On Tue, Dec 26, 2017 at 2:45 AM, Norman Walsh <[email protected]> wrote:
> Brad Fitzpatrick <[email protected]> writes: > > Norman, implementing by mimicking can get you pretty far, but be > > sure to actually read the protocol docs as well. > > Absolutely. I am, however, finding the documentation difficult to > understand. > For example, this page > > https://perkeep.org/doc/protocol/blob-upload > > Seems to be incomplete. Or at least, it doesn’t seem to describe what > camput actually does. > In my investigation, camput appears to POST the content addressable > blob then a moment later POST a blob of JSON that is apparently the > metadata for that blob. However, the metadata post is not, in any way > that I can see, distinguished as a metadata blob. And despite being > clearly a junk of JSON, it’s posted with a content-type of > application/octet-stream, so it’s inappropriate for the server to > attempt to parse it. > A blob server just stores blobs. It should not try to interpret blobs. In a blob, there is no difference between data and metadata, and there are not even content types in blobs, hence application/octet-stream for everything. The blob-upload protocol docs don't get into that, because that all happens at layers above blob storage. When you said you waited to write a "backend", I assumed you meant a blob storage backend, but perhaps you meant something else. Which part do you want to write, and for what reasons? Maybe we can point you in the right direction. -- You received this message because you are subscribed to the Google Groups "Camlistore" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
