Skip to content

fix(core): token usage and response event type for none stream mode. #524

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sunrabbit123 merged 1 commit into from
Jan 28, 2026
Merged

fix(core): token usage and response event type for none stream mode. #524

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 13 additions & 2 deletions packages/core/src/events/AgenticaResponseEvent.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,9 @@ export interface AgenticaResponseEvent extends AgenticaEventBase<"response"> {
body: OpenAI.ChatCompletionCreateParamsStreaming;
Copy link

Copilot AI Jan 28, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The body property is typed as OpenAI.ChatCompletionCreateParamsStreaming, but it should support both streaming and non-streaming requests to match the actual usage in the code. When a non-streaming response is created (when config.stream === false), the body will have stream: false and should be typed as OpenAI.ChatCompletionCreateParamsNonStreaming. Consider changing this to OpenAI.ChatCompletionCreateParamsStreaming | OpenAI.ChatCompletionCreateParamsNonStreaming to match the AgenticaRequestEvent.body type and accurately represent both streaming and non-streaming scenarios.

Suggested change
body: OpenAI.ChatCompletionCreateParamsStreaming;
body:
| OpenAI.ChatCompletionCreateParamsStreaming
| OpenAI.ChatCompletionCreateParamsNonStreaming;

Copilot uses AI. Check for mistakes.

/**
* The text content stream.
* The response data.
*/
stream: AsyncGenerator<OpenAI.ChatCompletionChunk, undefined, undefined>;
response: AgenticaResponseEvent.Response;

/**
* Options for the request.
Expand All @@ -31,3 +31,14 @@ export interface AgenticaResponseEvent extends AgenticaEventBase<"response"> {
*/
join: () => Promise<OpenAI.ChatCompletion>;
}
export namespace AgenticaResponseEvent {
export type Response = StreamResponse | NonStreamResponse;
export interface StreamResponse {
stream: true;
data: AsyncGenerator<OpenAI.ChatCompletionChunk, undefined, undefined>;
}
export interface NonStreamResponse {
stream: false;
data: OpenAI.ChatCompletion;
}
}
4 changes: 2 additions & 2 deletions packages/core/src/factory/events.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -330,7 +330,7 @@ export function createResponseEvent(props: {
source: AgenticaEventSource;
body: OpenAI.ChatCompletionCreateParamsStreaming;
options?: OpenAI.RequestOptions | undefined;
stream: AsyncGenerator<OpenAI.ChatCompletionChunk, undefined, undefined>;
response: AgenticaResponseEvent.Response;
join: () => Promise<OpenAI.ChatCompletion>;
}): AgenticaResponseEvent {
const id: string = v4();
Expand All @@ -343,7 +343,7 @@ export function createResponseEvent(props: {
source: props.source,
body: props.body,
options: props.options,
stream: props.stream,
response: props.response,
join: props.join,
};
}
26 changes: 25 additions & 1 deletion packages/core/src/utils/request.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,6 +63,27 @@ export function getChatCompletionFunction(props: {
})();

if ("toReadableStream" in completion === false) {
if (completion.usage != null) {
AgenticaTokenUsageAggregator.aggregate({
kind: source,
completionUsage: completion.usage,
usage: props.usage,
});
}
void props.dispatch({
id: v4(),
type: "response",
request_id: event.id,
source,
response: {
stream: false,
data: completion,
},
body: event.body as OpenAI.ChatCompletionCreateParamsStreaming,
options: event.options,
join: async () => completion,
created_at: new Date().toISOString(),
}).catch(() => {});
return {
type: "none-stream",
value: completion,
Expand Down Expand Up @@ -101,7 +122,10 @@ export function getChatCompletionFunction(props: {
type: "response",
request_id: event.id,
source,
stream: streamDefaultReaderToAsyncGenerator(streamForStream.getReader(), props.abortSignal),
response: {
stream: true,
data: streamDefaultReaderToAsyncGenerator(streamForStream.getReader(), props.abortSignal),
},
body: event.body as OpenAI.ChatCompletionCreateParamsStreaming,
options: event.options,
join: async () => {
Expand Down
86 changes: 43 additions & 43 deletions pnpm-lock.yaml
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

4 changes: 2 additions & 2 deletions pnpm-workspace.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,8 +21,8 @@ catalogs:
typescript: ~5.9.3
typia:
"@ryoppippi/unplugin-typia": ^2.6.4
"@samchon/openapi": ^6.0.0
typia: ^11.0.0
"@samchon/openapi": ^6.0.1
typia: ^11.0.3
vite:
vitest: ^3.0.9
onlyBuiltDependencies:
Expand Down
Loading
Loading