abstract-blob-store
abstract-blob-store copied to clipboard
Resumption of uploads
Resumption is likely to become a necessity for very large blobs in many use cases. To start off discussion, what about:
store.resumeWriteStream(resume_opts, opts, cb)
Where resume_opts are the options specifically needed to identify what and how to resume a write?
I could see something similar for resuming from a read stream, though maybe range requests are the right way to handle that, I'm not sure.
for inspiration: https://cloud.google.com/storage/docs/concepts-techniques#resumable
and for reference, I implemented this in s3-upload-stream: https://github.com/nathanpeck/s3-upload-stream/pull/25