jambonz-python-sdk 0.2.0__py3-none-any.whl

jambonz_sdk/types/components.py

@@ -0,0 +1,250 @@
+"""Shared component type definitions matching jambonz JSON schemas.
+
+All TypedDict keys use the exact same names as the jambonz JSON schemas
+(camelCase where applicable) to ensure serialization compatibility.
+"""
+
+from __future__ import annotations
+
+from typing import Any, TypedDict, Union
+
+
+class Auth(TypedDict, total=False):
+    """Authentication credentials."""
+
+    username: str
+    password: str
+    apiKey: str
+    agent_id: str
+    api_key: str
+
+
+class ActionHookObject(TypedDict, total=False):
+    """ActionHook as an object with URL and options."""
+
+    url: str
+    method: str  # "GET" | "POST"
+    basicAuth: Auth
+
+
+# ActionHook can be a simple URL string or an object
+ActionHook = Union[str, ActionHookObject]
+
+
+class Synthesizer(TypedDict, total=False):
+    """Text-to-speech configuration."""
+
+    vendor: str
+    label: str
+    language: str
+    voice: str | dict[str, Any]
+    fallbackVendor: str
+    fallbackLabel: str
+    fallbackLanguage: str
+    fallbackVoice: str | dict[str, Any]
+    engine: str  # "standard" | "neural" | "generative" | "long-form"
+    gender: str  # "MALE" | "FEMALE" | "NEUTRAL"
+    options: dict[str, Any]
+
+
+class Vad(TypedDict, total=False):
+    """Voice Activity Detection configuration."""
+
+    enable: bool
+    voiceMs: int
+    silenceMs: int
+    strategy: str
+    mode: int  # 0-3
+    vendor: str  # "webrtc" | "silero"
+    threshold: float  # 0-1
+    speechPadMs: int
+
+
+class Recognizer(TypedDict, total=False):
+    """Speech-to-text recognition configuration."""
+
+    vendor: str
+    label: str
+    language: str
+    fallbackVendor: str
+    fallbackLabel: str
+    fallbackLanguage: str
+    vad: Vad
+    autogeneratePrompt: bool
+    hints: list[str]
+    hintsBoost: float
+    altLanguages: list[str]
+    profanityFilter: bool
+    interim: bool
+    singleUtterance: bool
+    dualChannel: bool
+    separateRecognitionPerChannel: bool
+    punctuation: bool
+    enhancedModel: bool
+    words: bool
+    diarization: bool
+    diarizationMinSpeakers: int
+    diarizationMaxSpeakers: int
+    interactionType: str
+    naicsCode: int
+    identifyChannels: bool
+    vocabularyName: str
+    vocabularyFilterName: str
+    filterMethod: str  # "remove" | "mask" | "tag"
+    model: str
+    outputFormat: str  # "simple" | "detailed"
+    profanityOption: str  # "masked" | "removed" | "raw"
+    requestSnr: bool
+    initialSpeechTimeoutMs: int
+    azureServiceEndpoint: str
+    azureSttEndpointId: str
+    asrDtmfTerminationDigit: str
+    asrTimeout: int
+    fastRecognitionTimeout: int
+    minConfidence: float
+    deepgramOptions: dict[str, Any]
+    googleOptions: dict[str, Any]
+    awsOptions: dict[str, Any]
+    azureOptions: dict[str, Any]
+    nuanceOptions: dict[str, Any]
+    ibmOptions: dict[str, Any]
+    nvidiaOptions: dict[str, Any]
+    sonioxOptions: dict[str, Any]
+    cobaltOptions: dict[str, Any]
+    assemblyAiOptions: dict[str, Any]
+    speechmaticsOptions: dict[str, Any]
+    openaiOptions: dict[str, Any]
+    houndifyOptions: dict[str, Any]
+    gladiaOptions: dict[str, Any]
+    elevenlabsOptions: dict[str, Any]
+    verbioOptions: dict[str, Any]
+    customOptions: dict[str, Any]
+
+
+class FromHeader(TypedDict, total=False):
+    """SIP From header override."""
+
+    user: str
+    host: str
+
+
+class Target(TypedDict, total=False):
+    """Call target for the dial verb."""
+
+    type: str  # "phone" | "sip" | "user" | "teams"
+    number: str
+    sipUri: str
+    name: str
+    tenant: str
+    trunk: str
+    confirmHook: ActionHook
+    method: str  # "GET" | "POST"
+    headers: dict[str, str]
+    from_: FromHeader  # Note: 'from' is reserved in Python
+    auth: Auth
+    vmail: bool
+    overrideTo: str
+    proxy: str
+
+
+class AmdTimers(TypedDict, total=False):
+    """AMD timer configuration."""
+
+    noSpeechTimeoutMs: int
+    decisionTimeoutMs: int
+    toneTimeoutMs: int
+    greetingCompletionTimeoutMs: int
+
+
+class Amd(TypedDict, total=False):
+    """Answering machine detection configuration."""
+
+    actionHook: ActionHook
+    thresholdWordCount: int
+    digitCount: int
+    timers: AmdTimers
+    recognizer: Recognizer
+
+
+class BidirectionalAudio(TypedDict, total=False):
+    """Bidirectional audio streaming configuration."""
+
+    enabled: bool
+    streaming: bool
+    sampleRate: int
+
+
+class FillerNoise(TypedDict, total=False):
+    """Filler noise configuration."""
+
+    enable: bool
+    url: str
+    startDelaySecs: int
+
+
+class ActionHookDelayAction(TypedDict, total=False):
+    """Configuration for handling slow webhook responses."""
+
+    enabled: bool
+    noResponseTimeout: int
+    noResponseGiveUpTimeout: int
+    retries: int
+    actions: list[dict[str, Any]]
+
+
+class McpServer(TypedDict, total=False):
+    """MCP server configuration."""
+
+    url: str
+    auth: dict[str, Any]
+    roots: list[dict[str, Any]]
+
+
+class LlmBase(TypedDict, total=False):
+    """Shared properties for LLM/S2S verbs."""
+
+    vendor: str
+    model: str
+    auth: Auth
+    connectOptions: dict[str, Any]
+    llmOptions: dict[str, Any]
+    mcpServers: list[McpServer]
+    actionHook: ActionHook
+    eventHook: ActionHook
+    toolHook: ActionHook
+    events: list[str]
+
+
+class BargeIn(TypedDict, total=False):
+    """Barge-in configuration."""
+
+    enable: bool
+    sticky: bool
+    actionHook: ActionHook
+    input: list[str]
+    minBargeinWordCount: int
+
+
+class TtsStream(TypedDict, total=False):
+    """TTS streaming configuration."""
+
+    enable: bool
+    synthesizer: Synthesizer
+
+
+class NoiseIsolation(TypedDict, total=False):
+    """Noise isolation configuration."""
+
+    enable: bool
+    vendor: str
+    level: int
+    model: str
+
+
+class TurnTaking(TypedDict, total=False):
+    """Turn-taking detection configuration."""
+
+    enable: bool
+    vendor: str
+    threshold: float
+    model: str

jambonz_sdk/types/rest.py

@@ -0,0 +1,59 @@
+"""REST API request/response types."""
+
+from __future__ import annotations
+
+from typing import Any, TypedDict
+
+
+class CreateCallRequest(TypedDict, total=False):
+    """Request body for creating an outbound call."""
+
+    from_: str  # Serialized as 'from'
+    to: dict[str, Any]  # Target object
+    call_hook: str
+    call_status_hook: str
+    timeout: int
+    tag: dict[str, Any]
+    headers: dict[str, str]
+    caller_name: str
+
+
+class ListCallsFilter(TypedDict, total=False):
+    """Filter parameters for listing calls."""
+
+    direction: str  # "inbound" | "outbound"
+    from_: str
+    to: str
+    callStatus: str
+
+
+class CallStatus(TypedDict, total=False):
+    """Call status information."""
+
+    call_sid: str
+    account_sid: str
+    application_sid: str
+    direction: str
+    from_: str
+    to: str
+    call_id: str
+    sip_status: int
+    call_status: str
+    duration: int
+
+
+class CallInfo(TypedDict, total=False):
+    """Full call information returned by the API."""
+
+    call_sid: str
+    account_sid: str
+    application_sid: str
+    direction: str
+    from_: str
+    to: str
+    call_id: str
+    sip_status: int
+    call_status: str
+    caller_name: str
+    duration: int
+    trace_id: str

jambonz_sdk/types/session.py

@@ -0,0 +1,55 @@
+"""Call session and WebSocket message types."""
+
+from __future__ import annotations
+
+from enum import Enum
+from typing import Any, TypedDict
+
+
+class CallDirection(str, Enum):
+    """Call direction."""
+
+    INBOUND = "inbound"
+    OUTBOUND = "outbound"
+
+
+class CallSession(TypedDict, total=False):
+    """Call session data received on session:new."""
+
+    call_sid: str
+    account_sid: str
+    application_sid: str
+    direction: str  # "inbound" | "outbound"
+    from_: str  # Serialized as 'from'
+    to: str
+    call_id: str
+    sip_status: int
+    sip: dict[str, Any]
+    env_vars: dict[str, str]
+    defaults: dict[str, Any]
+    customerData: dict[str, Any]
+
+
+class WsMessageType(str, Enum):
+    """WebSocket message types from jambonz."""
+
+    SESSION_NEW = "session:new"
+    SESSION_REDIRECT = "session:redirect"
+    SESSION_RECONNECT = "session:reconnect"
+    SESSION_ADULTING = "session:adulting"
+    VERB_HOOK = "verb:hook"
+    VERB_STATUS = "verb:status"
+    CALL_STATUS = "call:status"
+    LLM_TOOL_CALL = "llm:tool-call"
+    LLM_EVENT = "llm:event"
+    TTS_TOKENS_RESULT = "tts:tokens-result"
+    TTS_STREAMING_EVENT = "tts:streaming-event"
+
+
+class WsMessage(TypedDict, total=False):
+    """WebSocket message structure."""
+
+    type: str
+    msgid: str
+    hook: str
+    data: dict[str, Any]