Sc 12927/implement streaming support for ai answers result #94
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
WalkthroughThis PR introduces a comprehensive AI Answers API client with streaming and non-streaming support, consolidates sentiment rating functionality, and adds a setting to enable streaming mode. The old ai-answers-interactions-api module is removed, with its functionality migrated into the new ai-answers-api module. The apifetch layer is refactored to delegate AI Answers requests to the new client. Changes
Sequence Diagram(s)sequenceDiagram
participant Client
participant apifetch
participant executeAiAnswersFetch
participant Endpoint as AI Answers Endpoint
participant Callback
Client->>apifetch: POST with conversation settings
apifetch->>executeAiAnswersFetch: delegate with useStreaming flag
alt Streaming Mode (useStreaming = true)
executeAiAnswersFetch->>Endpoint: GET streaming endpoint (SSE)
loop Parse SSE Events
Endpoint-->>executeAiAnswersFetch: data: {type, payload}
executeAiAnswersFetch->>executeAiAnswersFetch: accumulate (conversation/token/sources)
Note over executeAiAnswersFetch: Throttle tokens (500ms),<br/>immediate for sources/complete
executeAiAnswersFetch->>Callback: invoke with partial AiAnswersResponse
end
Endpoint-->>executeAiAnswersFetch: [DONE] or stream end
executeAiAnswersFetch->>Callback: final with is_streaming_complete=true
else Non-Streaming Mode (useStreaming = false)
executeAiAnswersFetch->>Endpoint: POST to non-streaming endpoint
Endpoint-->>executeAiAnswersFetch: AiAnswersResponse
executeAiAnswersFetch->>Callback: invoke with complete response
end
Note over executeAiAnswersFetch: Error handling: HTTP errors,<br/>parse failures, stream interruption
Callback-->>Client: populated response or error
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes
Possibly related PRs
Suggested reviewers
Pre-merge checks and finishing touches✅ Passed checks (3 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Comment |
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (5)
src/ai-answers-api.ts(1 hunks)src/ai-answers-interactions-api.ts(0 hunks)src/apifetch.ts(3 hunks)src/index.ts(2 hunks)src/settings.ts(2 hunks)
💤 Files with no reviewable changes (1)
- src/ai-answers-interactions-api.ts
🧰 Additional context used
🧬 Code graph analysis (2)
src/apifetch.ts (1)
src/ai-answers-api.ts (1)
executeAiAnswersFetch(59-71)
src/ai-answers-api.ts (4)
src/settings.ts (1)
Settings(35-72)src/apifetch.ts (1)
ApiFetchCallback(50-52)src/api.ts (2)
RESPONSE_SERVER_ERROR(44-44)aiAnswersInteractionsInstance(41-41)src/index.ts (1)
putSentimentClick(142-147)
Comment on lines
+99
to
+118
| fetch(streamingEndpoint, { | ||
| method: 'POST', | ||
| headers: { | ||
| 'Content-Type': 'application/json' | ||
| }, | ||
| body: JSON.stringify({ | ||
| question: settings?.keyword, | ||
| filter: settings?.aiAnswersFilterObject | ||
| }) | ||
| }) | ||
| .then(async (response) => { | ||
| if (!response.ok) { | ||
| throw new Error(`HTTP error! status: ${response.status}`); | ||
| } | ||
|
|
||
| const reader = response.body?.getReader(); | ||
| const decoder = new TextDecoder(); | ||
|
|
||
| if (!reader) { | ||
| throw new Error('No response body reader available'); |
There was a problem hiding this comment.
Streaming fetch bypasses request interceptors
Both the streaming and non‑streaming code paths call fetch directly. That bypasses the apiInstance (and its interceptor stack) that clients wire up through AddSearchClient.setApiRequestInterceptor. Today customers rely on that hook to inject private-key/authorization headers and other per-request mutations. With this change those headers are never applied, so AI Answers requests will start failing as soon as an interceptor is required. Please route these calls back through apiInstance (or otherwise execute the same interceptor pipeline before issuing the request) so configured interceptors continue to run.
Also applies to: 323-338
🤖 Prompt for AI Agents
In src/ai-answers-api.ts around lines 99 to 118 (and similarly around 323 to
338), the direct fetch calls bypass the apiInstance interceptor pipeline so
per-request interceptors (auth/private-key/etc.) are never applied; change the
code to route requests through apiInstance so the same interceptor stack is
executed before sending the request — for non-streaming use apiInstance.request
(or equivalent wrapper) with the same method, headers and JSON body; for
streaming where you currently call fetch and use response.body.getReader(), call
the apiInstance variant that returns a Response (or run the interceptor chain to
produce a finalized Request/RequestInit and then call fetch) so the
response/reader logic remains identical but interceptors are applied; update
both the streaming and non-streaming paths (lines noted) to use apiInstance and
preserve error/reader handling.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary by CodeRabbit
Release Notes