When uploading a large file and an error occurs midway, can I continue to Layout use the already built merkle-dag?
In this case there will be blocks already on disk, so they will not need to be written again when re-adding the file. This probably makes the process slightly faster. Still, file needs to be chunked and hashed from the start.
Thank you for quick reply.I did not clearly describe my requirements.The real requirement is that when the client (browser) uploads a large file (>1G), it allows the local or server to resume uploading after losing the network connection. For example, if I encounter an error after uploading 500MB, I only need to upload the remaining 500MB. after that, can I use the already built Merkle-DAG by IPFS to Layout the remaining 500MB instead of all.If I want to do this, is there any idea？
No, there is no way to resume an upload out-of-the-box that I know of. You’d need some sort of middleware that does it an only adds to ipfs once the upload has fully completed.
It’s true, middleware is needed. thanks a lot.