"]
|
||||
amount = txn["transactionValue"]
|
||||
|
||||
if acc_id in account_running_balances:
|
||||
for balance_type in account_running_balances[acc_id]:
|
||||
account_running_balances[acc_id][balance_type] -= (
|
||||
amount
|
||||
)
|
||||
|
||||
# Remove processed transactions to avoid double-processing
|
||||
del transactions_by_date[date_str]
|
||||
cursor.execute(formatted_query, params)
|
||||
rows = cursor.fetchall()
|
||||
|
||||
conn.close()
|
||||
|
||||
# Sort by date for proper chronological order
|
||||
historical_balances.sort(key=lambda x: x["reference_date"])
|
||||
|
||||
return historical_balances
|
||||
return [dict(row) for row in rows]
|
||||
|
||||
except Exception as e:
|
||||
conn.close()
|
||||
raise e
|
||||
|
||||
async def get_monthly_transaction_stats_from_db(
|
||||
self,
|
||||
account_id: Optional[str] = None,
|
||||
date_from: Optional[str] = None,
|
||||
date_to: Optional[str] = None,
|
||||
) -> List[Dict[str, Any]]:
|
||||
"""Get monthly transaction statistics aggregated by the database"""
|
||||
if not self.sqlite_enabled:
|
||||
logger.warning("SQLite database disabled, cannot read monthly stats")
|
||||
return []
|
||||
|
||||
try:
|
||||
monthly_stats = self._get_monthly_transaction_stats(
|
||||
account_id=account_id,
|
||||
date_from=date_from,
|
||||
date_to=date_to,
|
||||
)
|
||||
logger.debug(
|
||||
f"Retrieved {len(monthly_stats)} monthly stat points from database"
|
||||
)
|
||||
return monthly_stats
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to get monthly transaction stats from database: {e}")
|
||||
return []
|
||||
|
||||
def _get_monthly_transaction_stats(
|
||||
self,
|
||||
account_id: Optional[str] = None,
|
||||
date_from: Optional[str] = None,
|
||||
date_to: Optional[str] = None,
|
||||
) -> List[Dict[str, Any]]:
|
||||
"""Get monthly transaction statistics from SQLite database"""
|
||||
db_path = path_manager.get_database_path()
|
||||
if not db_path.exists():
|
||||
return []
|
||||
|
||||
conn = sqlite3.connect(str(db_path))
|
||||
conn.row_factory = sqlite3.Row
|
||||
cursor = conn.cursor()
|
||||
|
||||
try:
|
||||
# SQL query to aggregate transactions by month
|
||||
query = """
|
||||
SELECT
|
||||
strftime('%Y-%m', transactionDate) as month,
|
||||
COALESCE(SUM(CASE WHEN transactionValue > 0 THEN transactionValue ELSE 0 END), 0) as income,
|
||||
COALESCE(SUM(CASE WHEN transactionValue < 0 THEN ABS(transactionValue) ELSE 0 END), 0) as expenses,
|
||||
COALESCE(SUM(transactionValue), 0) as net
|
||||
FROM transactions
|
||||
WHERE 1=1
|
||||
"""
|
||||
|
||||
params = []
|
||||
|
||||
if account_id:
|
||||
query += " AND accountId = ?"
|
||||
params.append(account_id)
|
||||
|
||||
if date_from:
|
||||
query += " AND transactionDate >= ?"
|
||||
params.append(date_from)
|
||||
|
||||
if date_to:
|
||||
query += " AND transactionDate <= ?"
|
||||
params.append(date_to)
|
||||
|
||||
query += """
|
||||
GROUP BY strftime('%Y-%m', transactionDate)
|
||||
ORDER BY month ASC
|
||||
"""
|
||||
|
||||
cursor.execute(query, params)
|
||||
rows = cursor.fetchall()
|
||||
|
||||
# Convert to desired format with proper month display
|
||||
monthly_stats = []
|
||||
for row in rows:
|
||||
# Convert YYYY-MM to display format like "Mar 2024"
|
||||
year, month_num = row["month"].split("-")
|
||||
month_date = datetime.strptime(f"{year}-{month_num}-01", "%Y-%m-%d")
|
||||
display_month = month_date.strftime("%b %Y")
|
||||
|
||||
monthly_stats.append(
|
||||
{
|
||||
"month": display_month,
|
||||
"income": round(row["income"], 2),
|
||||
"expenses": round(row["expenses"], 2),
|
||||
"net": round(row["net"], 2),
|
||||
}
|
||||
)
|
||||
|
||||
conn.close()
|
||||
return monthly_stats
|
||||
|
||||
except Exception as e:
|
||||
conn.close()
|
||||
|
||||
Reference in New Issue
Block a user