Method to indicate that request has body

[zdm]

Is it possible to add method hasBody(), which will return true if request has body, in order to understand that we need to read it using onData.

[uasan]

Can check request.getHeader('content-length')

[zdm]

Data can be chunked, without known content length.

[uNetworkingAB]

hasBody is a good addition

[e3dio]

const hasBody = req => Number(req.getHeader('content-length')) || req.getHeader('transfer-encoding').includes('chunked');

[zdm]

yes, this will work
please, close, this issue, if you don;t want to add such method to the uws.

[uasan]

The client can send 0 bytes

!!+req.getHeader('content-length')

[e3dio]

@uasan yes '0' string needs to be properly accounted for, I fixed the function

[webcarrot]

// Several values can be listed, separated by a comma
Transfer-Encoding: gzip, chunked

req.getHeader('transfer-encoding')?.includes('chunked')

[zdm]

req.getHeader('transfer-encoding')?.toLowerCase().includes('chunked')

[uNetworkingAB]

Obviously this should be hasBody() as we already do standards compliant check for this internally. So it's just setting a boolean and returning it via hasBody()

[e3dio]

Usually you want to differentiate between standard request uploads with a 'content-length' and potentially unlimited stream of data with 'transfer-encoding: chunked', a simple hasBody would not help with that, encourages bad practice I think, maybe some people need it for something ?

const contentLength = Number(req.getHeader('content-length'));

const transferEncoding = req.getHeader('transfer-encoding');

if (contentLength) {

   const body = await getBody(res);

} else if (transferEncoding.includes('chunked')) {

   await processStream(res);

} else {

   console.log('no body');

}

res.end('done');

[zdm]

Fot me hasBody is enough. I already user methos, which is limit download size.

[uNetworkingAB]

bodyType() could return 0 for false, 1 for fix length, 2 for stream. The parser already knows all of this

[uasan]

bodyType() could return 0 for false, 1 for fix length, 2 for stream. The parser already knows all of this

With a fixed body length, it is very useful to know it in advance, then it will be better like this:

req.getBodyLength(): 0 | number | Infinity

[uNetworkingAB]

Agree, knowing the fixed length can help decide buffer size