Attempts to download a >1GB files fail

Steps to reproduce

• Attempt to download a large ~1GB file using the dropbox.filesDownload() method

Expected Result

• File gets downloaded fine.

Actual Result

• A V8 memory error is logged and the download fails:

Logs

/node_modules/dropbox/src/download-request.js:27
  res.on('data', function (chunk) { res.text += chunk; });
                          ^
RangeError: Invalid string length
    at IncomingMessage.<anonymous>  node_modules/dropbox/src/download-request.js:27:27)
    at emitOne (events.js:96:13)
    at IncomingMessage.emit (events.js:189:7)
    at readableAddChunk (_stream_readable.js:176:18)
    at IncomingMessage.Readable.push (_stream_readable.js:134:10)
    at HTTPParser.parserOnBody (_http_common.js:123:22)
    at TLSSocket.socketOnData (_http_client.js:411:20)
    at emitOne (events.js:96:13)
    at TLSSocket.emit (events.js:189:7)
    at readableAddChunk (_stream_readable.js:176:18)
    at TLSSocket.Readable.push (_stream_readable.js:134:10)
    at TLSWrap.onread (net.js:551:20)

Notes

AFAIK RangeError: Invalid string length is a V8 out-of-memory error. It basically complaints that a string’s length is too long. Perhaps res.on('data') response should accumulate on a Buffer instead to circumvent this?

[sidhuko]

@pran1990 duplicate of #101 and the PR #137 could update download-request.js and upload-request.js to return the Response object rather than buffered body. The Response object returns the stream (node-fetch) and has standardised methods to return different types of data, i.e. res.blob(), res.arrayBuffer() and res.json(). It also has properties for determining the response type (res.ok) and limiting duplicate consumption of the same stream (res.bodyUsed). Current methods in your SDK are very opposed to your company’s primary focus was on syncing large files between different operating systems…

Please read: https://developer.mozilla.org/en-US/docs/Web/API/Response

[BobDickinson]

I’m not sure I agree that this is a dupe of #101 (though certainly a streams-based solution should handle upload and download). I don’t have a problem with lack of stream support on upload, since I’m doing multipart upload on anything over 8Mb, and that doesn’t cause any heartburn. But not being able to download large files (due to lack of stream support) is kind of a deal killer. I also don’t see any support for doing a range request to get around this. It seems like the ability to download large files is pretty key and that not having that should be a pretty high priority fix.