Commit inicial: Projeto YuClusters

This commit is contained in:
Thiago Moura
2026-06-04 18:12:47 -03:00
commit 0a185c0223
28 changed files with 5093 additions and 0 deletions
+32
View File
@@ -0,0 +1,32 @@
# Node.js
node_modules/
dist/
build/
.env
.env.local
# Python
__pycache__/
*.pyc
*.pyo
*.pyd
.Python
env/
venv/
.venv/
env.bak/
venv.bak/
# Logs and databases
*.log
*.sqlite
*.db
# OS generated files
.DS_Store
.DS_Store?
._*
.Spotlight-V100
.Trashes
ehthumbs.db
Thumbs.db
+67
View File
@@ -0,0 +1,67 @@
# YuClusters Local 📊
YuClusters Local is a professional footprint chart and volumetric cluster analyzer integrated directly with your local **MetaTrader 5 (MT5)** terminal.
## Key Features
- **Real-Time Footprint Charting**: Ticks polled at ~100ms intervals and visualised dynamically using an optimized HTML5 Canvas element.
- **Diagonal Imbalances Detection**: Identifies aggressive buying or selling imbalances ($R \ge 3.0$).
- **Stacked Imbalances Highlights**: Dynamically spots 3+ consecutive imbalances in the same direction.
- **POC Highlight**: Identifies and highlights the highest volume nodes within each cluster using a gold border.
- **Interactive Drag & Pan UI**: Fully-featured HTML5 Canvas dashboard with drag-to-scroll horizontal/vertical panning and zoom adjustment.
---
## Installation & Setup
### 1. Prerequisites
- **Windows OS** (required by MetaTrader 5 API)
- **MetaTrader 5 Terminal** running locally and logged into your broker account
- **Python 3.11+** installed and added to PATH
- **Node.js 18+** installed
### 2. Backend Installation
1. From the project root, install Python dependencies:
```powershell
pip install -r requirements.txt
```
2. Set your environment configurations in `config/settings.py` or export them as environment variables (e.g. `MT5_LOGIN`, `MT5_PASSWORD`, `MT5_SERVER`).
### 3. Frontend Installation
1. Navigate to the `frontend/` folder:
```powershell
cd frontend
```
2. Install npm packages:
```powershell
npm install
```
---
## Execution Guide
### 1. Run the Backend Server
Start the FastAPI server (this will automatically launch the MetaTrader 5 polling collector):
```powershell
python backend/server.py
```
### 2. Run the React Frontend
Start the Vite development server:
```powershell
cd frontend
npm run dev
```
Open [http://localhost:3000](http://localhost:3000) in your web browser.
---
## Technical Calculations (from SKILL.md)
1. **Diagonal Buy Imbalance**: `ask_vol[i] >= R * bid_vol[i-1]` (compares with price level below)
2. **Diagonal Sell Imbalance**: `bid_vol[i] >= R * ask_vol[i+1]` (compares with price level above)
3. **POC (Point of Control)**: Price level with the highest total volume (`ask_vol + bid_vol`) inside a cluster. If volumes tie, the highest price wins.
4. **Stacked Imbalance Zone**: A vertical zone spanning 3 or more consecutive levels with the same imbalance direction.
+16
View File
@@ -0,0 +1,16 @@
# SKILL.md — YuClusters Local
## Regras de Cálculo (NÃO ALTERAR)
1. Imbalance diagonal BUY no nível i: ask_vol[i] >= R * bid_vol[i-1] (compara com nível ABAIXO)
2. Imbalance diagonal SELL no nível i: bid_vol[i] >= R * ask_vol[i+1] (compara com nível ACIMA)
3. POC = nível com MAIOR (ask_vol + bid_vol) dentro do cluster
4. Delta do cluster = soma de todos os delta[i] de todos os níveis
5. Stacked = mínimo de STACKED_MIN_COUNT níveis consecutivos com imbalance na MESMA direção
## Regras de Implementação
- NUNCA misturar lógica de agregação com lógica de UI
- NUNCA recalcular clusters históricos fechados
- SEMPRE usar tick_size do MT5 como granularidade mínima de nível de preço
- Conexão MT5 SEMPRE via lib oficial `MetaTrader5`, nunca via subprocess ou API REST externa
+389
View File
@@ -0,0 +1,389 @@
import uuid
import time
from typing import Dict, Any, List, Optional
from config import settings
class FootprintCluster:
def __init__(self, tick_size: float = 1.0):
self.tick_size = tick_size
self.cluster_id = str(uuid.uuid4())
self.status = "active"
self.open_time: Optional[int] = None
self.close_reason: Optional[str] = None
self.open_price: Optional[float] = None
self.close_price: Optional[float] = None
self.high: Optional[float] = None
self.low: Optional[float] = None
self.poc: Optional[float] = None
self.total_delta: float = 0.0
self.total_volume: float = 0.0
# levels: price_float -> { 'ask': float, 'bid': float }
# internally we keep float keys to make sorting and arithmetic easy
self.levels: Dict[float, Dict[str, float]] = {}
self.stacked: Dict[str, Any] = {
"buy": False,
"sell": False,
"price_range": []
}
self.advanced_metrics: Dict[str, Any] = {
"poc_position": None,
"pattern": None,
"top_extreme": None,
"bottom_extreme": None
}
def add_tick(self, price: float, volume: float, is_buy: bool, timestamp_msc: int) -> None:
if self.open_time is None:
self.open_time = timestamp_msc
self.open_price = price
self.close_price = price
# Round price to the nearest tick_size to avoid float precision issues
rounded_price = round(price / self.tick_size) * self.tick_size
if rounded_price not in self.levels:
self.levels[rounded_price] = {"ask": 0.0, "bid": 0.0}
if is_buy:
self.levels[rounded_price]["ask"] += volume
else:
self.levels[rounded_price]["bid"] += volume
# Update High/Low
if self.high is None or rounded_price > self.high:
self.high = rounded_price
if self.low is None or rounded_price < self.low:
self.low = rounded_price
# Recalculate totals, POC, imbalances and stacked imbalances
self._recalculate()
def _recalculate(self) -> None:
if not self.levels:
return
self.total_volume = 0.0
self.total_delta = 0.0
# First pass: calc totals and delta per level
for price, data in self.levels.items():
ask = data["ask"]
bid = data["bid"]
lvl_delta = ask - bid
lvl_total = ask + bid
self.total_volume += lvl_total
self.total_delta += lvl_delta
# Find POC: level with the highest total volume.
# Tie-breaker: choose the highest price level.
sorted_prices = sorted(self.levels.keys())
best_price = sorted_prices[0]
max_total = -1.0
for price in sorted_prices:
total_vol = self.levels[price]["ask"] + self.levels[price]["bid"]
if total_vol > max_total:
max_total = total_vol
best_price = price
elif total_vol == max_total:
if price > best_price:
best_price = price
self.poc = best_price
# Second pass: calculate imbalances diagonal and stacked imbalances
# To avoid division by zero:
# imbalance_buy[i] is True if ask_vol[i] >= R * bid_vol[i - tick_size]
# imbalance_sell[i] is True if bid_vol[i] >= R * ask_vol[i + tick_size]
R = settings.IMBALANCE_RATIO
imbalances_buy = {}
imbalances_sell = {}
for price in sorted_prices:
ask_val = self.levels[price]["ask"]
bid_val = self.levels[price]["bid"]
# Lower level price
lower_price = round((price - self.tick_size) / self.tick_size) * self.tick_size
bid_below = self.levels[lower_price]["bid"] if lower_price in self.levels else 0.0
# Upper level price
upper_price = round((price + self.tick_size) / self.tick_size) * self.tick_size
ask_above = self.levels[upper_price]["ask"] if upper_price in self.levels else 0.0
# Buy Imbalance (diagonal): ask vs bid_below
# Avoid triggering on 0 vs 0
if ask_val > 0 and ask_val >= R * bid_below:
imbalances_buy[price] = True
else:
imbalances_buy[price] = False
# Sell Imbalance (diagonal): bid vs ask_above
if bid_val > 0 and bid_val >= R * ask_above:
imbalances_sell[price] = True
else:
imbalances_sell[price] = False
# Detect stacked imbalances: 3+ consecutive levels with imbalance in the same direction
# Let's check contiguous price levels in steps of tick_size
stacked_buy = False
stacked_sell = False
stacked_buy_prices = []
stacked_sell_prices = []
min_consecutive = settings.STACKED_MIN_COUNT
# We need to check all possible price steps from low to high
if self.low is not None and self.high is not None:
current_price = self.low
consec_buy = []
consec_sell = []
while current_price <= self.high:
rounded_p = round(current_price / self.tick_size) * self.tick_size
# Check Buy
if imbalances_buy.get(rounded_p, False):
consec_buy.append(rounded_p)
else:
if len(consec_buy) >= min_consecutive:
stacked_buy = True
stacked_buy_prices.extend(consec_buy)
consec_buy = []
# Check Sell
if imbalances_sell.get(rounded_p, False):
consec_sell.append(rounded_p)
else:
if len(consec_sell) >= min_consecutive:
stacked_sell = True
stacked_sell_prices.extend(consec_sell)
consec_sell = []
current_price += self.tick_size
# Final check at the end of the loop
if len(consec_buy) >= min_consecutive:
stacked_buy = True
stacked_buy_prices.extend(consec_buy)
if len(consec_sell) >= min_consecutive:
stacked_sell = True
stacked_sell_prices.extend(consec_sell)
# Build output properties for stacked
self.stacked = {
"buy": stacked_buy,
"sell": stacked_sell,
"price_range": sorted(list(set(stacked_buy_prices + stacked_sell_prices)))
}
# Store imbalances back in levels for JSON output
for price in sorted_prices:
imb = None
if imbalances_buy.get(price, False) and imbalances_sell.get(price, False):
imb = "both"
elif imbalances_buy.get(price, False):
imb = "buy"
elif imbalances_sell.get(price, False):
imb = "sell"
self.levels[price]["delta"] = self.levels[price]["ask"] - self.levels[price]["bid"]
self.levels[price]["total"] = self.levels[price]["ask"] + self.levels[price]["bid"]
self.levels[price]["imbalance"] = imb
# Advanced Metrics Calculation
if self.high is not None and self.low is not None and self.high > self.low and self.total_volume > 0:
# 1. POC Position
poc_percent = (self.poc - self.low) / (self.high - self.low)
if poc_percent >= 0.65:
self.advanced_metrics["poc_position"] = "top"
elif poc_percent <= 0.35:
self.advanced_metrics["poc_position"] = "bottom"
else:
self.advanced_metrics["poc_position"] = "middle"
# 2. P and B Patterns
mid_price = (self.high + self.low) / 2.0
vol_above = sum(self.levels[p]["total"] for p in sorted_prices if p >= mid_price)
vol_below = sum(self.levels[p]["total"] for p in sorted_prices if p < mid_price)
if vol_above / self.total_volume > 0.65:
self.advanced_metrics["pattern"] = "P"
elif vol_below / self.total_volume > 0.65:
self.advanced_metrics["pattern"] = "B"
else:
self.advanced_metrics["pattern"] = "normal"
# 3. Extremes (Exhaustion / Absorption)
top_vol = self.levels[self.high]["total"]
next_top_price = round((self.high - self.tick_size) / self.tick_size) * self.tick_size
next_top_vol = self.levels.get(next_top_price, {}).get("total", 0.0)
avg_vol = self.total_volume / len(self.levels)
if top_vol < avg_vol * 0.2 and next_top_vol > avg_vol * 0.5:
self.advanced_metrics["top_extreme"] = "exhaustion"
elif top_vol > avg_vol * 2.5:
self.advanced_metrics["top_extreme"] = "absorption"
else:
self.advanced_metrics["top_extreme"] = "normal"
bottom_vol = self.levels[self.low]["total"]
next_bot_price = round((self.low + self.tick_size) / self.tick_size) * self.tick_size
next_bot_vol = self.levels.get(next_bot_price, {}).get("total", 0.0)
if bottom_vol < avg_vol * 0.2 and next_bot_vol > avg_vol * 0.5:
self.advanced_metrics["bottom_extreme"] = "exhaustion"
elif bottom_vol > avg_vol * 2.5:
self.advanced_metrics["bottom_extreme"] = "absorption"
else:
self.advanced_metrics["bottom_extreme"] = "normal"
# 4. Ratios (Exhaustion/Absorption quantification at extremes)
if next_top_vol > 0:
self.advanced_metrics["high_ratio"] = round(top_vol / next_top_vol, 2)
else:
self.advanced_metrics["high_ratio"] = round(top_vol, 2) if top_vol > 0 else 0.0
if next_bot_vol > 0:
self.advanced_metrics["low_ratio"] = round(bottom_vol / next_bot_vol, 2)
else:
self.advanced_metrics["low_ratio"] = round(bottom_vol, 2) if bottom_vol > 0 else 0.0
# 5. Delta Divergence (Price direction vs Order Flow Delta)
self.advanced_metrics["delta_divergence"] = False
if self.open_price is not None and self.close_price is not None:
is_bull = self.close_price > self.open_price
is_bear = self.close_price < self.open_price
if (is_bull and self.total_delta < 0) or (is_bear and self.total_delta > 0):
self.advanced_metrics["delta_divergence"] = True
def should_close(self, current_time_msc: int) -> Optional[str]:
# Check range
if self.high is not None and self.low is not None:
points = self.high - self.low
# If tick_size is 1.0, range is in points. Let's compare directly.
# (or points/tick_size >= CLUSTER_RANGE_POINTS)
# Standard MT5 points. If range in points exceeds setting:
if points >= settings.CLUSTER_RANGE_POINTS:
return "range"
# Check volume
if self.total_volume >= settings.CLUSTER_VOLUME_MAX:
return "volume"
# Check delta
if abs(self.total_delta) >= settings.CLUSTER_DELTA_MAX:
return "delta"
# Check time
if self.open_time is not None:
elapsed = (current_time_msc - self.open_time) / 1000.0
if elapsed >= settings.CLUSTER_TIME_SECONDS:
return "time"
return None
def close(self, reason: str) -> None:
self.status = "closed"
self.close_reason = reason
def to_json(self) -> Dict[str, Any]:
# Convert keys in levels to string for JSON compatibility
levels_str = {}
for price, data in sorted(self.levels.items(), reverse=True):
levels_str[f"{price:.5f}"] = {
"ask": float(data.get("ask", 0.0)),
"bid": float(data.get("bid", 0.0)),
"delta": float(data.get("delta", 0.0)),
"total": float(data.get("total", 0.0)),
"imbalance": data.get("imbalance", None)
}
return {
"cluster_id": self.cluster_id,
"tick_size": self.tick_size,
"status": self.status,
"open_time": int(self.open_time) if self.open_time is not None else None,
"close_reason": self.close_reason,
"open_price": float(self.open_price) if self.open_price is not None else None,
"close_price": float(self.close_price) if self.close_price is not None else None,
"high": float(self.high) if self.high is not None else None,
"low": float(self.low) if self.low is not None else None,
"poc": float(self.poc) if self.poc is not None else None,
"total_delta": float(self.total_delta),
"total_volume": float(self.total_volume),
"levels": levels_str,
"stacked": {
"buy": bool(self.stacked.get("buy", False)),
"sell": bool(self.stacked.get("sell", False)),
"price_range": [float(p) for p in self.stacked.get("price_range", [])]
},
"advanced_metrics": self.advanced_metrics
}
class Aggregator:
def __init__(self, tick_size: float = 1.0):
self.tick_size = tick_size
self.active_cluster = FootprintCluster(tick_size=self.tick_size)
self.history: List[Dict[str, Any]] = []
def process_tick(self, price: float, volume: float, is_buy: bool, timestamp_msc: int) -> tuple[Dict[str, Any], Optional[Dict[str, Any]]]:
"""
Process a single tick.
Returns:
(active_cluster_json, closed_cluster_json_if_just_closed)
"""
self.active_cluster.add_tick(price, volume, is_buy, timestamp_msc)
# Check if the active cluster should be closed
close_reason = self.active_cluster.should_close(timestamp_msc)
closed_json = None
if close_reason:
self.active_cluster.close(close_reason)
closed_json = self.active_cluster.to_json()
# Save to history
self.history.append(closed_json)
if len(self.history) > settings.HISTORY_BUFFER_SIZE:
self.history.pop(0)
# Start new cluster
self.active_cluster = FootprintCluster(tick_size=self.tick_size)
# The next tick will set the open_time of the new cluster.
return self.active_cluster.to_json(), closed_json
def classify_tick(last: float, bid: float, ask: float, flags: int) -> bool:
"""
Classify tick aggressiveness.
TICK_FLAG_BUY = 32 -> BUY (True)
TICK_FLAG_SELL = 64 -> SELL (False)
TICK_FLAG_ASK = 4 -> ASK (True)
TICK_FLAG_BID = 2 -> BID (False)
"""
if flags & 32:
return True
elif flags & 64:
return False
if last > 0:
if last >= ask:
return True
elif last <= bid:
return False
# Forex quote ticks fallback
if flags & 4:
return True
elif flags & 2:
return False
# Standard fallback if between spread: closer to ask is BUY
if ask > bid and last > 0:
return (last - bid) >= (ask - last)
return True
+175
View File
@@ -0,0 +1,175 @@
import asyncio
import logging
import time
import sys
import os
from typing import Callable, Optional, Any
sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
import MetaTrader5 as mt5
from config import settings
from backend.aggregator import Aggregator, classify_tick
logging.basicConfig(level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s")
logger = logging.getLogger("mt5_collector")
class MT5Collector:
def __init__(self, aggregator: Aggregator, on_update_callback: Callable[[dict, Optional[dict]], Any]):
self.aggregator = aggregator
self.on_update_callback = on_update_callback
self.symbol = settings.MT5_SYMBOL
self.running = False
self.connected = False
self.last_tick_time_msc = 0
self.seen_ticks_buffer = set()
async def connect_mt5(self) -> bool:
"""
Attempts to initialize and login to the MetaTrader 5 terminal.
All MT5 calls are blocking, so they run in a thread executor.
"""
try:
mt5_path = r"C:\Program Files\MetaTrader 5\terminal64.exe"
initialized = await asyncio.to_thread(mt5.initialize, path=mt5_path)
if not initialized:
err = await asyncio.to_thread(mt5.last_error)
logger.error(f"MT5 initialize failed: {err}")
return False
if settings.MT5_LOGIN > 0:
login_success = await asyncio.to_thread(
mt5.login,
settings.MT5_LOGIN,
password=settings.MT5_PASSWORD,
server=settings.MT5_SERVER,
)
if not login_success:
err = await asyncio.to_thread(mt5.last_error)
logger.error(f"MT5 login failed: {err}")
await asyncio.to_thread(mt5.shutdown)
return False
symbol_info = await asyncio.to_thread(mt5.symbol_info, self.symbol)
if symbol_info is None:
logger.error(f"Symbol {self.symbol} not found.")
await asyncio.to_thread(mt5.shutdown)
return False
if not symbol_info.visible:
selected = await asyncio.to_thread(mt5.symbol_select, self.symbol, True)
if not selected:
logger.error(f"Failed to select/make visible symbol {self.symbol}.")
await asyncio.to_thread(mt5.shutdown)
return False
tick_size = symbol_info.trade_tick_size
if tick_size > 0:
self.aggregator.tick_size = tick_size
self.aggregator.active_cluster.tick_size = tick_size
logger.info(f"Set aggregator tick size to {tick_size}")
logger.info("Successfully connected to MetaTrader 5 and logged in.")
self.connected = True
return True
except Exception as e:
logger.error(f"Exception during MT5 connection: {e}")
return False
async def disconnect_mt5(self):
try:
await asyncio.to_thread(mt5.shutdown)
except Exception as e:
logger.error(f"Error during MT5 shutdown: {e}")
self.connected = False
async def start(self):
self.running = True
backoff = 1.0
while self.running:
if not self.connected:
success = await self.connect_mt5()
if not success:
logger.info(f"Reconnecting to MT5 in {backoff:.1f}s...")
await asyncio.sleep(backoff)
backoff = min(backoff * 2, 60.0)
continue
else:
backoff = 1.0
# Fetch from 6 hours ago so the chart isn't empty when started
from datetime import datetime, timedelta
start_time_dt = datetime.now() - timedelta(hours=6)
ticks = await asyncio.to_thread(
mt5.copy_ticks_from, self.symbol, start_time_dt, 100000, mt5.COPY_TICKS_ALL
)
if ticks is not None and len(ticks) > 0:
self.last_tick_time_msc = ticks[0]['time_msc']
else:
self.last_tick_time_msc = int(start_time_dt.timestamp() * 1000)
# Polling loop
try:
from datetime import datetime
polling_dt = datetime.fromtimestamp(self.last_tick_time_msc / 1000.0)
ticks = await asyncio.to_thread(
mt5.copy_ticks_from,
self.symbol,
polling_dt,
1000,
mt5.COPY_TICKS_ALL,
)
if ticks is None:
err = await asyncio.to_thread(mt5.last_error)
logger.error(f"MT5 copy_ticks_from returned None: {err}")
self.connected = False
await self.disconnect_mt5()
continue
if len(ticks) > 0:
logger.info(f"Fetched {len(ticks)} ticks starting at {ticks[0]['time_msc']}")
for tick in ticks:
msc = tick['time_msc']
if msc < self.last_tick_time_msc:
continue
tick_id = (msc, tick['bid'], tick['ask'], tick['last'], tick['volume_real'], tick['flags'])
if msc == self.last_tick_time_msc and tick_id in self.seen_ticks_buffer:
continue
if msc > self.last_tick_time_msc:
self.seen_ticks_buffer.clear()
self.last_tick_time_msc = msc
self.seen_ticks_buffer.add(tick_id)
price = tick['last'] if tick['last'] > 0 else (tick['bid'] if tick['bid'] > 0 else tick['ask'])
volume = tick['volume_real'] if tick['volume_real'] > 0 else float(tick['volume'])
flags = tick['flags']
# Forex ticks often have volume=0 (they are just quote updates).
# We count each quote update as 1 unit of tick volume to build the footprint.
if volume == 0:
volume = 1.0
is_buy = classify_tick(price, tick['bid'], tick['ask'], flags)
active_json, closed_json = self.aggregator.process_tick(price, volume, is_buy, msc)
self.on_update_callback(active_json, closed_json)
await asyncio.sleep(0.1)
except Exception as e:
logger.error(f"Error during tick polling loop: {e}")
self.connected = False
await self.disconnect_mt5()
await asyncio.sleep(2.0)
async def stop(self):
self.running = False
await self.disconnect_mt5()
logger.info("MT5 Collector stopped.")
+116
View File
@@ -0,0 +1,116 @@
import asyncio
import logging
import sys
import os
from contextlib import asynccontextmanager
from fastapi import FastAPI, WebSocket, WebSocketDisconnect
from fastapi.middleware.cors import CORSMiddleware
from typing import Set, Dict, Any, Optional
# Add project root to sys.path
sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
from config import settings
from backend.aggregator import Aggregator
from backend.mt5_collector import MT5Collector
logging.basicConfig(level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s")
logger = logging.getLogger("server")
# Shared state
aggregator = Aggregator(tick_size=1.0)
active_connections: Set[WebSocket] = set()
collector_task: Optional[asyncio.Task] = None
def broadcast_update(active_json: dict, closed_json: Optional[dict]):
"""
Callback executed by MT5Collector when a new tick is processed.
Schedules WebSocket sends on the event loop.
"""
message = {
"type": "tick",
"active": active_json,
"closed": closed_json
}
for connection in list(active_connections):
try:
asyncio.create_task(connection.send_json(message))
except Exception as e:
logger.error(f"Error sending update to client: {e}")
# Initialize MT5 Collector
collector = MT5Collector(aggregator, on_update_callback=broadcast_update)
async def _start_collector_delayed():
"""Wait a moment for the server to fully start, then begin polling MT5."""
await asyncio.sleep(1)
logger.info("Starting MT5 collector background task...")
await collector.start()
@asynccontextmanager
async def lifespan(app: FastAPI):
global collector_task
# Fire-and-forget: start collector AFTER yielding so uvicorn is ready
collector_task = asyncio.create_task(_start_collector_delayed())
logger.info("Server is starting up...")
yield
# Shutdown
logger.info("Stopping MT5 collector task...")
collector.running = False
collector_task.cancel()
try:
await collector_task
except asyncio.CancelledError:
pass
await collector.disconnect_mt5()
app = FastAPI(title="YuClusters Local Server", lifespan=lifespan)
# Add CORS Middleware so local frontend can query history
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
@app.get("/history")
async def get_history():
"""Returns the buffer of historical closed clusters."""
return aggregator.history
@app.websocket("/ws")
async def websocket_endpoint(websocket: WebSocket):
await websocket.accept()
active_connections.add(websocket)
logger.info(f"Client connected. Active connections: {len(active_connections)}")
# Send the current active cluster state on connection
try:
await websocket.send_json({
"type": "init",
"active": aggregator.active_cluster.to_json()
})
except Exception as e:
logger.error(f"Error sending init state: {e}")
try:
while True:
await websocket.receive_text()
except WebSocketDisconnect:
active_connections.discard(websocket)
logger.info(f"Client disconnected. Active connections: {len(active_connections)}")
except Exception as e:
logger.error(f"WebSocket error: {e}")
active_connections.discard(websocket)
if __name__ == "__main__":
import uvicorn
logger.info(f"Starting YuClusters server on port {settings.WS_PORT}...")
uvicorn.run(
"backend.server:app",
host="0.0.0.0",
port=settings.WS_PORT,
log_level="info",
)
+22
View File
@@ -0,0 +1,22 @@
import os
# MT5
MT5_LOGIN = int(os.environ.get("MT5_LOGIN", 0)) # seu número de conta
MT5_PASSWORD = os.environ.get("MT5_PASSWORD", "") # sua senha
MT5_SERVER = os.environ.get("MT5_SERVER", "") # nome do servidor (ex: "XPInvestimentos-Real")
MT5_SYMBOL = os.environ.get("MT5_SYMBOL", "EURUSD") # símbolo padrão
# Cluster
CLUSTER_RANGE_POINTS = int(os.environ.get("CLUSTER_RANGE_POINTS", 10))
CLUSTER_VOLUME_MAX = float(os.environ.get("CLUSTER_VOLUME_MAX", 1000))
CLUSTER_DELTA_MAX = float(os.environ.get("CLUSTER_DELTA_MAX", 500))
CLUSTER_TIME_SECONDS = float(os.environ.get("CLUSTER_TIME_SECONDS", 60))
CLUSTER_CLOSE_MODE = os.environ.get("CLUSTER_CLOSE_MODE", "range") # "range" | "volume" | "delta" | "time"
# Imbalance
IMBALANCE_RATIO = float(os.environ.get("IMBALANCE_RATIO", 3.0))
STACKED_MIN_COUNT = int(os.environ.get("STACKED_MIN_COUNT", 3))
# WebSocket
WS_PORT = int(os.environ.get("WS_PORT", 6002))
HISTORY_BUFFER_SIZE = int(os.environ.get("HISTORY_BUFFER_SIZE", 50))
+21
View File
@@ -0,0 +1,21 @@
version: '3.8'
services:
mt5-gateway:
build:
context: ./docker/mt5-gateway
dockerfile: Dockerfile
container_name: mt5-gateway-service
ports:
- "5000:5000"
environment:
# These variables can be injected via a .env file securely in production
- MT5_SERVER=
- MT5_LOGIN=0
- MT5_PASSWORD=
deploy:
resources:
limits:
cpus: '1.0'
memory: 1G
restart: unless-stopped
+40
View File
@@ -0,0 +1,40 @@
FROM ubuntu:22.04
# Prevent interactive prompts
ENV DEBIAN_FRONTEND=noninteractive
# Add 32-bit architecture for Wine and install dependencies
RUN dpkg --add-architecture i386 && \
apt-get update && \
apt-get install -y --no-install-recommends \
wine64 \
wine32 \
xvfb \
wget \
cabextract \
winbind \
curl \
&& apt-get clean \
&& rm -rf /var/lib/apt/lists/*
# Set up user for Wine
RUN useradd -m -s /bin/bash mt5user
USER mt5user
WORKDIR /home/mt5user
# Copy scripts
COPY --chown=mt5user:mt5user scripts/ /app/scripts/
RUN chmod +x /app/scripts/*.sh
# Run provisioning steps
RUN /app/scripts/05_install_python.sh
RUN /app/scripts/06_install_libraries.sh
RUN /app/scripts/06b_install_mt5.sh
# Copy application source
COPY --chown=mt5user:mt5user app.py /app/app.py
EXPOSE 5000
# Start script
ENTRYPOINT ["/app/scripts/07_start_wine_flask.sh"]
+102
View File
@@ -0,0 +1,102 @@
import os
from flask import Flask, jsonify, request
import MetaTrader5 as mt5
app = Flask(__name__)
# Basic settings from environment or defaults
MT5_PATH = os.getenv("MT5_PATH", "C:\\Program Files\\MetaTrader 5\\terminal64.exe")
MT5_SERVER = os.getenv("MT5_SERVER", "")
MT5_LOGIN = int(os.getenv("MT5_LOGIN", "0"))
MT5_PASSWORD = os.getenv("MT5_PASSWORD", "")
def init_mt5():
# If login is provided, connect with credentials
if MT5_LOGIN != 0 and MT5_PASSWORD:
if not mt5.initialize(path=MT5_PATH, login=MT5_LOGIN, server=MT5_SERVER, password=MT5_PASSWORD):
return False, mt5.last_error()
else:
# Just initialize whatever is there
if not mt5.initialize(path=MT5_PATH):
return False, mt5.last_error()
return True, None
@app.route('/health', methods=['GET'])
def health_check():
success, error = init_mt5()
if not success:
return jsonify({"status": "error", "message": "Failed to connect to MT5", "error_code": error}), 500
info = mt5.terminal_info()
if info is None:
return jsonify({"status": "error", "message": "Failed to get terminal info"}), 500
return jsonify({
"status": "ok",
"terminal_connected": info.connected,
"trade_allowed": info.trade_allowed,
"build": info.build
})
@app.route('/symbol/<ticker>', methods=['GET'])
def symbol_info(ticker):
init_mt5()
info = mt5.symbol_info(ticker)
if info is None:
return jsonify({"status": "error", "message": f"Symbol {ticker} not found"}), 404
return jsonify({
"symbol": info.name,
"bid": info.bid,
"ask": info.ask,
"spread": info.spread,
"trade_mode": info.trade_mode
})
@app.route('/order', methods=['POST'])
def place_order():
init_mt5()
data = request.json
# Very basic order payload (can be extended with full Swagger spec later)
# Expects: {"symbol": "EURUSD", "action": "buy", "volume": 1.0}
symbol = data.get("symbol")
action = data.get("action")
volume = float(data.get("volume", 0.0))
if action == "buy":
type = mt5.ORDER_TYPE_BUY
price = mt5.symbol_info_tick(symbol).ask
else:
type = mt5.ORDER_TYPE_SELL
price = mt5.symbol_info_tick(symbol).bid
order_request = {
"action": mt5.TRADE_ACTION_DEAL,
"symbol": symbol,
"volume": volume,
"type": type,
"price": price,
"deviation": 20,
"magic": 234000,
"comment": "python api",
"type_time": mt5.ORDER_TIME_GTC,
"type_filling": mt5.ORDER_FILLING_IOC,
}
result = mt5.order_send(order_request)
if result is None:
return jsonify({"status": "error", "message": "Order failed entirely", "error": mt5.last_error()}), 500
# Translate MT5 Return Codes to friendly API responses
# Mapping can be expanded as needed
if result.retcode == mt5.TRADE_RETCODE_DONE:
return jsonify({"status": "ok", "retcode": result.retcode, "deal": result.deal, "message": "Order placed successfully"})
elif result.retcode == mt5.TRADE_RETCODE_MARKET_CLOSED:
return jsonify({"status": "error", "retcode": result.retcode, "message": "Market is closed"}), 400
else:
return jsonify({"status": "error", "retcode": result.retcode, "message": "Order failed", "comment": result.comment}), 400
if __name__ == '__main__':
app.run(host='0.0.0.0', port=5000)
@@ -0,0 +1,10 @@
#!/bin/bash
set -e
echo "=== Installing Python for Windows via Wine ==="
# We need to install the Windows version of Python inside Wine for MT5 library compatibility
WINEPREFIX=$HOME/.wine wine msiexec /i https://www.python.org/ftp/python/3.10.11/python-3.10.11-amd64.msi /quiet InstallAllUsers=1 PrependPath=1 Include_test=0
# Verify python installation
WINEPREFIX=$HOME/.wine wine python --version
echo "Python installed successfully."
@@ -0,0 +1,12 @@
#!/bin/bash
set -e
echo "=== Installing Python Libraries via Wine ==="
# Ensure pip is up to date
WINEPREFIX=$HOME/.wine wine python -m pip install --upgrade pip
# Install Flask and MetaTrader5
# The user explicitly warned to be careful with the case sensitivity of MetaTrader5!
WINEPREFIX=$HOME/.wine wine python -m pip install Flask MetaTrader5
echo "Libraries installed successfully."
@@ -0,0 +1,27 @@
#!/bin/bash
set -e
echo "=== Downloading and Installing MetaTrader 5 ==="
export WINEPREFIX=$HOME/.wine
export WINEDLLOVERRIDES="mscoree,mshtml="
# Start Xvfb temporarily for the installation
# Some silent Windows installers still crash if there's no display available
Xvfb :99 -screen 0 1024x768x16 &
XVFB_PID=$!
export DISPLAY=:99
sleep 2
# Download MT5 setup from MetaQuotes official CDN
wget -O mt5setup.exe "https://download.mql5.com/cdn/web/metaquotes.software.corp/mt5/mt5setup.exe"
# Install silently ( /auto )
echo "Installing MT5 silently (this may take a minute)..."
wine mt5setup.exe /auto
# Wait for background installation tasks to complete and shutdown Wine safely
wineserver -w
kill $XVFB_PID || true
rm mt5setup.exe
echo "MT5 installed successfully."
@@ -0,0 +1,18 @@
#!/bin/bash
set -e
echo "=== Starting MT5 Gateway ==="
export WINEPREFIX=$HOME/.wine
export WINEDLLOVERRIDES="mscoree,mshtml="
# Start Xvfb in background
Xvfb :0 -screen 0 1024x768x16 &
export DISPLAY=:0
# Wait for X11
sleep 2
# We start the Flask server via Python in Wine
# The MT5 logic inside app.py will initialize MT5
echo "Starting Flask API Bridge..."
wine python /app/app.py
+17
View File
@@ -0,0 +1,17 @@
<!doctype html>
<html lang="en" class="h-full bg-darkBg">
<head>
<meta charset="UTF-8" />
<link rel="icon" type="image/svg+xml" href="data:image/svg+xml,<svg xmlns=%22http://www.w3.org/2000/svg%22 viewBox=%220 0 100 100%22><text y=%22.9em%22 font-size=%2290%22>📊</text></svg>" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>YuClusters Local — Pro Footprint Chart</title>
<!-- Outfit & Inter Fonts -->
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700&family=Outfit:wght@400;600;800&family=JetBrains+Mono:wght@400;700&display=swap" rel="stylesheet">
</head>
<body class="h-full text-slate-100 font-sans selection:bg-neonGreen/20 selection:text-neonGreen antialiased overflow-hidden">
<div id="root" class="h-full"></div>
<script type="module" src="/src/main.jsx"></script>
</body>
</html>
+2665
View File
File diff suppressed because it is too large Load Diff
+24
View File
@@ -0,0 +1,24 @@
{
"name": "yuclusters-frontend",
"private": true,
"version": "0.1.0",
"type": "module",
"scripts": {
"dev": "vite",
"build": "vite build",
"preview": "vite preview"
},
"dependencies": {
"react": "^18.2.0",
"react-dom": "^18.2.0"
},
"devDependencies": {
"@types/react": "^18.2.66",
"@types/react-dom": "^18.2.22",
"@vitejs/plugin-react": "^4.2.1",
"autoprefixer": "^10.4.18",
"postcss": "^8.4.35",
"tailwindcss": "^3.4.1",
"vite": "^5.2.0"
}
}
+6
View File
@@ -0,0 +1,6 @@
export default {
plugins: {
tailwindcss: {},
autoprefixer: {},
},
}
+275
View File
@@ -0,0 +1,275 @@
import React, { useState, useEffect, useCallback } from 'react';
import { useWebSocket } from './useWebSocket';
import FootprintCanvas from './FootprintCanvas';
import AlertEngine from './utils/AlertEngine';
const BACKEND_PORT = 6002;
const WS_URL = `ws://localhost:${BACKEND_PORT}/ws`;
const API_URL = `http://localhost:${BACKEND_PORT}`;
export default function App() {
const [history, setHistory] = useState([]);
const [activeCluster, setActiveCluster] = useState(null);
const [lastTickTime, setLastTickTime] = useState(null);
// Phase 2 & 4 Settings
const [stepMultiplier, setStepMultiplier] = useState(1);
const [viewMode, setViewMode] = useState('bidask'); // 'bidask' or 'delta'
const [imbalanceRatio, setImbalanceRatio] = useState(300); // percentage
// Toasts
const [toasts, setToasts] = useState([]);
const pushToast = useCallback((msg, type = 'info') => {
const id = Date.now() + Math.random();
setToasts(prev => [...prev, { id, msg, type }]);
setTimeout(() => {
setToasts(prev => prev.filter(t => t.id !== id));
}, 5000);
}, []);
// Fetch initial history when backend becomes reachable
const fetchHistory = useCallback(async () => {
try {
const res = await fetch(`${API_URL}/history`);
if (res.ok) {
const data = await res.json();
setHistory(data);
}
} catch (e) {
console.warn("Could not fetch cluster history:", e);
}
}, []);
// Message Handler for WebSocket
const handleWebSocketMessage = useCallback((msg) => {
setLastTickTime(new Date());
if (msg.type === 'init') {
setActiveCluster(msg.active);
fetchHistory(); // Sync history on connection
} else if (msg.type === 'tick') {
setActiveCluster(msg.active);
// If a cluster has just closed, we receive the closed state
if (msg.closed) {
setHistory(prev => {
const updated = [...prev, msg.closed];
if (updated.length > 50) {
updated.shift();
}
return updated;
});
// Dispatch alerts for the newly closed cluster
AlertEngine.processClusters([msg.closed], pushToast);
}
}
}, [fetchHistory]);
const wsStatus = useWebSocket(WS_URL, handleWebSocketMessage);
useEffect(() => {
fetchHistory();
}, [fetchHistory, wsStatus]);
// Aggregate stats from history
const totalVolume = history.reduce((acc, c) => acc + (c.total_volume || 0), 0) + (activeCluster?.total_volume || 0);
const avgVolumePerCluster = history.length > 0 ? (history.reduce((acc, c) => acc + (c.total_volume || 0), 0) / history.length).toFixed(0) : 0;
const allClusters = [...history];
if (activeCluster) allClusters.push(activeCluster);
return (
<div className="flex flex-col h-full bg-[#0B0E14] text-slate-100 relative">
{/* Toast Container */}
<div className="absolute top-4 right-4 z-50 flex flex-col gap-2">
{toasts.map(t => (
<div key={t.id} className={`px-4 py-3 rounded-md shadow-lg font-medium text-sm flex items-center gap-2 border ${
t.type === 'error' ? 'bg-red-500/10 border-red-500 text-red-500' :
t.type === 'warning' ? 'bg-amber-500/10 border-amber-500 text-amber-400' :
'bg-blue-500/10 border-blue-500 text-blue-400'
}`}>
<span>{t.type === 'error' ? '🚨' : t.type === 'warning' ? '⚠️' : '️'}</span>
{t.msg}
</div>
))}
</div>
{/* Premium Header */}
<header className="flex items-center justify-between px-6 py-4 bg-[#151B26] border-b border-slate-800 shadow-md">
<div className="flex items-center gap-3">
<div className="w-9 h-9 bg-gradient-to-tr from-[#00E676] to-[#00B0FF] rounded-lg flex items-center justify-center font-bold text-lg text-white shadow-lg">
Yu
</div>
<div>
<h1 className="text-lg font-bold tracking-tight bg-gradient-to-r from-white to-slate-400 bg-clip-text text-transparent">
YuClusters Local
</h1>
<p className="text-xs text-slate-400">Order Flow Footprint Analyzer 2.0</p>
</div>
</div>
{/* Connection Status indicator */}
<div className="flex items-center gap-6 text-sm">
{lastTickTime && (
<div className="text-slate-400 text-xs hidden sm:block">
Last Update: <span className="font-mono text-slate-300">{lastTickTime.toLocaleTimeString()}</span>
</div>
)}
<div className="flex items-center gap-2">
<span className="text-xs text-slate-400 font-medium">MT5 Bridge:</span>
<div className={`flex items-center gap-1.5 px-3 py-1 rounded-full text-xs font-semibold ${
wsStatus === 'connected'
? 'bg-[#00E676]/10 text-[#00E676] border border-[#00E676]/20'
: wsStatus === 'connecting'
? 'bg-amber-500/10 text-amber-500 border border-amber-500/20'
: 'bg-red-500/10 text-red-500 border border-red-500/20'
}`}>
<span className={`w-1.5 h-1.5 rounded-full ${
wsStatus === 'connected' ? 'bg-[#00E676] animate-pulse' : wsStatus === 'connecting' ? 'bg-amber-500 animate-pulse' : 'bg-red-500'
}`} />
{wsStatus.toUpperCase()}
</div>
</div>
</div>
</header>
{/* Main Body Layout */}
<main className="flex-1 flex overflow-hidden p-6 gap-6">
{/* Footprint Chart Panel */}
<div className="flex-1 flex flex-col h-full bg-[#151B26]/30 rounded-xl overflow-hidden">
<FootprintCanvas
clusters={allClusters}
tickSize={0.00001}
stepMultiplier={stepMultiplier}
viewMode={viewMode}
imbalanceRatio={imbalanceRatio}
/>
</div>
{/* Info Sidebar Panel */}
<aside className="w-80 flex flex-col gap-6 hidden lg:flex">
{/* Active Statistics Card */}
<section className="bg-[#151B26] border border-slate-800 rounded-xl p-5 shadow-lg">
<h2 className="text-sm font-semibold text-slate-300 mb-4 border-b border-slate-800 pb-2">
System Overview
</h2>
<div className="space-y-4">
<div className="flex justify-between items-center text-xs">
<span className="text-slate-400">Target Symbol:</span>
<span className="font-mono font-semibold text-[#00B0FF] bg-[#00B0FF]/10 px-2 py-0.5 rounded">
EURUSD
</span>
</div>
<div className="flex flex-col sm:flex-row items-center gap-6">
<div className="flex gap-8">
<div className="flex flex-col">
<span className="text-xs text-slate-500 font-medium">TOTAL VOLUME</span>
<span className="text-lg font-bold text-slate-200">{(totalVolume / 1000).toFixed(1)}k</span>
</div>
<div className="flex flex-col">
<span className="text-xs text-slate-500 font-medium">AVG VOL / CLUSTER</span>
<span className="text-lg font-bold text-slate-200">{avgVolumePerCluster}</span>
</div>
</div>
<div className="h-8 w-px bg-slate-800 hidden sm:block"></div>
<div className="flex items-center gap-4">
<div className="flex flex-col">
<span className="text-[10px] text-slate-500 font-bold mb-1">PRICE STEP</span>
<div className="flex items-center bg-[#151B26] border border-slate-700 rounded-md overflow-hidden">
<button onClick={() => setStepMultiplier(Math.max(1, stepMultiplier - 1))} className="px-2 py-1 text-slate-400 hover:text-white hover:bg-slate-800">-</button>
<div className="px-3 py-1 text-sm font-bold text-white min-w-[30px] text-center">{stepMultiplier}</div>
<button onClick={() => setStepMultiplier(stepMultiplier + 1)} className="px-2 py-1 text-slate-400 hover:text-white hover:bg-slate-800">+</button>
</div>
</div>
<div className="flex flex-col">
<span className="text-[10px] text-slate-500 font-bold mb-1">VIEW MODE</span>
<div className="flex bg-[#151B26] border border-slate-700 rounded-md overflow-hidden">
<button
onClick={() => setViewMode('bidask')}
className={`px-3 py-1 text-xs font-semibold ${viewMode === 'bidask' ? 'bg-[#00E676] text-slate-900' : 'text-slate-400 hover:bg-slate-800'}`}
>
Bid x Ask
</button>
<button
onClick={() => setViewMode('delta')}
className={`px-3 py-1 text-xs font-semibold ${viewMode === 'delta' ? 'bg-[#00E676] text-slate-900' : 'text-slate-400 hover:bg-slate-800'}`}
>
Delta
</button>
</div>
</div>
</div>
<div className="flex items-center gap-4 mt-4">
<div className="flex flex-col flex-1">
<div className="flex justify-between mb-1">
<span className="text-[10px] text-slate-500 font-bold">IMBALANCE RATIO</span>
<span className="text-[10px] font-mono text-slate-300">{imbalanceRatio}%</span>
</div>
<input
type="range"
min="150"
max="500"
step="10"
value={imbalanceRatio}
onChange={(e) => setImbalanceRatio(parseInt(e.target.value))}
className="w-full accent-[#00B0FF] bg-slate-800 rounded-lg h-1.5 appearance-none cursor-pointer"
/>
</div>
</div>
</div>
<div className="flex justify-between items-center text-xs mt-4 border-t border-slate-800 pt-4">
<span className="text-slate-400">Closed Clusters:</span>
<span className="font-mono text-slate-200">{history.length}</span>
</div>
</div>
</section>
{/* Aggregator Settings Rules Indicator */}
<section className="bg-[#151B26] border border-slate-800 rounded-xl p-5 shadow-lg flex-1">
<h2 className="text-sm font-semibold text-slate-300 mb-4 border-b border-slate-800 pb-2">
Aggregator Rules
</h2>
<div className="text-xs space-y-3.5 text-slate-400">
<div>
<p className="text-slate-300 font-medium mb-1">Diagonal Imbalance</p>
<code className="block bg-[#0B0E14] p-2 rounded text-[10px] text-slate-400 leading-relaxed font-mono">
BUY: ask[i] &ge; 3.0 &times; bid[i-1]<br />
SELL: bid[i] &ge; 3.0 &times; ask[i+1]
</code>
</div>
<div>
<p className="text-slate-300 font-medium mb-1">Stacked Imbalance Zone</p>
<p className="leading-relaxed">
Triggered on <span className="text-white font-semibold">3+</span> consecutive diagonal imbalances in the same direction. Highlighting horizontal zones.
</p>
</div>
<div>
<p className="text-slate-300 font-medium mb-1">POC (Point of Control)</p>
<p className="leading-relaxed">
Level containing the highest total volume within the cluster. Highlighted with a <span className="text-[#FFD600] font-semibold">Gold border</span>.
</p>
</div>
</div>
</section>
{/* Quick Guide Footer */}
<footer className="text-[11px] text-slate-500 text-center leading-relaxed">
Drag to pan horizontally & vertically.<br />
Use scroll wheel to move vertical scale.<br />
Hold Shift + scroll wheel to scroll horizontal.
</footer>
</aside>
</main>
</div>
);
}
+699
View File
@@ -0,0 +1,699 @@
import React, { useRef, useEffect, useState } from 'react';
export default function FootprintCanvas({ clusters, tickSize = 1.0, stepMultiplier = 1, viewMode = 'bidask', imbalanceRatio = 300 }) {
const canvasRef = useRef(null);
// Navigation & Scale State
const [scrollOffset, setScrollOffset] = useState({ x: 50, y: 0 }); // X: horizontal offset, Y: vertical offset
const [zoom, setZoom] = useState(1); // Zoom level
const [isDragging, setIsDragging] = useState(false);
const dragStart = useRef({ x: 0, y: 0 });
const dragOffsetStart = useRef({ x: 0, y: 0 });
// Apply zoom to sizes
const colWidth = 140 * zoom; // width of each cluster column
const colGap = 15 * zoom; // gap between columns
const rowHeight = 26 * zoom; // height of each price cell
const axisWidth = 70; // width of the vertical price axis on the right
// Handle auto-scroll to the right (most recent cluster) on new cluster load
const lastClusterCount = useRef(0);
useEffect(() => {
if (clusters && clusters.length > lastClusterCount.current && canvasRef.current) {
const canvas = canvasRef.current;
// Scroll to show the active cluster at the right side
const rightmostX = canvas.width - axisWidth - (clusters.length * (colWidth + colGap)) - 50;
setScrollOffset(prev => ({ ...prev, x: Math.min(160, rightmostX) }));
lastClusterCount.current = clusters.length;
}
}, [clusters?.length]);
// Main Render Loop
useEffect(() => {
const canvas = canvasRef.current;
if (!canvas) return;
const ctx = canvas.getContext('2d');
// Handle High DPI displays
const dpr = window.devicePixelRatio || 1;
const rect = canvas.getBoundingClientRect();
canvas.width = rect.width * dpr;
canvas.height = rect.height * dpr;
ctx.scale(dpr, dpr);
const width = rect.width;
const height = rect.height;
// Split View Layout
const bottomPanelHeight = 100;
const chartHeight = height - bottomPanelHeight;
// Clear screen
ctx.fillStyle = '#0B0E14';
ctx.fillRect(0, 0, width, height);
// Draw Grid Background
ctx.strokeStyle = '#151B26';
ctx.lineWidth = 1;
for (let x = 0; x < width; x += 50) {
ctx.beginPath();
ctx.moveTo(x, 0);
ctx.lineTo(x, height);
ctx.stroke();
}
for (let y = 0; y < height; y += 50) {
ctx.beginPath();
ctx.moveTo(0, y);
ctx.lineTo(width, y);
ctx.stroke();
}
if (!clusters || clusters.length === 0) {
ctx.fillStyle = '#64748B';
ctx.font = '16px Outfit, sans-serif';
ctx.textAlign = 'center';
ctx.textBaseline = 'middle';
ctx.fillText('Waiting for market data from MetaTrader 5...', width / 2, height / 2);
return;
}
// Determine the baseline price to align Y coordinates
const latestCluster = clusters[clusters.length - 1];
const basePrice = latestCluster.poc || 0;
const actualTickSize = (latestCluster.tick_size || tickSize) * stepMultiplier;
const centerY = chartHeight / 2 + scrollOffset.y;
// Helper to get Y coordinate for a given price
const getPriceY = (price) => {
const diffTicks = (price - basePrice) / actualTickSize;
return centerY - (diffTicks * rowHeight);
};
// Helper to get price from Y coordinate (for axis rendering)
const getYPrice = (y) => {
const diffTicks = (centerY - y) / rowHeight;
return basePrice + (diffTicks * actualTickSize);
};
// Cumulative Delta Tracking
let cumulativeDelta = 0;
// Draw Columns (Clusters)
clusters.forEach((cluster, index) => {
// Calculate column X position
// Offset starting after the Volume Profile panel (width 140)
const colX = scrollOffset.x + index * (colWidth + colGap);
// Don't render if outside canvas bounds (horizontal clipping)
if (colX + colWidth < 0 || colX > width - axisWidth) return;
let levels = cluster.levels || {};
// Dynamic Binning based on stepMultiplier
if (stepMultiplier > 1) {
const binnedLevels = {};
Object.keys(levels).forEach(pStr => {
const p = parseFloat(pStr);
const data = levels[pStr];
const binPrice = Math.round(p / actualTickSize) * actualTickSize;
if (!binnedLevels[binPrice]) {
binnedLevels[binPrice] = { ask: 0, bid: 0, total: 0, delta: 0, imbalance: null };
}
binnedLevels[binPrice].ask += data.ask || 0;
binnedLevels[binPrice].bid += data.bid || 0;
binnedLevels[binPrice].total += data.total || 0;
binnedLevels[binPrice].delta += data.delta || 0;
const ratio = imbalanceRatio / 100.0;
if (binnedLevels[binPrice].ask >= binnedLevels[binPrice].bid * ratio && binnedLevels[binPrice].ask > 0) {
binnedLevels[binPrice].imbalance = 'buy';
} else if (binnedLevels[binPrice].bid >= binnedLevels[binPrice].ask * ratio && binnedLevels[binPrice].bid > 0) {
binnedLevels[binPrice].imbalance = 'sell';
}
});
levels = binnedLevels;
}
// Sort string keys numerically, but keep them as strings to avoid trailing zero lookup issues
const pricesStr = Object.keys(levels).sort((a, b) => Number(b) - Number(a));
if (pricesStr.length === 0) return;
const numericPrices = pricesStr.map(Number);
const highestPrice = Math.max(...numericPrices);
const lowestPrice = Math.min(...numericPrices);
// Draw Stacked Imbalance background zone if present
if (cluster.stacked && (cluster.stacked.buy || cluster.stacked.sell)) {
const stackedPrices = cluster.stacked.price_range || [];
if (stackedPrices.length > 0) {
const sHigh = Math.max(...stackedPrices);
const sLow = Math.min(...stackedPrices);
const yTop = getPriceY(sHigh) - rowHeight / 2;
const yBottom = getPriceY(sLow) + rowHeight / 2;
const grad = ctx.createLinearGradient(colX - 8, yTop, colX, yTop);
if (cluster.stacked.buy) {
grad.addColorStop(0, 'rgba(0, 230, 118, 0.4)');
grad.addColorStop(1, 'rgba(0, 230, 118, 0.05)');
ctx.fillStyle = grad;
} else {
grad.addColorStop(0, 'rgba(255, 23, 68, 0.4)');
grad.addColorStop(1, 'rgba(255, 23, 68, 0.05)');
ctx.fillStyle = grad;
}
ctx.fillRect(colX - 10, yTop, 10, yBottom - yTop);
// Draw thin outline
ctx.strokeStyle = cluster.stacked.buy ? '#00E676' : '#FF1744';
ctx.lineWidth = 1;
ctx.beginPath();
ctx.moveTo(colX - 10, yTop);
ctx.lineTo(colX - 10, yBottom);
ctx.stroke();
}
}
// Draw Cluster Header (Info Card at the top)
const headerY = getPriceY(highestPrice) - rowHeight - 35;
// Header Background
ctx.fillStyle = 'rgba(21, 27, 38, 0.85)';
ctx.strokeStyle = cluster.status === 'active' ? 'rgba(0, 230, 118, 0.4)' : '#2A364F';
ctx.lineWidth = 1.5;
ctx.beginPath();
ctx.roundRect(colX, headerY, colWidth, 45, 6);
ctx.fill();
ctx.stroke();
// Header Text
ctx.fillStyle = '#94A3B8';
ctx.font = '10px JetBrains Mono, monospace';
ctx.textAlign = 'left';
const pattern = cluster.advanced_metrics?.pattern;
const divergence = cluster.advanced_metrics?.delta_divergence;
let patternTag = '';
if (pattern === 'P') patternTag = '[P] ';
if (pattern === 'B') patternTag = '[B] ';
// Divergence Tag
if (divergence) patternTag += '⚠️ ';
// Volume & Delta
const volK = (cluster.total_volume || 0).toFixed(0);
const deltaStr = (cluster.total_delta >= 0 ? '+' : '') + (cluster.total_delta || 0).toFixed(0);
ctx.fillText(`${patternTag}VOL: ${volK}`, colX + 8, headerY + 18);
ctx.fillStyle = cluster.total_delta >= 0 ? '#00E676' : '#FF1744';
ctx.fillText(`DEL: ${deltaStr}`, colX + 8, headerY + 32);
// Time or reason
ctx.fillStyle = '#64748B';
const timeStr = cluster.open_time ? new Date(cluster.open_time).toLocaleTimeString([], { hour: '2-digit', minute: '2-digit', second: '2-digit' }) : '--:--:--';
ctx.fillText(timeStr, colX + colWidth - 55, headerY + 18);
// Find max volume in this cluster to calculate relative opacities
const maxTotalVolumeInCluster = Math.max(...Object.values(levels).map(l => l.total || 1));
// OHLC Candlestick Skeleton (Background)
if (cluster.open_price !== undefined && cluster.close_price !== undefined) {
const isBull = cluster.close_price >= cluster.open_price;
ctx.strokeStyle = isBull ? 'rgba(0, 230, 118, 0.4)' : 'rgba(255, 23, 68, 0.4)';
ctx.fillStyle = isBull ? 'rgba(0, 230, 118, 0.1)' : 'rgba(255, 23, 68, 0.1)';
ctx.lineWidth = 1;
const openY = getPriceY(cluster.open_price);
const closeY = getPriceY(cluster.close_price);
const highY = getPriceY(highestPrice);
const lowY = getPriceY(lowestPrice);
const candleTop = Math.min(openY, closeY) - rowHeight / 2;
const candleBottom = Math.max(openY, closeY) + rowHeight / 2;
const bodyHeight = Math.max(2, candleBottom - candleTop);
// Draw Wick (Pavio)
ctx.beginPath();
ctx.moveTo(colX + colWidth / 2, highY - rowHeight / 2);
ctx.lineTo(colX + colWidth / 2, lowY + rowHeight / 2);
ctx.stroke();
// Draw Body Background
ctx.fillRect(colX - 4, candleTop, colWidth + 8, bodyHeight);
ctx.strokeRect(colX - 4, candleTop, colWidth + 8, bodyHeight);
}
// Draw Cells
pricesStr.forEach((priceStr, i) => {
const price = Number(priceStr);
const cellData = levels[priceStr];
const cellY = getPriceY(price) - rowHeight / 2;
// Skip if vertically out of bounds
if (cellY + rowHeight < 0 || cellY > height) return;
const bid = cellData.bid || 0;
const ask = cellData.ask || 0;
const total = cellData.total || 0;
// Dynamic Imbalance Calculation (Frontend)
const ratio = imbalanceRatio / 100.0;
let dynImbalance = null;
// ask vs lower bid
const lowerData = i + 1 < pricesStr.length ? levels[pricesStr[i + 1]] : null;
const lowerBid = lowerData ? (lowerData.bid || 0) : 0;
const isBuyImbalance = ask >= lowerBid * ratio && ask > 0;
// bid vs higher ask
const upperData = i - 1 >= 0 ? levels[pricesStr[i - 1]] : null;
const upperAsk = upperData ? (upperData.ask || 0) : 0;
const isSellImbalance = bid >= upperAsk * ratio && bid > 0;
if (isBuyImbalance && isSellImbalance) dynImbalance = 'both';
else if (isBuyImbalance) dynImbalance = 'buy';
else if (isSellImbalance) dynImbalance = 'sell';
// Volume-based opacity
const relOpacity = maxTotalVolumeInCluster > 0 ? (total / maxTotalVolumeInCluster) : 0;
// Base fill color with opacity
let cellColor = `rgba(59, 130, 246, ${0.05 + relOpacity * 0.25})`; // Dark Slate Blue default
if (ask >= bid * ratio && ask > 0) {
cellColor = `rgba(0, 230, 118, ${0.2 + relOpacity * 0.5})`; // Strong Green Heatmap (Horizontal)
} else if (bid >= ask * ratio && bid > 0) {
cellColor = `rgba(255, 23, 68, ${0.2 + relOpacity * 0.5})`; // Strong Red Heatmap (Horizontal)
} else if (dynImbalance === 'buy') {
cellColor = `rgba(0, 230, 118, ${0.1 + relOpacity * 0.35})`; // Neon Green tint
} else if (dynImbalance === 'sell') {
cellColor = `rgba(255, 23, 68, ${0.1 + relOpacity * 0.35})`; // Neon Red tint
} else if (dynImbalance === 'both') {
cellColor = `rgba(168, 85, 247, ${0.1 + relOpacity * 0.35})`; // Purple
}
ctx.fillStyle = cellColor;
ctx.fillRect(colX, cellY, colWidth, rowHeight - 2);
// Imbalance border outline
if (dynImbalance === 'buy') {
ctx.strokeStyle = 'rgba(0, 230, 118, 0.8)';
ctx.lineWidth = 1;
ctx.strokeRect(colX + 0.5, cellY + 0.5, colWidth - 1, rowHeight - 3);
} else if (dynImbalance === 'sell') {
ctx.strokeStyle = 'rgba(255, 23, 68, 0.8)';
ctx.lineWidth = 1;
ctx.strokeRect(colX + 0.5, cellY + 0.5, colWidth - 1, rowHeight - 3);
}
// Draw POC outline (Thick Gold Border)
if (price === cluster.poc) {
ctx.strokeStyle = '#FFD600';
ctx.lineWidth = 2;
ctx.strokeRect(colX + 1, cellY + 1, colWidth - 2, rowHeight - 4);
}
// Draw Text (Bid x Ask OR Delta)
const fontSize = Math.max(6, 11 * zoom);
ctx.font = `${fontSize}px JetBrains Mono, monospace`;
ctx.textAlign = 'center';
ctx.textBaseline = 'middle';
if (dynImbalance === 'buy') {
ctx.fillStyle = '#00E676';
} else if (dynImbalance === 'sell') {
ctx.fillStyle = '#FF1744';
} else {
ctx.fillStyle = '#E2E8F0';
}
// Hide text if zoomed out too much to avoid clutter
if (zoom >= 0.5) {
if (viewMode === 'delta') {
const deltaStr = (cellData.delta >= 0 ? '+' : '') + (cellData.delta || 0).toFixed(0);
ctx.fillText(deltaStr, colX + colWidth / 2, cellY + rowHeight / 2);
} else {
ctx.fillText(`${bid.toFixed(0)} × ${ask.toFixed(0)}`, colX + colWidth / 2, cellY + rowHeight / 2);
}
}
});
// Render Extremes Ratios (Informers)
const adv = cluster.advanced_metrics;
if (adv) {
ctx.font = '10px JetBrains Mono, monospace';
ctx.textAlign = 'center';
ctx.textBaseline = 'middle';
const highY = getPriceY(highestPrice) - rowHeight;
const lowY = getPriceY(lowestPrice) + rowHeight;
// Top Ratio
if (adv.high_ratio !== undefined) {
const hRatio = adv.high_ratio;
ctx.fillStyle = hRatio < 0.7 ? '#FFD600' : '#64748B'; // Gold if exhaustion
ctx.fillText(hRatio.toFixed(2), colX + colWidth / 2, highY);
}
// Bottom Ratio
if (adv.low_ratio !== undefined) {
const lRatio = adv.low_ratio;
ctx.fillStyle = lRatio < 0.7 ? '#FFD600' : '#64748B'; // Gold if exhaustion
ctx.fillText(lRatio.toFixed(2), colX + colWidth / 2, lowY);
}
}
// Draw Top/Bottom Extreme Markers
if (adv) {
if (adv.top_extreme === 'exhaustion' || adv.top_extreme === 'absorption') {
const topY = getPriceY(highestPrice) - rowHeight / 2;
ctx.strokeStyle = adv.top_extreme === 'absorption' ? '#FF9800' : '#2196F3'; // Orange for Absorption, Blue for Exhaustion
ctx.lineWidth = adv.top_extreme === 'absorption' ? 3 : 1.5;
ctx.beginPath();
ctx.moveTo(colX, topY);
ctx.lineTo(colX + colWidth, topY);
ctx.stroke();
}
if (adv.bottom_extreme === 'exhaustion' || adv.bottom_extreme === 'absorption') {
const bottomY = getPriceY(lowestPrice) + rowHeight / 2;
ctx.strokeStyle = adv.bottom_extreme === 'absorption' ? '#FF9800' : '#2196F3';
ctx.lineWidth = adv.bottom_extreme === 'absorption' ? 3 : 1.5;
ctx.beginPath();
ctx.moveTo(colX, bottomY);
ctx.lineTo(colX + colWidth, bottomY);
ctx.stroke();
}
}
// Calculate Cumulative Delta for bottom panel
cumulativeDelta += (cluster.total_delta || 0);
// Bottom Panel (Delta Histogram)
const panelY = height - bottomPanelHeight;
const cvdBaseline = panelY + bottomPanelHeight / 2;
// Delta cluster bar
const deltaVol = cluster.total_delta || 0;
const deltaColor = deltaVol >= 0 ? 'rgba(0, 230, 118, 0.7)' : 'rgba(255, 23, 68, 0.7)';
ctx.fillStyle = deltaColor;
// Scale: 1000 volume = 20px
const scaleFactor = 30 / 1000;
const barH = Math.min(Math.abs(deltaVol) * scaleFactor, bottomPanelHeight / 2 - 5);
const startY = deltaVol >= 0 ? cvdBaseline - barH : cvdBaseline;
ctx.fillRect(colX + 5, startY, colWidth - 10, barH);
// CVD line (Cumulative Delta)
ctx.fillStyle = cumulativeDelta >= 0 ? '#00E676' : '#FF1744';
ctx.font = '10px JetBrains Mono, monospace';
ctx.fillText(`CVD: ${cumulativeDelta.toFixed(0)}`, colX + colWidth / 2, panelY + 15);
});
// Draw Volume Profile Overlay Panel (Left Side)
const volProfileWidth = 140;
const volumeProfile = {};
let maxProfileVolume = 0;
clusters.forEach(cluster => {
const lvls = cluster.levels || {};
Object.keys(lvls).forEach(pStr => {
const p = parseFloat(pStr);
const data = lvls[pStr];
const binPrice = Math.round(p / actualTickSize) * actualTickSize;
volumeProfile[binPrice] = (volumeProfile[binPrice] || 0) + (data.total || 0);
if (volumeProfile[binPrice] > maxProfileVolume) {
maxProfileVolume = volumeProfile[binPrice];
}
});
});
// Value Area Calculation (70% of total volume)
let totalDayVolume = 0;
let dayPocPrice = null;
let dayPocVol = -1;
Object.keys(volumeProfile).forEach(pStr => {
const vol = volumeProfile[pStr];
totalDayVolume += vol;
if (vol > dayPocVol) {
dayPocVol = vol;
dayPocPrice = parseFloat(pStr);
}
});
let vah = dayPocPrice;
let val = dayPocPrice;
if (totalDayVolume > 0 && dayPocPrice !== null) {
let currentValVol = dayPocVol;
const targetVol = totalDayVolume * 0.70;
let upperPrice = dayPocPrice + actualTickSize;
let lowerPrice = dayPocPrice - actualTickSize;
while (currentValVol < targetVol) {
const upperVol = volumeProfile[upperPrice.toString()] || 0;
const lowerVol = volumeProfile[lowerPrice.toString()] || 0;
if (upperVol === 0 && lowerVol === 0) {
break; // No more volume to add
}
if (upperVol >= lowerVol) {
currentValVol += upperVol;
vah = upperPrice;
upperPrice += actualTickSize;
} else {
currentValVol += lowerVol;
val = lowerPrice;
lowerPrice -= actualTickSize;
}
}
}
// Draw VAH, VAL, and POC Lines across the chart
if (dayPocPrice !== null) {
const pocY = getPriceY(dayPocPrice);
const vahY = getPriceY(vah);
const valY = getPriceY(val);
// VAH Line
ctx.strokeStyle = 'rgba(148, 163, 184, 0.5)'; // Slate-400 dashed
ctx.lineWidth = 1;
ctx.setLineDash([4, 4]);
ctx.beginPath(); ctx.moveTo(0, vahY); ctx.lineTo(width, vahY); ctx.stroke();
// VAL Line
ctx.beginPath(); ctx.moveTo(0, valY); ctx.lineTo(width, valY); ctx.stroke();
ctx.setLineDash([]);
// POC Line
ctx.strokeStyle = 'rgba(255, 214, 0, 0.6)'; // Gold solid
ctx.lineWidth = 1.5;
ctx.beginPath(); ctx.moveTo(0, pocY); ctx.lineTo(width, pocY); ctx.stroke();
// POC Label
ctx.fillStyle = '#FFD600';
ctx.font = '10px JetBrains Mono, monospace';
ctx.textAlign = 'right';
ctx.fillText('POC', width - axisWidth - 5, pocY - 5);
}
if (maxProfileVolume > 0) {
// Solid background for the panel to cover grid/clusters underneath
ctx.fillStyle = 'rgba(15, 20, 30, 0.85)';
ctx.fillRect(0, 0, volProfileWidth, chartHeight);
// Right border of panel
ctx.strokeStyle = '#2A364F';
ctx.lineWidth = 1;
ctx.beginPath();
ctx.moveTo(volProfileWidth, 0);
ctx.lineTo(volProfileWidth, chartHeight);
ctx.stroke();
// Histogram bars
Object.keys(volumeProfile).forEach(pStr => {
const price = parseFloat(pStr);
const vol = volumeProfile[pStr];
const y = getPriceY(price) - rowHeight / 2;
const barWidth = (vol / maxProfileVolume) * (volProfileWidth - 5);
// Highlight bars inside Value Area
if (price <= vah && price >= val) {
ctx.fillStyle = 'rgba(59, 130, 246, 0.6)'; // Stronger blue for Value Area
} else {
ctx.fillStyle = 'rgba(59, 130, 246, 0.2)'; // Faded blue outside Value Area
}
ctx.fillRect(0, y, barWidth, rowHeight - 2);
});
// Panel Title
ctx.fillStyle = '#94A3B8';
ctx.font = '10px JetBrains Mono, monospace';
ctx.textAlign = 'center';
ctx.fillText('VOL PROFILE', volProfileWidth / 2, 20);
}
// Draw Vertical Price Axis (Right Side)
ctx.fillStyle = 'rgba(11, 14, 20, 0.95)';
ctx.fillRect(width - axisWidth, 0, axisWidth, chartHeight);
ctx.strokeStyle = '#1E293B';
ctx.lineWidth = 1.5;
ctx.beginPath();
ctx.moveTo(width - axisWidth, 0);
ctx.lineTo(width - axisWidth, chartHeight);
ctx.stroke();
// Render price tags along the axis
ctx.fillStyle = '#94A3B8';
ctx.font = '10px JetBrains Mono, monospace';
ctx.textAlign = 'left';
ctx.textBaseline = 'middle';
// Step every rowHeight
const startY = 0;
const endY = chartHeight;
// Draw tick labels on axis
for (let y = startY; y < endY; y += rowHeight) {
const price = getYPrice(y);
// Align price to tickSize
const roundedPrice = Math.round(price / tickSize) * tickSize;
const labelY = getPriceY(roundedPrice);
// Draw grid line connection to axis
ctx.strokeStyle = 'rgba(30, 41, 59, 0.5)';
ctx.lineWidth = 0.5;
ctx.beginPath();
ctx.moveTo(0, labelY);
ctx.lineTo(width - axisWidth, labelY);
ctx.stroke();
const decimals = tickSize < 1 ? Math.max(0, -Math.floor(Math.log10(tickSize))) : 2;
ctx.fillStyle = '#64748B';
ctx.fillText(`${roundedPrice.toFixed(decimals)}`, width - axisWidth + 8, labelY);
}
// Bottom Panel separator
const panelY = height - bottomPanelHeight;
ctx.fillStyle = '#111827';
ctx.fillRect(0, panelY, width, bottomPanelHeight);
ctx.strokeStyle = '#1E293B';
ctx.lineWidth = 2;
ctx.beginPath();
ctx.moveTo(0, panelY);
ctx.lineTo(width, panelY);
ctx.stroke();
// Draw CVD zero line
const cvdBaseline = panelY + bottomPanelHeight / 2;
ctx.strokeStyle = 'rgba(255, 255, 255, 0.1)';
ctx.lineWidth = 1;
ctx.setLineDash([5, 5]);
ctx.beginPath();
ctx.moveTo(0, cvdBaseline);
ctx.lineTo(width, cvdBaseline);
ctx.stroke();
ctx.setLineDash([]);
// CVD Label
ctx.fillStyle = '#94A3B8';
ctx.font = '12px Outfit, sans-serif';
ctx.textAlign = 'right';
ctx.fillText('CVD / Delta', width - 20, panelY + 20);
}, [clusters, scrollOffset]);
// Mouse Interaction: Panning/Scrolling
const handleMouseDown = (e) => {
const rect = canvasRef.current.getBoundingClientRect();
const x = e.clientX - rect.left;
const y = e.clientY - rect.top;
setIsDragging(true);
dragStart.current = { x, y };
dragOffsetStart.current = { ...scrollOffset };
};
const handleMouseMove = (e) => {
if (!isDragging) return;
const rect = canvasRef.current.getBoundingClientRect();
const x = e.clientX - rect.left;
const y = e.clientY - rect.top;
const dx = x - dragStart.current.x;
const dy = y - dragStart.current.y;
setScrollOffset({
x: dragOffsetStart.current.x + dx,
y: dragOffsetStart.current.y + dy
});
};
const handleMouseUp = () => {
setIsDragging(false);
};
// Wheel interaction for scrolling and zooming
const handleWheel = (e) => {
if (e.ctrlKey) {
// Zoom in/out
if (e.deltaY < 0) {
setZoom(z => Math.min(2.5, z + 0.1));
} else {
setZoom(z => Math.max(0.3, z - 0.1));
}
return;
}
// shift + wheel = horizontal scroll, normal wheel = vertical scroll
if (e.shiftKey) {
setScrollOffset(prev => ({ ...prev, x: prev.x - e.deltaY }));
} else {
setScrollOffset(prev => ({ ...prev, y: prev.y - e.deltaY * 0.5 }));
}
};
return (
<div className="relative w-full h-full cursor-grab active:cursor-grabbing select-none overflow-hidden rounded-xl border border-slate-800 bg-darkBg shadow-2xl">
<canvas
ref={canvasRef}
className="w-full h-full block"
onMouseDown={handleMouseDown}
onMouseMove={handleMouseMove}
onMouseUp={handleMouseUp}
onMouseLeave={handleMouseUp}
onWheel={handleWheel}
/>
<div className="absolute bottom-4 left-4 flex gap-2 items-center bg-darkBg/50 p-2 rounded-xl backdrop-blur-md border border-slate-800">
<button
onClick={() => setZoom(z => Math.max(0.3, z - 0.1))}
className="w-8 h-8 flex items-center justify-center bg-darkPanel border border-slate-700 rounded-lg text-lg font-bold text-slate-300 hover:text-white hover:bg-slate-800 transition"
title="Zoom Out"
>
-
</button>
<span className="text-xs font-mono text-slate-400 min-w-[35px] text-center">
{Math.round(zoom * 100)}%
</span>
<button
onClick={() => setZoom(z => Math.min(2.5, z + 0.1))}
className="w-8 h-8 flex items-center justify-center bg-darkPanel border border-slate-700 rounded-lg text-lg font-bold text-slate-300 hover:text-white hover:bg-slate-800 transition"
title="Zoom In"
>
+
</button>
<div className="w-px h-6 bg-slate-700 mx-1"></div>
<button
onClick={() => { setScrollOffset({ x: 50, y: 0 }); setZoom(1); }}
className="px-4 py-1.5 bg-darkPanel border border-slate-700 rounded-lg text-xs font-semibold text-slate-300 hover:text-white hover:bg-slate-800 transition"
>
Reset View
</button>
</div>
</div>
);
}
+33
View File
@@ -0,0 +1,33 @@
@tailwind base;
@tailwind components;
@tailwind utilities;
@layer base {
body {
background-color: #0B0E14;
font-family: 'Inter', -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, sans-serif;
}
h1, h2, h3, h4 {
font-family: 'Outfit', sans-serif;
}
}
/* Custom scrollbars */
::-webkit-scrollbar {
width: 6px;
height: 6px;
}
::-webkit-scrollbar-track {
background: #0B0E14;
}
::-webkit-scrollbar-thumb {
background: #1C2331;
border-radius: 4px;
}
::-webkit-scrollbar-thumb:hover {
background: #2D3748;
}
+10
View File
@@ -0,0 +1,10 @@
import React from 'react'
import ReactDOM from 'react-dom/client'
import App from './App.jsx'
import './index.css'
ReactDOM.createRoot(document.getElementById('root')).render(
<React.StrictMode>
<App />
</React.StrictMode>,
)
+55
View File
@@ -0,0 +1,55 @@
import { useEffect, useRef, useState } from 'react';
export function useWebSocket(url, onMessageCallback) {
const [status, setStatus] = useState('disconnected');
const wsRef = useRef(null);
const reconnectTimeoutRef = useRef(null);
useEffect(() => {
let active = true;
function connect() {
if (!active) return;
setStatus('connecting');
const ws = new WebSocket(url);
wsRef.current = ws;
ws.onopen = () => {
if (!active) return;
setStatus('connected');
};
ws.onmessage = (event) => {
if (!active) return;
try {
const data = JSON.parse(event.data);
onMessageCallback(data);
} catch (e) {
console.error("Error parsing WS message:", e);
}
};
ws.onclose = () => {
if (!active) return;
setStatus('disconnected');
reconnectTimeoutRef.current = setTimeout(connect, 3000);
};
ws.onerror = (err) => {
console.error("WS connection error:", err);
ws.close();
};
}
connect();
return () => {
active = false;
if (wsRef.current) wsRef.current.close();
if (reconnectTimeoutRef.current) clearTimeout(reconnectTimeoutRef.current);
};
}, [url, onMessageCallback]);
return status;
}
+86
View File
@@ -0,0 +1,86 @@
class AlertEngine {
constructor() {
this.processedClusters = new Set();
this.audioCache = {};
// We could load actual mp3/wav files here. For this implementation,
// we'll use the browser's SpeechSynthesis API as a fallback to actually say the alert out loud,
// which is very useful for trading without looking at the screen.
}
playAudio(message) {
if ('speechSynthesis' in window) {
const msg = new SpeechSynthesisUtterance(message);
msg.rate = 1.2;
msg.pitch = 1.1;
window.speechSynthesis.speak(msg);
} else {
console.log('Audio Alert:', message);
}
}
processClusters(clusters, pushToast) {
if (!clusters || clusters.length === 0) return;
clusters.forEach(cluster => {
// Only alert on closed clusters that we haven't processed yet
if (cluster.status === 'closed' && cluster.open_time && !this.processedClusters.has(cluster.open_time)) {
this.processedClusters.add(cluster.open_time);
const adv = cluster.advanced_metrics;
if (!adv) return;
let alerts = [];
// 1. P/B Pattern Alerts
if (adv.pattern === 'P') {
alerts.push({ type: 'info', msg: 'P Pattern Detected (Possible Short Covering)' });
} else if (adv.pattern === 'B') {
alerts.push({ type: 'info', msg: 'B Pattern Detected (Possible Long Liquidation)' });
}
// 2. Exhaustion/Absorption Alerts
if (adv.top_extreme === 'absorption') {
alerts.push({ type: 'warning', msg: 'Heavy Absorption at the Highs!' });
} else if (adv.top_extreme === 'exhaustion') {
alerts.push({ type: 'info', msg: 'Exhaustion at the Highs.' });
}
if (adv.bottom_extreme === 'absorption') {
alerts.push({ type: 'warning', msg: 'Heavy Absorption at the Lows!' });
} else if (adv.bottom_extreme === 'exhaustion') {
alerts.push({ type: 'info', msg: 'Exhaustion at the Lows.' });
}
// 3. Divergence Alerts
if (adv.delta_divergence) {
alerts.push({ type: 'error', msg: 'Delta Divergence! Price moving against Order Flow.' });
}
// 4. Ratio Extremes
if (adv.high_ratio && adv.high_ratio < 0.5) {
alerts.push({ type: 'warning', msg: `High Ratio Alert: ${adv.high_ratio}` });
}
if (adv.low_ratio && adv.low_ratio < 0.5) {
alerts.push({ type: 'warning', msg: `Low Ratio Alert: ${adv.low_ratio}` });
}
// Dispatch alerts
if (alerts.length > 0) {
// Play highest priority audio
const hasWarning = alerts.some(a => a.type === 'warning' || a.type === 'error');
if (hasWarning) {
this.playAudio(alerts[0].msg);
}
// Push UI toasts
if (pushToast) {
alerts.forEach(a => pushToast(a.msg, a.type));
}
}
}
});
}
}
export default new AlertEngine();
+19
View File
@@ -0,0 +1,19 @@
/** @type {import('tailwindcss').Config} */
export default {
content: [
"./index.html",
"./src/**/*.{js,ts,jsx,tsx}",
],
theme: {
extend: {
colors: {
darkBg: "#0B0E14",
darkPanel: "#151B26",
neonGreen: "#00E676",
neonRed: "#FF1744",
goldPOC: "#FFD600",
}
},
},
plugins: [],
}
+10
View File
@@ -0,0 +1,10 @@
import { defineConfig } from 'vite'
import react from '@vitejs/plugin-react'
// https://vitejs.dev/config/
export default defineConfig({
plugins: [react()],
server: {
port: 3000,
}
})
+4
View File
@@ -0,0 +1,4 @@
fastapi>=0.109.0
uvicorn>=0.27.0
MetaTrader5>=5.0.33
pandas>=2.0.0
+143
View File
@@ -0,0 +1,143 @@
import sys
import os
import unittest
# Add root folder to path so we can import config and backend
sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
from config import settings
from backend.aggregator import Aggregator, FootprintCluster, classify_tick
class TestAggregator(unittest.TestCase):
def test_tick_classification(self):
# TICK_FLAG_BUY = 0x2
# TICK_FLAG_SELL = 0x4
# 1. With flags
self.assertTrue(classify_tick(last=10.0, bid=9.0, ask=11.0, flags=0x2)) # Buy flag
self.assertFalse(classify_tick(last=10.0, bid=9.0, ask=11.0, flags=0x4)) # Sell flag
# Buy flag should take priority even if last <= bid
self.assertTrue(classify_tick(last=8.0, bid=9.0, ask=11.0, flags=0x2))
# 2. Fallbacks (no flags or flag=0)
self.assertTrue(classify_tick(last=11.5, bid=9.0, ask=11.0, flags=0)) # last >= ask
self.assertFalse(classify_tick(last=8.5, bid=9.0, ask=11.0, flags=0)) # last <= bid
# Mid-spread fallbacks
self.assertTrue(classify_tick(last=10.5, bid=9.0, ask=11.0, flags=0)) # closer to ask
self.assertFalse(classify_tick(last=9.5, bid=9.0, ask=11.0, flags=0)) # closer to bid
def test_poc_calculation_and_tie(self):
cluster = FootprintCluster(tick_size=1.0)
# Add ticks
cluster.add_tick(price=10.0, volume=100.0, is_buy=True, timestamp_msc=1000)
cluster.add_tick(price=11.0, volume=150.0, is_buy=False, timestamp_msc=1100)
cluster.add_tick(price=12.0, volume=50.0, is_buy=True, timestamp_msc=1200)
# POC should be 11.0 (highest volume 150)
self.assertEqual(cluster.poc, 11.0)
# Add more volume to 10.0 to create a tie of 150.0 with 11.0
cluster.add_tick(price=10.0, volume=50.0, is_buy=True, timestamp_msc=1300)
# 10.0 total = 150.0. 11.0 total = 150.0.
# Tie breaker rule: highest price level wins, so 11.0 should still be POC
self.assertEqual(cluster.poc, 11.0)
# Now let 12.0 tie with 150.0 (currently 50.0, add 100.0)
cluster.add_tick(price=12.0, volume=100.0, is_buy=False, timestamp_msc=1400)
# Tie between 10.0, 11.0, and 12.0. Highest price level is 12.0
self.assertEqual(cluster.poc, 12.0)
def test_diagonal_imbalances(self):
cluster = FootprintCluster(tick_size=1.0)
# Let's seed level 10.0 and 11.0
# At level 10.0, bid_vol = 10.0
# At level 11.0, ask_vol = 30.0
# ask_vol[11.0] (30.0) >= 3.0 * bid_vol[10.0] (10.0) -> Buy imbalance at 11.0!
cluster.add_tick(price=10.0, volume=10.0, is_buy=False, timestamp_msc=1000)
cluster.add_tick(price=11.0, volume=30.0, is_buy=True, timestamp_msc=1010)
levels_data = cluster.to_json()["levels"]
self.assertEqual(levels_data["11.00"]["imbalance"], "buy")
# At level 12.0, ask_vol = 50.0
# At level 11.0, bid_vol = 150.0
# bid_vol[11.0] (150.0) >= 3.0 * ask_vol[12.0] (50.0) -> Sell imbalance at 11.0!
cluster.add_tick(price=12.0, volume=50.0, is_buy=True, timestamp_msc=1020)
cluster.add_tick(price=11.0, volume=150.0, is_buy=False, timestamp_msc=1030)
levels_data = cluster.to_json()["levels"]
self.assertEqual(levels_data["11.00"]["imbalance"], "both")
def test_stacked_imbalances(self):
cluster = FootprintCluster(tick_size=1.0)
# Seed bid levels
cluster.add_tick(price=9.0, volume=10.0, is_buy=False, timestamp_msc=1000)
cluster.add_tick(price=10.0, volume=10.0, is_buy=False, timestamp_msc=1000)
cluster.add_tick(price=11.0, volume=10.0, is_buy=False, timestamp_msc=1000)
# Seed ask levels to trigger buy imbalances at 10.0, 11.0, 12.0
# ask[10.0] >= 3 * bid[9.0] -> ask[10.0] >= 30
cluster.add_tick(price=10.0, volume=30.0, is_buy=True, timestamp_msc=1000)
# ask[11.0] >= 3 * bid[10.0] -> ask[11.0] >= 30
cluster.add_tick(price=11.0, volume=30.0, is_buy=True, timestamp_msc=1000)
# ask[12.0] >= 3 * bid[11.0] -> ask[12.0] >= 30
cluster.add_tick(price=12.0, volume=30.0, is_buy=True, timestamp_msc=1000)
res = cluster.to_json()
self.assertTrue(res["stacked"]["buy"])
self.assertIn(10.0, res["stacked"]["price_range"])
self.assertIn(11.0, res["stacked"]["price_range"])
self.assertIn(12.0, res["stacked"]["price_range"])
def test_closure_criteria(self):
# Override settings programmatically
settings.CLUSTER_RANGE_POINTS = 10
settings.CLUSTER_VOLUME_MAX = 1000
settings.CLUSTER_DELTA_MAX = 500
settings.CLUSTER_TIME_SECONDS = 60
agg = Aggregator(tick_size=1.0)
# 1. Test closure by range
active, closed = agg.process_tick(price=100.0, volume=1.0, is_buy=True, timestamp_msc=1000)
self.assertIsNone(closed)
active, closed = agg.process_tick(price=110.0, volume=1.0, is_buy=True, timestamp_msc=1050)
self.assertIsNotNone(closed)
self.assertEqual(closed["close_reason"], "range")
# 2. Test closure by volume
agg = Aggregator(tick_size=1.0)
# Add 400 buy (delta=400, vol=400)
active, closed = agg.process_tick(price=100.0, volume=400.0, is_buy=True, timestamp_msc=1000)
self.assertIsNone(closed)
# Add 400 sell (delta=0, vol=800)
active, closed = agg.process_tick(price=100.0, volume=400.0, is_buy=False, timestamp_msc=1010)
self.assertIsNone(closed)
# Add 200 buy (delta=200, vol=1000) -> Should close due to volume
active, closed = agg.process_tick(price=100.0, volume=200.0, is_buy=True, timestamp_msc=1020)
self.assertIsNotNone(closed)
self.assertEqual(closed["close_reason"], "volume")
# 3. Test closure by delta
agg = Aggregator(tick_size=1.0)
active, closed = agg.process_tick(price=100.0, volume=499.0, is_buy=True, timestamp_msc=1000)
self.assertIsNone(closed)
active, closed = agg.process_tick(price=100.0, volume=1.0, is_buy=True, timestamp_msc=1010)
self.assertIsNotNone(closed)
self.assertEqual(closed["close_reason"], "delta")
# 4. Test closure by time
agg = Aggregator(tick_size=1.0)
active, closed = agg.process_tick(price=100.0, volume=1.0, is_buy=True, timestamp_msc=1000)
self.assertIsNone(closed)
active, closed = agg.process_tick(price=100.0, volume=1.0, is_buy=True, timestamp_msc=1000 + 60000)
self.assertIsNotNone(closed)
self.assertEqual(closed["close_reason"], "time")
if __name__ == "__main__":
unittest.main()