nanakura/moeka-project
1
0
Fork 0
Code Issues Pull Requests 1 Actions 1 Packages Projects Releases Wiki Activity

chore(server): disable audio endpoint

Browse Source
This commit is contained in:
RainbowBird
2026-03-28 02:25:44 +08:00
committed by RainbowBird
parent 4673494e73
commit 28f5c87fb2
1 changed files with 120 additions and 121 deletions
Download Patch File Download Diff File Expand all files Collapse all files
apps/server/src/routes/openai/v1/index.ts
+120 -121
View File
@@ -12,7 +12,6 @@ import type { HonoEnv } from '../../../types/hono'
import { useLogger } from '@guiiai/logg'
import { context, SpanStatusCode, trace } from '@opentelemetry/api'
import { Hono } from 'hono'
import { bodyLimit } from 'hono/body-limit'
import { authGuard } from '../../../middlewares/auth'
import { configGuard } from '../../../middlewares/config-guard'
@@ -315,148 +314,148 @@ export function createV1CompletionsRoutes(fluxService: FluxService, billingServi
return c.json(responseBody)
}
async function handleTTS(c: Context<HonoEnv>) {
const user = c.get('user')!
const flux = await fluxService.getFlux(user.id)
if (flux.flux <= 0) {
throw createPaymentRequiredError('Insufficient flux')
}
// async function handleTTS(c: Context<HonoEnv>) {
// const user = c.get('user')!
// const flux = await fluxService.getFlux(user.id)
// if (flux.flux <= 0) {
// throw createPaymentRequiredError('Insufficient flux')
// }
const body = await c.req.json()
const gatewayBaseUrl = await configKV.getOrThrow('GATEWAY_BASE_URL')
const baseUrl = normalizeBaseUrl(gatewayBaseUrl)
const serverAttributes = getServerConnectionAttributes(baseUrl)
const requestModel = body.model || 'auto'
// const body = await c.req.json()
// const gatewayBaseUrl = await configKV.getOrThrow('GATEWAY_BASE_URL')
// const baseUrl = normalizeBaseUrl(gatewayBaseUrl)
// const serverAttributes = getServerConnectionAttributes(baseUrl)
// const requestModel = body.model || 'auto'
const span = tracer.startSpan('llm.gateway.tts', {
attributes: {
[GEN_AI_ATTR_REQUEST_MODEL]: requestModel,
[AIRI_ATTR_GEN_AI_OPERATION_KIND]: 'text_to_speech',
...serverAttributes,
},
})
// const span = tracer.startSpan('llm.gateway.tts', {
// attributes: {
// [GEN_AI_ATTR_REQUEST_MODEL]: requestModel,
// [AIRI_ATTR_GEN_AI_OPERATION_KIND]: 'text_to_speech',
// ...serverAttributes,
// },
// })
const startedAt = Date.now()
// const startedAt = Date.now()
const response = await context.with(trace.setSpan(context.active(), span), () =>
fetch(`${baseUrl}audio/speech`, {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify(body),
}))
// const response = await context.with(trace.setSpan(context.active(), span), () =>
// fetch(`${baseUrl}audio/speech`, {
// method: 'POST',
// headers: { 'Content-Type': 'application/json' },
// body: JSON.stringify(body),
// }))
const durationMs = Date.now() - startedAt
span.setAttribute('http.response.status_code', response.status)
// const durationMs = Date.now() - startedAt
// span.setAttribute('http.response.status_code', response.status)
if (!response.ok) {
span.setStatus({ code: SpanStatusCode.ERROR, message: `Gateway ${response.status}` })
span.end()
recordMetrics({ model: requestModel, status: response.status, type: 'tts', durationMs, fluxConsumed: 0 })
return new Response(response.body, {
status: response.status,
headers: buildSafeResponseHeaders(response),
})
}
// if (!response.ok) {
// span.setStatus({ code: SpanStatusCode.ERROR, message: `Gateway ${response.status}` })
// span.end()
// recordMetrics({ model: requestModel, status: response.status, type: 'tts', durationMs, fluxConsumed: 0 })
// return new Response(response.body, {
// status: response.status,
// headers: buildSafeResponseHeaders(response),
// })
// }
const fluxPerRequest = await configKV.getOrThrow('FLUX_PER_REQUEST_TTS')
await billingService.consumeFluxForLLM({
userId: user.id,
amount: fluxPerRequest,
requestId: nanoid(),
description: `tts:${requestModel}`,
})
// const fluxPerRequest = await configKV.getOrThrow('FLUX_PER_REQUEST_TTS')
// await billingService.consumeFluxForLLM({
// userId: user.id,
// amount: fluxPerRequest,
// requestId: nanoid(),
// description: `tts:${requestModel}`,
// })
span.setAttribute(AIRI_ATTR_BILLING_FLUX_CONSUMED, fluxPerRequest)
span.end()
recordMetrics({ model: requestModel, status: response.status, type: 'tts', durationMs, fluxConsumed: fluxPerRequest })
// span.setAttribute(AIRI_ATTR_BILLING_FLUX_CONSUMED, fluxPerRequest)
// span.end()
// recordMetrics({ model: requestModel, status: response.status, type: 'tts', durationMs, fluxConsumed: fluxPerRequest })
publishRequestLog({
userId: user.id,
model: requestModel,
status: response.status,
durationMs,
fluxConsumed: fluxPerRequest,
})
// publishRequestLog({
// userId: user.id,
// model: requestModel,
// status: response.status,
// durationMs,
// fluxConsumed: fluxPerRequest,
// })
return new Response(response.body, {
status: response.status,
headers: buildSafeResponseHeaders(response),
})
}
// return new Response(response.body, {
// status: response.status,
// headers: buildSafeResponseHeaders(response),
// })
// }
async function handleTranscription(c: Context<HonoEnv>) {
const user = c.get('user')!
const flux = await fluxService.getFlux(user.id)
if (flux.flux <= 0) {
throw createPaymentRequiredError('Insufficient flux')
}
// async function handleTranscription(c: Context<HonoEnv>) {
// const user = c.get('user')!
// const flux = await fluxService.getFlux(user.id)
// if (flux.flux <= 0) {
// throw createPaymentRequiredError('Insufficient flux')
// }
const gatewayBaseUrl = await configKV.getOrThrow('GATEWAY_BASE_URL')
const baseUrl = normalizeBaseUrl(gatewayBaseUrl)
const serverAttributes = getServerConnectionAttributes(baseUrl)
// const gatewayBaseUrl = await configKV.getOrThrow('GATEWAY_BASE_URL')
// const baseUrl = normalizeBaseUrl(gatewayBaseUrl)
// const serverAttributes = getServerConnectionAttributes(baseUrl)
const span = tracer.startSpan('llm.gateway.asr', {
attributes: {
[GEN_AI_ATTR_REQUEST_MODEL]: 'auto',
[AIRI_ATTR_GEN_AI_OPERATION_KIND]: 'speech_to_text',
...serverAttributes,
},
})
// const span = tracer.startSpan('llm.gateway.asr', {
// attributes: {
// [GEN_AI_ATTR_REQUEST_MODEL]: 'auto',
// [AIRI_ATTR_GEN_AI_OPERATION_KIND]: 'speech_to_text',
// ...serverAttributes,
// },
// })
const startedAt = Date.now()
// const startedAt = Date.now()
const rawBody = await c.req.arrayBuffer()
const contentType = c.req.header('content-type') || 'multipart/form-data'
// const rawBody = await c.req.arrayBuffer()
// const contentType = c.req.header('content-type') || 'multipart/form-data'
const response = await context.with(trace.setSpan(context.active(), span), () =>
fetch(`${baseUrl}audio/transcriptions`, {
method: 'POST',
headers: { 'Content-Type': contentType },
body: rawBody,
}))
// const response = await context.with(trace.setSpan(context.active(), span), () =>
// fetch(`${baseUrl}audio/transcriptions`, {
// method: 'POST',
// headers: { 'Content-Type': contentType },
// body: rawBody,
// }))
const durationMs = Date.now() - startedAt
span.setAttribute('http.response.status_code', response.status)
// const durationMs = Date.now() - startedAt
// span.setAttribute('http.response.status_code', response.status)
if (!response.ok) {
span.setStatus({ code: SpanStatusCode.ERROR, message: `Gateway ${response.status}` })
span.end()
recordMetrics({ model: 'auto', status: response.status, type: 'asr', durationMs, fluxConsumed: 0 })
return new Response(response.body, {
status: response.status,
headers: buildSafeResponseHeaders(response),
})
}
// if (!response.ok) {
// span.setStatus({ code: SpanStatusCode.ERROR, message: `Gateway ${response.status}` })
// span.end()
// recordMetrics({ model: 'auto', status: response.status, type: 'asr', durationMs, fluxConsumed: 0 })
// return new Response(response.body, {
// status: response.status,
// headers: buildSafeResponseHeaders(response),
// })
// }
const fluxPerRequest = await configKV.getOrThrow('FLUX_PER_REQUEST_ASR')
await billingService.consumeFluxForLLM({
userId: user.id,
amount: fluxPerRequest,
requestId: nanoid(),
description: `asr:auto`,
})
// const fluxPerRequest = await configKV.getOrThrow('FLUX_PER_REQUEST_ASR')
// await billingService.consumeFluxForLLM({
// userId: user.id,
// amount: fluxPerRequest,
// requestId: nanoid(),
// description: `asr:auto`,
// })
span.setAttribute(AIRI_ATTR_BILLING_FLUX_CONSUMED, fluxPerRequest)
span.end()
recordMetrics({ model: 'auto', status: response.status, type: 'asr', durationMs, fluxConsumed: fluxPerRequest })
// span.setAttribute(AIRI_ATTR_BILLING_FLUX_CONSUMED, fluxPerRequest)
// span.end()
// recordMetrics({ model: 'auto', status: response.status, type: 'asr', durationMs, fluxConsumed: fluxPerRequest })
publishRequestLog({
userId: user.id,
model: 'auto',
status: response.status,
durationMs,
fluxConsumed: fluxPerRequest,
})
// publishRequestLog({
// userId: user.id,
// model: 'auto',
// status: response.status,
// durationMs,
// fluxConsumed: fluxPerRequest,
// })
return new Response(response.body, {
status: response.status,
headers: buildSafeResponseHeaders(response),
})
}
// return new Response(response.body, {
// status: response.status,
// headers: buildSafeResponseHeaders(response),
// })
// }
const chatGuard = configGuard(configKV, ['FLUX_PER_REQUEST', 'GATEWAY_BASE_URL', 'DEFAULT_CHAT_MODEL'], 'Service is not available yet')
const ttsGuard = configGuard(configKV, ['FLUX_PER_REQUEST_TTS', 'GATEWAY_BASE_URL'], 'TTS service is not available yet')
const asrGuard = configGuard(configKV, ['FLUX_PER_REQUEST_ASR', 'GATEWAY_BASE_URL'], 'ASR service is not available yet')
// const ttsGuard = configGuard(configKV, ['FLUX_PER_REQUEST_TTS', 'GATEWAY_BASE_URL'], 'TTS service is not available yet')
// const asrGuard = configGuard(configKV, ['FLUX_PER_REQUEST_ASR', 'GATEWAY_BASE_URL'], 'ASR service is not available yet')
// 60 requests per minute per user for LLM completions
const completionsRateLimit = rateLimiter({ max: 60, windowSec: 60 })
@@ -465,6 +464,6 @@ export function createV1CompletionsRoutes(fluxService: FluxService, billingServi
.use('*', authGuard)
.post('/chat/completions', completionsRateLimit, chatGuard, handleCompletion)
.post('/chat/completion', completionsRateLimit, chatGuard, handleCompletion)
.post('/audio/speech', ttsGuard, handleTTS)
.post('/audio/transcriptions', bodyLimit({ maxSize: 25 * 1024 * 1024 }), asrGuard, handleTranscription)
// .post('/audio/speech', ttsGuard, handleTTS)
// .post('/audio/transcriptions', bodyLimit({ maxSize: 25 * 1024 * 1024 }), asrGuard, handleTranscription)
}