Class: StreamProcessor#
Defined in: activities/chat/stream/processor.ts:168
StreamProcessor - State machine for processing AI response streams
Manages the full UIMessage[] conversation and emits events on changes.
State tracking:
- Full message array
- Current assistant message being streamed
- Text content accumulation
- Multiple parallel tool calls
- Tool call completion detection
Tool call completion is detected when:
- A new tool call starts at a different index
- Text content arrives
- Stream ends
Constructors#
Constructor#
new StreamProcessor(options): StreamProcessor;
Defined in: activities/chat/stream/processor.ts:197
Parameters#
options#
Returns#
StreamProcessor
Methods#
addToolApprovalResponse()#
addToolApprovalResponse(approvalId, approved): void;
Defined in: activities/chat/stream/processor.ts:311
Add an approval response (called by client after handling onApprovalRequest)
Parameters#
approvalId#
string
approved#
boolean
Returns#
void
addToolResult()#
addToolResult(
toolCallId,
output,
error?): void;
Defined in: activities/chat/stream/processor.ts:267
Add a tool result (called by client after handling onToolCall)
Parameters#
toolCallId#
string
output#
any
error?#
string
Returns#
void
addUserMessage()#
addUserMessage(content): UIMessage;
Defined in: activities/chat/stream/processor.ts:225
Add a user message to the conversation
Parameters#
content#
string
Returns#
areAllToolsComplete()#
areAllToolsComplete(): boolean;
Defined in: activities/chat/stream/processor.ts:342
Check if all tool calls in the last assistant message are complete Useful for auto-continue logic
Returns#
boolean
clearMessages()#
clearMessages(): void;
Defined in: activities/chat/stream/processor.ts:374
Clear all messages
Returns#
void
finalizeStream()#
finalizeStream(): void;
Defined in: activities/chat/stream/processor.ts:948
Finalize the stream - complete all pending operations
Returns#
void
getMessages()#
getMessages(): UIMessage[];
Defined in: activities/chat/stream/processor.ts:334
Get current messages
Returns#
getRecording()#
getRecording(): ChunkRecording | null;
Defined in: activities/chat/stream/processor.ts:1034
Get the current recording
Returns#
ChunkRecording | null
getState()#
getState(): ProcessorState;
Defined in: activities/chat/stream/processor.ts:1007
Get current processor state (legacy)
Returns#
process()#
process(stream): Promise<ProcessorResult>;
Defined in: activities/chat/stream/processor.ts:387
Process a stream and emit events through handlers
Parameters#
stream#
AsyncIterable<any>
Returns#
Promise<ProcessorResult>
processChunk()#
processChunk(chunk): void;
Defined in: activities/chat/stream/processor.ts:415
Process a single chunk from the stream
Parameters#
chunk#
Returns#
void
removeMessagesAfter()#
removeMessagesAfter(index): void;
Defined in: activities/chat/stream/processor.ts:366
Remove messages after a certain index (for reload/retry)
Parameters#
index#
number
Returns#
void
reset()#
reset(): void;
Defined in: activities/chat/stream/processor.ts:1057
Full reset (including messages)
Returns#
void
setMessages()#
setMessages(messages): void;
Defined in: activities/chat/stream/processor.ts:217
Set the messages array (e.g., from persisted state)
Parameters#
messages#
Returns#
void
startAssistantMessage()#
startAssistantMessage(): string;
Defined in: activities/chat/stream/processor.ts:243
Start streaming a new assistant message Returns the message ID
Returns#
string
startRecording()#
startRecording(): void;
Defined in: activities/chat/stream/processor.ts:1021
Start recording chunks
Returns#
void
toModelMessages()#
toModelMessages(): ModelMessage<
| string
| ContentPart<unknown, unknown, unknown, unknown, unknown>[]
| null>[];
Defined in: activities/chat/stream/processor.ts:323
Get the conversation as ModelMessages (for sending to LLM)
Returns#
ModelMessage<
| string
| ContentPart<unknown, unknown, unknown, unknown, unknown>[]
| null>[]
replay()#
static replay(recording, options?): Promise<ProcessorResult>;
Defined in: activities/chat/stream/processor.ts:1066
Replay a recording through the processor
Parameters#
recording#
options?#
Returns#
Promise<ProcessorResult>