Realtime Orderbook Streaming¶
The orderbook is the raw microstructure of the market.
Every algo that optimises:
- entry quality
- slippage
- execution timing
- liquidity sensing
ultimately depends on live bid–ask depth, not just last traded price.
Nubra’s WebSocket Orderbook stream gives you tick-by-tick depth updates with deterministic structure, identical in UAT and LIVE.
What Is the Orderbook?¶
The 3 Layers of Market Truth¶
1 Last Trade
What just happened
Represents the most recent executed transaction in the market.
2 Bid Depth
Buy-side liquidity
Aggregated buy orders waiting in the market.
3 Ask Depth
Sell-side liquidity
Aggregated sell orders waiting in the market.
Realtime Orderbook — Visualised
This is a live WebSocket orderbook rendered in real time using
matplotlib.
- Tick-by-tick depth updates via Nubra WebSocket
- 20 levels of Bid and Ask depth
- Prices converted from paise → rupees
- No polling, no snapshots — pure streaming
Ideal for execution analysis, liquidity sensing, and slippage-aware strategies.
WebSocket OrderBook Stream and Visualization¶
Below is a ready-to-use example that streams a realtime 20-level orderbook over WebSocket and visualizes it using Matplotlib.
The WebSocket runs in a separate background thread so socket.keep_running() can receive data continuously without blocking the UI. Incoming ticks update a shared snapshot, while Matplotlib periodically reads this snapshot to redraw the orderbook.
Change the REF_ID to visualize the orderbook for any instrument.
#!/usr/bin/env python3
import threading
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation
from nubra_python_sdk.ticker import websocketdata
from nubra_python_sdk.start_sdk import InitNubraSdk, NubraEnv
# ============================================================
# CONFIG
# ============================================================
REF_ID = "1842210"
DEPTH = 20
UPDATE_MS = 300
# ============================================================
# Shared snapshot
# ============================================================
snapshot = {
"ltp": "–",
"ltq": "–",
"volume": "–",
"bids": [],
"asks": [],
}
lock = threading.Lock()
# ============================================================
# WebSocket callback
# ============================================================
def on_orderbook_data(msg):
with lock:
snapshot["ltp"] = f"{msg.last_traded_price / 100:.2f}"
snapshot["ltq"] = msg.last_traded_quantity
snapshot["volume"] = msg.volume
snapshot["bids"] = [
(lvl.price / 100, lvl.quantity, lvl.num_orders)
for lvl in msg.bids[:DEPTH]
]
snapshot["asks"] = [
(lvl.price / 100, lvl.quantity, lvl.num_orders)
for lvl in msg.asks[:DEPTH]
]
def on_connect(msg):
print("[WebSocket] Connected")
# ============================================================
# WebSocket thread
# ============================================================
def start_socket():
nubra = InitNubraSdk(NubraEnv.UAT,env_creds=True)
socket = websocketdata.NubraDataSocket(
client=nubra,
on_orderbook_data=on_orderbook_data,
on_connect=on_connect,
)
socket.connect()
socket.subscribe([REF_ID], data_type="orderbook")
socket.keep_running()
threading.Thread(target=start_socket, daemon=True).start()
# ============================================================
# Matplotlib Layout (ULTRA COMPACT)
# ============================================================
plt.style.use("dark_background")
fig = plt.figure(figsize=(10, 6), facecolor="black")
# Header cards (shorter)
ax_ltp = fig.add_axes([0.05, 0.84, 0.25, 0.095])
ax_ltq = fig.add_axes([0.375, 0.84, 0.25, 0.095])
ax_vol = fig.add_axes([0.70, 0.84, 0.25, 0.095])
# Tables (taller)
ax_bid = fig.add_axes([0.05, 0.08, 0.42, 0.74])
ax_ask = fig.add_axes([0.53, 0.08, 0.42, 0.74])
for ax in (ax_ltp, ax_ltq, ax_vol, ax_bid, ax_ask):
ax.axis("off")
# ============================================================
# UI helpers (SMALL TEXT)
# ============================================================
def draw_card(ax, title, value):
ax.clear()
ax.axis("off")
ax.add_patch(
plt.Rectangle(
(0, 0), 1, 1,
facecolor="#121212",
edgecolor="#2a2a2a",
linewidth=0.8
)
)
ax.text(
0.5, 0.62, title,
ha="center", va="center",
fontsize=8, color="#aaaaaa"
)
ax.text(
0.5, 0.30, f"{value}",
ha="center", va="center",
fontsize=12, weight="bold", color="white"
)
def draw_table(ax, title, rows, bg_color):
ax.clear()
ax.axis("off")
ax.text(
0.5, 0.985, title,
ha="center", va="bottom",
fontsize=9, weight="bold",
transform=ax.transAxes
)
if not rows:
rows = [["–", "–", "–"]]
formatted = []
for p, q, n in rows:
price = f"{p:.2f}" if isinstance(p, (int, float)) else p
formatted.append([price, q, n])
table = ax.table(
cellText=formatted,
colLabels=["Price", "Qty", "Orders"],
loc="center",
cellLoc="center",
colLoc="center",
)
table.scale(1, 1.05)
for (row, col), cell in table.get_celld().items():
cell.set_edgecolor("#2a2a2a")
cell.set_linewidth(0.6)
if row == 0:
cell.set_facecolor("#1f1f1f")
cell.set_text_props(color="white", weight="bold", fontsize=7)
else:
cell.set_facecolor(bg_color)
cell.set_text_props(color="white", fontsize=7)
cell.set_height(0.045) # 🔥 fits 20 rows cleanly
# ============================================================
# UI Update Loop
# ============================================================
def update(_):
with lock:
ltp = snapshot["ltp"]
ltq = snapshot["ltq"]
vol = snapshot["volume"]
bids = snapshot["bids"]
asks = snapshot["asks"]
draw_card(ax_ltp, "LTP", ltp)
draw_card(ax_ltq, "LTQ", ltq)
draw_card(ax_vol, "Volume", vol)
draw_table(ax_bid, "Bid", bids, "#0f2f1b")
draw_table(ax_ask, "Ask", asks, "#2f1414")
ani = FuncAnimation(
fig,
update,
interval=UPDATE_MS,
cache_frame_data=False
)
plt.show()
