feat(embeddings): emit UsageEvent on /v1/embeddings 200 (#226 MVP)

[moonming]

Summary

Refs #226 — embeddings only; remaining 5 endpoints tracked under #226 with checkboxes ticked as each follow-up PR lands.

Pre-fix, `/v1/embeddings` dropped the `UsageEvent` entirely. Every embeddings request was invisible to cp-api's budget ledger and the customer-facing `/logs` analytics. `chat.rs` has emitted `UsageEvent` since #302 M17 — the gap was specifically on the non-chat OpenAI-shape handlers.

This PR ships the embeddings half: a successful `/v1/embeddings` call now emits a `UsageEvent` on the configured sink with the upstream-reported `prompt_tokens`, the resolved `model_id`, the authenticated `api_key_id`, the status code, and `inbound_protocol = "openai"`.

Scope

MVP: /v1/embeddings only. Follow-ups for /v1/completions (#403), /v1/responses (#404), /v1/rerank (#405), /v1/audio/* (#406), /v1/images/* (#407) — same shape, same emit-on-success-only convention.

Per-PK telemetry attribution (`provider_kind` / `featured` / `branded_provider` / `pk_label` / `byo_label`) is wired in `chat.rs` only today. Adding it to the non-chat handlers is also a follow-up; backward-compat is preserved here because the `UsageEvent` defaults map empty/false → cp-api stores NULL.

Emission convention

Mirrors `chat.rs::emit_usage_event` — emit when an upstream call was made:

Path	Emit?	Reason
200 OK	yes	upstream returned a response — attribute spend (even when `prompt_tokens=0`)
501 NotImplemented	no	no upstream call happened (`upstream_called: false`) — skip
upstream 4xx/5xx error	no	no usage data to attribute

Gating is on the explicit `EmbedDispatchSuccess.upstream_called` flag (audit M1 fix — see commit 164c364) rather than a `prompt_tokens > 0` sentinel, so a future provider that legitimately reports zero tokens on a 200 still gets attributed.

Audit response

Independent agent review found 2 MEDIUM + helpful LOWs. Resolutions:

• M1 — `prompt_tokens > 0` sentinel: addressed in commit 164c364. New `upstream_called: bool` flag + regression test `emits_usage_event_on_200_with_zero_prompt_tokens_audit_m1`.
• M2 — `provider.to_ascii_lowercase()` compilation: dispelled. `dispatch::require_provider` returns `&str`, so `str::to_ascii_lowercase()` is valid (returns `String`). `cargo build` + `cargo test` green on every revision.
• L1 — no e2e observable for emission: filed as follow-up (will be added with the prometheus counter when the rest of Non-chat endpoints don't emit UsageEvents — cp-api's spend ledger misses spend from /v1/responses, /v1/embeddings, /v1/audio*, /v1/images*, /v1/rerank #226 lands).
• L5 — `Fixes Non-chat endpoints don't emit UsageEvents — cp-api's spend ledger misses spend from /v1/responses, /v1/embeddings, /v1/audio*, /v1/images*, /v1/rerank #226` would auto-close: addressed by switching to `Refs Non-chat endpoints don't emit UsageEvents — cp-api's spend ledger misses spend from /v1/responses, /v1/embeddings, /v1/audio*, /v1/images*, /v1/rerank #226` (above).

Test plan

• Unit test: `emits_usage_event_on_200_with_prompt_tokens_issue_226` — drives a successful 200 with `prompt_tokens=42`, pins the wire shape (`prompt_tokens`, `completion_tokens=0`, `status_code=200`, `model_id`, `api_key_id`, `inbound_protocol="openai"`, non-empty `request_id` + `occurred_at`).
• Audit M1 regression: `emits_usage_event_on_200_with_zero_prompt_tokens_audit_m1` — drives a successful 200 with `prompt_tokens=0` and asserts the event still arrives, pinning the `upstream_called`-based gate.
• All 12 pre-existing `embeddings::tests` still pass — no regression in dispatch, input-shape preservation, or upstream error mapping.
• `cargo clippy -p aisix-proxy -- -D warnings` clean.

References

• Issue: Non-chat endpoints don't emit UsageEvents — cp-api's spend ledger misses spend from /v1/responses, /v1/embeddings, /v1/audio*, /v1/images*, /v1/rerank #226
• Follow-ups: UsageEvent emission missing on /v1/completions (#226 follow-up) #403 UsageEvent emission missing on /v1/responses (#226 follow-up) #404 UsageEvent emission missing on /v1/rerank (#226 follow-up) #405 UsageEvent emission missing on /v1/audio/* (#226 follow-up) #406 UsageEvent emission missing on /v1/images/* (#226 follow-up) #407
• Upstream contract: OpenAI Embeddings response shape — `usage.prompt_tokens` per https://platform.openai.com/docs/api-reference/embeddings/create
• Internal: `crates/aisix-proxy/src/chat.rs::emit_usage_event` (the convention this mirrors)
• cp-api receiving end: `dpmgr_usage_events` table; nullable columns mean older DP images / handlers without all fields still ingest cleanly.

[coderabbitai]

Warning

Review limit reached

@moonming, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 3 minutes and 16 seconds. Learn how PR review limits work.

Your organization has run out of usage credits. Purchase more in the billing tab.

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Free

Run ID: bf0d9617-d2ab-4f08-a962-9c9651507583

📥 Commits

Reviewing files that changed from the base of the PR and between 869f617 and 164c364.

📒 Files selected for processing (1)

• crates/aisix-proxy/src/embeddings.rs

📝 Walkthrough

Walkthrough

This PR enhances the embeddings endpoint to emit UsageEvent for successful requests. The handler now conditionally emits usage events when prompt_tokens > 0, backed by an internal dispatch refactor that captures and returns token counts. A new helper function builds and emits events to the usage sink and OTLP exporters, with test coverage validating the feature.

Changes

Embeddings usage event emission

Layer / File(s)	Summary
Handler and imports update crates/aisix-proxy/src/embeddings.rs	Updated embeddings handler to conditionally emit UsageEvent on successful requests when prompt_tokens > 0, using model, provider, api-key, status, latency, and resolved request identifiers. Added RequestOutcome to imports.
Dispatch structure and implementation crates/aisix-proxy/src/embeddings.rs	Introduced EmbedDispatchSuccess struct containing Response, provider label, model_id, and prompt_tokens. Updated dispatch function to return this struct instead of a tuple. Refactored "not supported" error path to return prompt_tokens: 0 with 501 envelope, suppressing usage emission.
Usage event emission helper crates/aisix-proxy/src/embeddings.rs	Added emit_usage_event helper that constructs a UsageEvent with embeddings-specific defaults and pushes it to the usage_sink and OTLP/HTTP exporter fan-out from the live snapshot, mirroring chat.rs conventions.
Regression test for usage event emission crates/aisix-proxy/src/embeddings.rs	Added test validating that a successful 200 embeddings request emits exactly one UsageEvent with expected prompt_tokens, zero completion_tokens, HTTP status code, api_key_id, model_id, inbound_protocol = "openai", and valid timestamps/request IDs.

---

🎯 3 (Moderate) | ⏱️ ~25 minutes

---

Note

🎁 Summarized by CodeRabbit Free

Your organization has reached its limit of developer seats under the Pro Plan. For new users, CodeRabbit will generate a high-level summary and a walkthrough for each pull request. For a comprehensive line-by-line review, please add seats to your subscription by visiting https://app.coderabbit.ai/login.If you believe this is a mistake and have available seats, please assign one to the pull request author through the subscription management page using the link above.

_{Comment @coderabbitai help to get the list of available commands and usage tips.}