jbilcke-hf's picture
jbilcke-hf HF staff
oh yeah
bb04353
import { Blob } from "node:buffer"
import express from "express"
import queryString from "query-string"
import { parseClap, ClapProject } from "@aitube/clap"
import { clapToTmpVideoFilePath } from "./main"
// import { defaultExportFormat, type SupportedExportFormat } from "@aitube/ffmpeg"
import { defaultExportFormat, type SupportedExportFormat } from "./bug-in-bun/aitube_ffmpeg"
import { deleteFile } from "@aitube/io"
const app = express()
const port = 3000
process.on('unhandledRejection', (reason: string, p: Promise<any>) => {
console.error('Unhandled Rejection at:', p, 'reason:', reason);
})
process.on('uncaughtException', (error: Error) => {
console.error(`Caught exception: ${error}\n` + `Exception origin: ${error.stack}`);
})
// fix this error: "PayloadTooLargeError: request entity too large"
// there are multiple version because.. yeah well, it's Express!
// app.use(bodyParser.json({limit: '50mb'}));
//app.use(bodyParser.urlencoded({limit: '50mb', extended: true}));
app.use(express.json({ limit: '200mb' }));
app.use(express.urlencoded({ limit: '200mb', extended: true }));
app.get("/", async (req, res) => {
res.status(200)
const documentation = `<html>
<head></head>
<body style="display: flex;
align-items: center;
justify-content: center;
background-color: #000000;
opacity: 1;
background-image: repeating-radial-gradient( circle at 0 0, transparent 0, #000000 7px ), repeating-linear-gradient( #34353655, #343536 );
">
<div style="">
<p style="">
<h1 style="
color: rgba(255,255,255,0.9);
font-size: 4.5vw;
text-shadow: #000 1px 0 3px;
font-family: Helvetica Neue, Helvetica, sans-serif;
font-weight: 100;
">Clap Exporter <span style="font-weight: 400">API</span></h1>
<pre style="color: rgba(255,255,255,0.7); font-size: 2vw; text-shadow: #000 1px 0 3px; font-family: monospace;">
$ curl -o movie.mp4 \\
-X POST \\
--data-binary @path/to/movie.clap \\
https://jbilcke-hf-ai-tube-clap-exporter.hf.space?f=mp4
</pre>
<br/>
</p>
</div>
</body>
<html>`
res.write(documentation)
res.end()
})
// the export robot has only one job: to export .clap files
app.post("/", async (req, res) => {
console.log("receiving POST request")
const qs = queryString.parseUrl(req.url || "")
const query = (qs || {}).query
let format: SupportedExportFormat = defaultExportFormat
try {
format = decodeURIComponent(query?.f?.toString() || defaultExportFormat).trim() as SupportedExportFormat
if (format !== "mp4" && format !== "webm") {
format = defaultExportFormat
}
} catch (err) {}
let data: Uint8Array[] = [];
req.on("data", (chunk) => {
data.push(chunk);
});
req.on("end", async () => {
try {
let fileData = Buffer.concat(data)
const clap: ClapProject = await parseClap(new Blob([fileData]));
// not! that is too large!!!
console.log("got a clap project:", clap?.meta?.description)
const {
tmpWorkDir,
outputFilePath,
} = await clapToTmpVideoFilePath({ clap, format })
console.log(`got an output ${format} file at:`, outputFilePath)
res.download(outputFilePath, async () => {
// clean-up after ourselves (we clear the whole tmp directory)
await deleteFile(tmpWorkDir)
// console.log("cleared the temporary folder")
})
return
} catch (err) {
console.log(`failed to process the request\n${err}`)
res.status(500)
res.write(JSON.stringify({ "error": `${err}` }))
res.end()
return
}
});
})
app.listen(port, () => {
console.log(`Open http://localhost:${port}`)
})