Refactor: Separate data sync from UI rendering, change colunm width

This commit is contained in:
2026-01-27 07:12:09 +08:00
parent 61b32bbdd7
commit 9e7f474d5e
5 changed files with 1691 additions and 144 deletions
+177 -28
View File
@@ -1,48 +1,179 @@
import pandas as pd
import requests
import os
from datetime import datetime
import shutil
from datetime import datetime, time
from ta.trend import EMAIndicator
from ta.momentum import StochasticOscillator
class DataEngine:
def __init__(self, symbol, url, provider):
def __init__(self, symbol=None, url=None, provider=None, data_dir='data_cache'):
self.symbol = symbol
self.url = url
self.provider = provider
# 1. Get the folder where engine.py lives
# Use your robust path logic
base_path = os.path.dirname(os.path.abspath(__file__))
# 2. Define the cache directory path
self.cache_dir = os.path.join(base_path, "data_cache")
# 3. Create the folder if it doesn't exist (safety-first)
self.cache_dir = os.path.join(base_path, data_dir) # Use data_dir variable
os.makedirs(self.cache_dir, exist_ok=True)
# 4. Set the full path for this specific instrument's CSV
self.file_path = os.path.join(self.cache_dir, f"{self.symbol}.csv")
# 4. Only set file_path if we actually have a symbol
if self.symbol:
self.file_path = os.path.join(self.cache_dir, f"{self.symbol}.csv")
else:
self.file_path = None
def global_sync(self):
"""The 'One-Click' background loop."""
# 1. Get the latest list of instruments from your CSV
all_instruments = self.load_instruments_from_csv()
def load_instruments_from_csv(self, file_path):
import csv
instruments = []
for item in all_instruments:
# 2. Update the 'Current' target for the engine
self.symbol = item['symbol']
self.cusip = item['cusip']
self.provider = item['provider']
# Updated templates for maximum historical reach
TEMPLATES = {
'jpm': "https://am.jpmorgan.com/FundsMarketingHandler/historicalData?cusip={cusip}&country=hk&role=per",
# period1=0 fetches from the earliest available date; interval=1d is daily
'yahoo': "https://query1.finance.yahoo.com/v8/finance/chart/{cusip}?period1=0&period2=9999999999&interval=1d&events=history",
# FT remains 30-day window; Smart Append logic in fetch_data handles the history
'agi': "https://markets.ft.com/data/funds/tearsheet/historical?s={cusip}"
}
try:
abs_path = os.path.join(os.path.dirname(__file__), file_path)
# 3. Regenerate the URL and File Path for THIS specific instrument
self.url = self.generate_url()
self.file_path = os.path.join(self.data_dir, f"{self.symbol}.csv")
# 4. Run the robust fetch/merge logic we built
print(f"Syncing {self.symbol}...")
self.fetch_data()
print("Global Sync Complete.")
if not os.path.exists(abs_path):
print(f"Error: {file_path} not found.")
return []
with open(abs_path, mode='r', encoding='utf-8-sig') as csvfile:
reader = csv.DictReader(csvfile)
reader.fieldnames = [name.strip().lower() for name in reader.fieldnames]
for row in reader:
symbol = row.get('symbol', '').strip()
cusip = row.get('cusip', '').strip()
provider = row.get('provider', 'jpm').strip().lower()
if symbol and cusip:
template = TEMPLATES.get(provider, TEMPLATES['jpm'])
url = template.format(cusip=cusip)
instruments.append({
"symbol": symbol,
"url": url,
"provider": provider,
"cusip": cusip # Added this so sync_all can use it if needed
})
except Exception as e:
print(f"CSV Loading Error: {e}")
return instruments
# URL_CONFIG = load_instruments_from_csv('instruments.csv')
def global_sync(self):
"""Backup, Sync all instruments, and return a summary report."""
# 1. Run Maintenance/Backup
self.run_pre_sync_maintenance()
# FIX 1: Add 'self.' so it calls the method inside this class
instruments = self.load_instruments_from_csv('instruments.csv')
report = {
"total": len(instruments),
"updated": 0,
"failed": 0,
"details": []
}
for item in instruments:
try:
self.symbol = item['symbol']
self.provider = item['provider']
self.url = item['url']
# FIX 2: Use 'self.cache_dir' to match your __init__ logic
self.file_path = os.path.join(self.cache_dir, f"{self.symbol}.csv")
print(f"Updating {self.symbol}...")
# fetch_data now returns the updated DataFrame or None
result_df = self.fetch_data()
time.sleep(1)
if result_df is not None and not result_df.empty:
report["updated"] += 1
last_price = result_df['close'].iloc[-1]
report["details"].append(f"{self.symbol}: Updated (Price: {last_price})")
else:
report["failed"] += 1
report["details"].append(f"{self.symbol}: No new data found")
except Exception as e:
report["failed"] += 1
report["details"].append(f"⚠️ {self.symbol}: Error ({str(e)})")
return report
def run_pre_sync_maintenance(self):
"""Backs up files and reports current data health."""
import os
import shutil
import pandas as pd
from datetime import datetime
# 1. Setup paths correctly
base_dir = os.path.dirname(os.path.abspath(__file__))
backup_dir = os.path.join(base_dir, 'backups')
# 2. Create the timestamped folder path FIRST
timestamp = datetime.now().strftime("%Y%m%d_%H%M")
current_backup_path = os.path.join(backup_dir, f"sync_backup_{timestamp}")
# 3. Create the directories (safety-first)
os.makedirs(current_backup_path, exist_ok=True)
print(f"\n--- Pre-Sync Health Check ({timestamp}) ---")
stats = []
# 4. Check if cache exists to avoid errors
if not os.path.exists(self.cache_dir):
print(f"⚠️ Cache directory not found at {self.cache_dir}")
return pd.DataFrame()
# 5. Backup loop
for filename in os.listdir(self.cache_dir):
if filename.endswith(".csv"):
src = os.path.join(self.cache_dir, filename)
dst = os.path.join(current_backup_path, filename)
try:
# Perform copy
shutil.copy2(src, dst)
# Read data for health check
df = pd.read_csv(src)
# Store stats
stats.append({
"Fund": filename.replace(".csv", ""),
"Rows": len(df),
"Start": df['date'].min() if 'date' in df.columns else "N/A",
"End": df['date'].max() if 'date' in df.columns else "N/A"
})
print(f"📦 Backed up: {filename} ({len(df)} rows)")
except Exception as e:
print(f"⚠️ Could not backup {filename}: {e}")
continue
# 6. Display and return report
if stats:
stats_df = pd.DataFrame(stats)
print("\n" + stats_df.to_string(index=False))
print(f"\n✅ All backups saved to: {current_backup_path}")
return stats_df
else:
print("📭 No CSV files found to backup.")
return pd.DataFrame()
def _parse_jpm(self, json_data):
if isinstance(json_data, dict) and "historicalNAVList" in json_data:
@@ -165,6 +296,24 @@ class DataEngine:
print(f"Network error for {self.symbol}: {e}")
return local_df
def get_local_metrics(self):
"""Reads ONLY from local CSV and returns metrics immediately."""
if not os.path.exists(self.file_path):
return {"error": "Missing Local Data", "status": "needs_sync"}
try:
df = pd.read_csv(self.file_path)
# Ensure columns are clean
df.columns = [c.lower().strip() for c in df.columns]
df['date'] = pd.to_datetime(df['date'], errors='coerce')
df = df.dropna(subset=['date', 'close']).sort_values('date')
# Pass this local dataframe to your existing calculation function
return self.calculate_table_metrics(df)
except Exception as e:
print(f"Error reading local data for {self.symbol}: {e}")
return None
def calculate_table_metrics(self, df):
if df is None or df.empty or len(df) < 2: