amit-a / formidable-serverless Goto Github PK

In the docs it says it uses the same API as formidable. There's an easy and straightforward to extend the type definitions from formidable to formidable-serverless?

I was thinking in installing like this:

npm i --save formidable-serverless
i --save-dev @types/formidable

And somehow instruct Node to use formidable's types for formidable-serverless.

Hi, @Amit-A ,

AWS lambda is throwing:

> 2024-02-21T08:21:47.729Z	d74793a1-1596-4182-9ce5-6eb0370f8941	ERROR	Invoke Error 	
> {
>     "errorType": "TypeError",
>     "errorMessage": "req.on is not a function",
>     "stack": [
>         "TypeError: req.on is not a function",
>         "    at formidable.IncomingForm.parse (/var/task/node_modules/formidable-serverless/lib/index.js:39:7)",
>         "    at response.statusCode (/var/task/index.js:89:12)",
>         "    at new Promise (<anonymous>)",
>         "    at exports.handler (/var/task/index.js:88:12)"
>     ]
> }

I have a really plain lambda so I’m not sure what I’m missing.

My AWS lambda config:

Compatible architectures - Both:
• x86_64
• arm64

Compatible runtimes:
• nodejs 20.x

console.log('Node version is: ' + process.version);
console.log(process.versions);

Shows in CloudWatch:

Node version is: v20.11.0
{
node: '20.11.0',
acorn: '8.11.2',
ada: '2.7.4',
ares: '1.20.1',
base64: '0.5.1',
brotli: '1.0.9',
cjs_module_lexer: '1.2.2',
cldr: '43.1',
icu: '73.2',
llhttp: '8.1.1',
modules: '115',
napi: '9',
nghttp2: '1.58.0',
openssl: '3.1.4',
simdutf: '4.0.4',
tz: '2023c',
undici: '5.27.2',
unicode: '15.0',
uv: '1.46.0',
uvwasi: '0.0.19',
v8: '11.3.244.8-node.17',
zlib: '1.2.13.1-motley-5daffc7'
}

Here is my index.js:

console.log('Node version is: ' + process.version);
console.log(process.versions);

const isCustom = require('/opt/nodejs/myCustom');

const formidable = require('formidable-serverless');
const http = require('http');
const util = require('util');


const { S3Client } = require("@aws-sdk/client-s3");
const s3 = new S3Client({ region: "us-east-1" });


exports.handler = async(event) => {
    const res = await isCustom.handler(event);
    if (!res.isValid) { return res.failed; }



    const form = new formidable.IncomingForm();

    return new Promise((resolve, reject) => {
      form.parse(event, (err, fields, files) => {
        if(err) return reject(err);

        console.log(fields);  
        console.log(files);  
  
        resolve({status: 'ok'});
      });
    });
  


    return res.def; 
};

I have an easily reproducible setup:
• Windows 10 with VS Code.
• Just installed node 20.11.1 LTS from nodejs.org using the big button which downloads:
node-v20.11.1-x64.msi
• In my VS Code terminal:
npm install npm@latest -g
• So now
npm -v
Shows version: 10.4.0
node -v
Shows version: v20.11.1

1. Created my folder "c:\git\lmbd-202-api" and added in my "index.js" as shown above.
2. In VS Code: File, Open [that] folder.
3. In VS Code: Terminal, New terminal and type:
   npm init -y
4. Then type:
   npm install formidable-serverless

The output was:
npm WARN deprecated [email protected]: Please upgrade to latest, formidable@v2 or formidable@v3! Check these notes: https://bit.ly/2ZEqIau

Is there a npm command to just upgrade the parts of formidable that are needed (like your install only pulls 43k) or do I have to pull the entire 400k? Note, I still received the same error after downloading all of it.
npm install formidable@v3

5. Selected all the files in my "lmbd-202-api" folder, right mouse click: Send to: Compressed zip folder, and name it "lmbd-202-api.zip."
6. In the AWS Lambda console, on the “Code source” page, click Upload from, .zip file. Select the lmbd-202-api.zip. Note "lmbd-202-api" is also the exact name of my Lambda function.
7. In the AWS Lambda console, add a space after any semicolon so the Deploy button becomes active; Then click it.
8. Hitting the end point with a Postman PUT with multipart Form-Data that contains a 1k json file and two 50k images throws the 500 level error.

Note, my CloudFront origin is passing the request to my API Gateway which has Lambda Proxy Integration on and is in turn passing the request to my Lambda. Note, I logged the entire event to CloudWatch and the last line in the event is: isBase64Encoded: false.

Thank you

I like formidable-serverless, it really solved my pain with Firebase functions.
It is used as following:

const formidable = require("formidable-serverless");

function getBodyFields(req) {
  var form = new formidable.IncomingForm();
  return new Promise((resolve, reject) => {
    form.parse(req, (err, fields) => {
      if (err) reject(err);
      else if (Object.keys(fields).length === 0) reject(new Error("No data to upload."));
      else resolve(fields);
    });
  });
}

api.post(apiPath, async (req, res) => {
  try {
    const { base64, width, height, ext } = await getBodyFields(req);
  } catch (error) {
    console.error(error);
    res.status(500).send(error);
  }
});

Having sent an empty request without any field and file the unhandled error happened:

TypeError: this._parser.write is not a function
at IncomingForm.write (/functions/node_modules/formidable/lib/incoming_form.js:159:34)
at IncomingForm.formidable.IncomingForm.parse (/functions/node_modules/formidable-serverless/lib/index.js:58:10)
at Promise (/functions/api.js:114:10)
at new Promise ()
at getBodyFields (/functions/api.js:113:10)
at api.post (/functions/api.js:31:50)
at Layer.handle [as handle_request] (/functions/node_modules/express/lib/router/layer.js:95:5)
at next (/functions/node_modules/express/lib/router/route.js:137:13)
at Route.dispatch (/functions/node_modules/express/lib/router/route.js:112:3)
at Layer.handle [as handle_request] (/functions/node_modules/express/lib/router/layer.js:95:5)

"Unhandled" means that error was not cought with neither
if (err) reject(err);
nor
if (Object.keys(fields).length === 0) reject(new Error("No data to upload."));

It would be good to catch this error in callback function as err.

Thank you!

warning formidable-serverless > [email protected]: Please upgrade to latest, formidable@v2 or formidable@v3! Check these notes: https://bit.ly/2ZEqIau

I am trying to upload image to s3 bucket with aws-sdk and formidable-serverless, but I got error when uploading the image from frontend, here is my code in use:

const aws = require("aws-sdk")
const uuid = require("uuid")
import formidable from "formidable-serverless"

const Bucket = "bucket"

const s3 = new aws.S3({
	accessKeyId: "xxx",
	secretAccessKey: "xxx",
	region: "xxx",
	endpoint: "s3.xxx",
	signatureVersion: "v4",
})

export default async (req, res) => {
	const form = new formidable.IncomingForm()
	form.parse(req, (err, fields, files) => {
		const file = {
			Bucket: Bucket,
			Key: uuid.v4() + " - " + files.file.name,
			Body: files.file,
			ContentType: files.file.type,
		}
		console.log(file)
		s3.putObject(file, function (err, data) {
			if (err) console.log("Error: " + err)
			else res.send(data)
		})
	})
}

Here is the error I can't handle with:

NetworkingError [ERR_INVALID_ARG_TYPE]: The "chunk" argument must be of type string or an instance of Buffer. Received an instance of File

console.log(file) shows:

{
  Bucket: 'bucket',
  Key: 'd5f53b62-dbc3-4206-99c4-482970d0f16a - test.png',
  Body: File {
    _events: [Object: null prototype] {},
    _eventsCount: 0,
    _maxListeners: undefined,
    size: 89865,
    path: '/var/folders/7_/8fbg7my57lx3z46ccfwjddj80000gn/T/upload_3463316247a7d167eadba05a7c0d01b3',
    name: 'test.png',
    type: 'image/png',
    hash: null,
    lastModifiedDate: 2021-04-22T14:20:11.945Z,
    _writeStream: WriteStream {
      _writableState: [WritableState],
      _events: [Object: null prototype] {},
      _eventsCount: 0,
      _maxListeners: undefined,
      path: '/var/folders/7_/8fbg7my57lx3z46ccfwjddj80000gn/T/upload_3463316247a7d167eadba05a7c0d01b3',
      fd: 57,
      flags: 'w',
      mode: 438,
      start: undefined,
      autoClose: true,
      pos: undefined,
      bytesWritten: 89865,
      closed: false,
      [Symbol(kFs)]: [Object],
      [Symbol(kCapture)]: false,
      [Symbol(kIsPerformingIO)]: false
    },
    [Symbol(kCapture)]: false
  },
  ContentType: 'image/png'
}

When I set Body to Body: toString(files.file) then I can upload but the uploaded image can not be opened.