Presets

Pre-configured WebSocket profiles for common use cases.

Optimized for exchange WebSocket connections.

Features: - Fast reconnect (0.5s base, 30s cap, infinite retries) - Aggressive heartbeat (20s interval) - Large message buffer (10K messages) with deduplication - Proactive reconnect before 24h exchange limit - Compression enabled

Parameters:

Name	Type	Description	Default
uri	str	WebSocket URI.	required
**overrides	Any	Override any default setting.	{}

Returns:

Type	Description
WebSocket[Any]	Configured WebSocket.

Source code in src/jetsocket/presets.py

def trading(uri: str, **overrides: Any) -> WebSocket[Any]:
    """Optimized for exchange WebSocket connections.

    Features:
    - Fast reconnect (0.5s base, 30s cap, infinite retries)
    - Aggressive heartbeat (20s interval)
    - Large message buffer (10K messages) with deduplication
    - Proactive reconnect before 24h exchange limit
    - Compression enabled

    Args:
        uri: WebSocket URI.
        **overrides: Override any default setting.

    Returns:
        Configured WebSocket.
    """
    defaults: dict[str, Any] = {
        "reconnect": True,
        "backoff": BackoffConfig(
            base=0.5,
            multiplier=2.0,
            cap=30.0,
            jitter=True,
            max_attempts=0,
            reset_after=60.0,
        ),
        "heartbeat": HeartbeatConfig(interval=20.0, timeout=10.0),
        "buffer": BufferConfig(
            capacity=10_000,
            overflow_policy="drop_oldest",
            enable_dedup=True,
            dedup_window=1000,
        ),
        "max_connection_age": 85800.0,
        "compress": True,
    }
    defaults.update(overrides)
    return WebSocket(uri, **defaults)

Optimized for LLM streaming (large messages, low latency).

Features: - Quick retry with 5 max attempts - Longer heartbeat (30s, less overhead) - Smaller buffer (500, sequential streaming) - Large max message size (128MB) - Compression disabled (token streaming doesn't benefit)

Parameters:

Name	Type	Description	Default
uri	str	WebSocket URI.	required
**overrides	Any	Override any default setting.	{}

Returns:

Type	Description
WebSocket[Any]	Configured WebSocket.

Source code in src/jetsocket/presets.py

def llm_stream(uri: str, **overrides: Any) -> WebSocket[Any]:
    """Optimized for LLM streaming (large messages, low latency).

    Features:
    - Quick retry with 5 max attempts
    - Longer heartbeat (30s, less overhead)
    - Smaller buffer (500, sequential streaming)
    - Large max message size (128MB)
    - Compression disabled (token streaming doesn't benefit)

    Args:
        uri: WebSocket URI.
        **overrides: Override any default setting.

    Returns:
        Configured WebSocket.
    """
    defaults: dict[str, Any] = {
        "reconnect": True,
        "backoff": BackoffConfig(
            base=1.0,
            multiplier=2.0,
            cap=10.0,
            jitter=True,
            max_attempts=5,
            reset_after=30.0,
        ),
        "heartbeat": HeartbeatConfig(interval=30.0, timeout=15.0),
        "buffer": BufferConfig(capacity=500, overflow_policy="drop_oldest"),
        "max_message_size": 128 * 1024 * 1024,
        "compress": False,
    }
    defaults.update(overrides)
    return WebSocket(uri, **defaults)

Optimized for real-time dashboard feeds.

Features: - Relaxed reconnect (2s base, 60s cap) - Standard heartbeat (30s) - Small buffer (100, latest data only) - Drop oldest on overflow

Parameters:

Name	Type	Description	Default
uri	str	WebSocket URI.	required
**overrides	Any	Override any default setting.	{}

Returns:

Type	Description
WebSocket[Any]	Configured WebSocket.

Source code in src/jetsocket/presets.py

def dashboard(uri: str, **overrides: Any) -> WebSocket[Any]:
    """Optimized for real-time dashboard feeds.

    Features:
    - Relaxed reconnect (2s base, 60s cap)
    - Standard heartbeat (30s)
    - Small buffer (100, latest data only)
    - Drop oldest on overflow

    Args:
        uri: WebSocket URI.
        **overrides: Override any default setting.

    Returns:
        Configured WebSocket.
    """
    defaults: dict[str, Any] = {
        "reconnect": True,
        "backoff": BackoffConfig(
            base=2.0,
            multiplier=2.0,
            cap=60.0,
            jitter=True,
            max_attempts=0,
            reset_after=120.0,
        ),
        "heartbeat": HeartbeatConfig(interval=30.0, timeout=20.0),
        "buffer": BufferConfig(capacity=100, overflow_policy="drop_oldest"),
    }
    defaults.update(overrides)
    return WebSocket(uri, **defaults)

Minimal configuration with basic reconnect.

Features: - Basic reconnect with 10 max attempts - No heartbeat - No buffer

Parameters:

Name	Type	Description	Default
uri	str	WebSocket URI.	required
**overrides	Any	Override any default setting.	{}

Returns:

Type	Description
WebSocket[Any]	Configured WebSocket.

Source code in src/jetsocket/presets.py

def minimal(uri: str, **overrides: Any) -> WebSocket[Any]:
    """Minimal configuration with basic reconnect.

    Features:
    - Basic reconnect with 10 max attempts
    - No heartbeat
    - No buffer

    Args:
        uri: WebSocket URI.
        **overrides: Override any default setting.

    Returns:
        Configured WebSocket.
    """
    defaults: dict[str, Any] = {
        "reconnect": True,
        "backoff": BackoffConfig(
            base=1.0,
            multiplier=2.0,
            cap=30.0,
            jitter=True,
            max_attempts=10,
            reset_after=60.0,
        ),
        "heartbeat": None,
        "buffer": None,
    }
    defaults.update(overrides)
    return WebSocket(uri, **defaults)

Usage

from jetsocket.presets import trading, llm_stream, dashboard

# For crypto trading
ws = trading("wss://stream.binance.com/ws")

# For LLM streaming
ws = llm_stream("wss://api.openai.com/v1/realtime")

# For dashboards
ws = dashboard("wss://dashboard.example.com/ws")

Preset Details

trading()

Optimized for exchange WebSocket connections:

Setting	Value	Rationale
Backoff base	0.5s	Fast initial reconnect
Backoff cap	30s	Don't wait too long
Max attempts	Infinite	Never give up
Heartbeat interval	20s	Aggressive keep-alive
Buffer capacity	10,000	Large for high throughput
Deduplication	Enabled	Prevent duplicate trades
Max connection age	23h 50m	Rotate before 24h limit
Compression	Enabled	Reduce bandwidth

llm_stream()

Optimized for LLM streaming APIs:

Setting	Value	Rationale
Backoff base	1s	Quick retry
Backoff cap	10s	Fast fail for UX
Max attempts	5	Don't retry forever
Heartbeat interval	30s	Less overhead
Buffer capacity	500	Smaller (sequential)
Max message size	128MB	Large responses
Compression	Disabled	Token streaming doesn't benefit

dashboard()

Optimized for real-time dashboard feeds:

Setting	Value	Rationale
Backoff base	2s	Relaxed reconnect
Backoff cap	60s	Patient retry
Max attempts	Infinite	Always reconnect
Heartbeat interval	30s	Standard keep-alive
Buffer capacity	100	Small (latest data only)
Overflow policy	drop_oldest	Stale data is useless

minimal()

Basic configuration with sensible defaults:

Setting	Value
Reconnect	Enabled
Heartbeat interval	30s
Buffer capacity	100

Overriding Defaults

All presets accept **overrides to customize settings:

from jetsocket.presets import trading, llm_stream

# Trading preset with custom buffer
ws = trading(
    "wss://stream.binance.com/ws",
    buffer=BufferConfig(capacity=50000),  # Override buffer
)

# LLM preset with longer timeout
ws = llm_stream(
    "wss://api.openai.com/v1/realtime",
    heartbeat=HeartbeatConfig(interval=60.0, timeout=30.0),
)