trentuna/candle-annotator
8
0
Fork 0
Code Issues Pull requests Releases Packages Activity Actions

feat(api): add Next.js proxy routes for ML inference service

Browse source
This commit is contained in:
Marko Djordjevic 2026-02-15 14:30:09 +01:00
parent 3a83fd38e9
commit 205021e810
5 changed files with 210 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

70
src/app/api/model/info/route.ts Normal file
View file

@ -0,0 +1,70 @@
import { NextRequest, NextResponse } from 'next/server';
const INFERENCE_API_URL = process.env.INFERENCE_API_URL || 'http://localhost:8001';
const INFERENCE_API_TIMEOUT = parseInt(process.env.INFERENCE_API_TIMEOUT || '10000', 10);
export async function GET(request: NextRequest) {
try {
// Forward request to Python inference service
const controller = new AbortController();
const timeoutId = setTimeout(() => controller.abort(), INFERENCE_API_TIMEOUT);
try {
const response = await fetch(`${INFERENCE_API_URL}/model/info`, {
method: 'GET',
headers: {
'Content-Type': 'application/json',
},
signal: controller.signal,
});
clearTimeout(timeoutId);
// Forward the response from the inference service
const data = await response.json();
if (!response.ok) {
// Handle 503 (no model available) specifically
if (response.status === 503) {
return NextResponse.json(
{ error: 'No model available' },
{ status: 503 }
);
}
return NextResponse.json(
{ error: data.detail || 'Failed to get model info' },
{ status: response.status }
);
}
return NextResponse.json(data);
} catch (fetchError: any) {
clearTimeout(timeoutId);
if (fetchError.name === 'AbortError') {
return NextResponse.json(
{ error: 'Model info request timed out' },
{ status: 504 }
);
}
throw fetchError;
}
} catch (error: any) {
console.error('Model info proxy error:', error);
// Check if it's a connection error
if (error.cause?.code === 'ECONNREFUSED' || error.message?.includes('fetch failed')) {
return NextResponse.json(
{ error: 'Inference service unavailable' },
{ status: 503 }
);
}
return NextResponse.json(
{ error: 'Internal server error' },
{ status: 500 }
);
}
}

65
src/app/api/predict/batch/route.ts Normal file
View file

@ -0,0 +1,65 @@
import { NextRequest, NextResponse } from 'next/server';
const INFERENCE_API_URL = process.env.INFERENCE_API_URL || 'http://localhost:8001';
const INFERENCE_BATCH_TIMEOUT = parseInt(process.env.INFERENCE_BATCH_TIMEOUT || '120000', 10);
export async function POST(request: NextRequest) {
try {
const body = await request.json();
// Forward request to Python inference service
const controller = new AbortController();
const timeoutId = setTimeout(() => controller.abort(), INFERENCE_BATCH_TIMEOUT);
try {
const response = await fetch(`${INFERENCE_API_URL}/predict/batch`, {
method: 'POST',
headers: {
'Content-Type': 'application/json',
},
body: JSON.stringify(body),
signal: controller.signal,
});
clearTimeout(timeoutId);
// Forward the response from the inference service
const data = await response.json();
if (!response.ok) {
return NextResponse.json(
{ error: data.detail || 'Batch prediction failed' },
{ status: response.status }
);
}
return NextResponse.json(data);
} catch (fetchError: any) {
clearTimeout(timeoutId);
if (fetchError.name === 'AbortError') {
return NextResponse.json(
{ error: 'Batch prediction timed out' },
{ status: 504 }
);
}
throw fetchError;
}
} catch (error: any) {
console.error('Batch predict proxy error:', error);
// Check if it's a connection error
if (error.cause?.code === 'ECONNREFUSED' || error.message?.includes('fetch failed')) {
return NextResponse.json(
{ error: 'Inference service unavailable' },
{ status: 503 }
);
}
return NextResponse.json(
{ error: 'Internal server error' },
{ status: 500 }
);
}
}

65
src/app/api/predict/route.ts Normal file
View file

@ -0,0 +1,65 @@
import { NextRequest, NextResponse } from 'next/server';
const INFERENCE_API_URL = process.env.INFERENCE_API_URL || 'http://localhost:8001';
const INFERENCE_API_TIMEOUT = parseInt(process.env.INFERENCE_API_TIMEOUT || '30000', 10);
export async function POST(request: NextRequest) {
try {
const body = await request.json();
// Forward request to Python inference service
const controller = new AbortController();
const timeoutId = setTimeout(() => controller.abort(), INFERENCE_API_TIMEOUT);
try {
const response = await fetch(`${INFERENCE_API_URL}/predict`, {
method: 'POST',
headers: {
'Content-Type': 'application/json',
},
body: JSON.stringify(body),
signal: controller.signal,
});
clearTimeout(timeoutId);
// Forward the response from the inference service
const data = await response.json();
if (!response.ok) {
return NextResponse.json(
{ error: data.detail || 'Prediction failed' },
{ status: response.status }
);
}
return NextResponse.json(data);
} catch (fetchError: any) {
clearTimeout(timeoutId);
if (fetchError.name === 'AbortError') {
return NextResponse.json(
{ error: 'Prediction request timed out' },
{ status: 504 }
);
}
throw fetchError;
}
} catch (error: any) {
console.error('Predict proxy error:', error);
// Check if it's a connection error
if (error.cause?.code === 'ECONNREFUSED' || error.message?.includes('fetch failed')) {
return NextResponse.json(
{ error: 'Inference service unavailable' },
{ status: 503 }
);
}
return NextResponse.json(
{ error: 'Internal server error' },
{ status: 500 }
);
}
}