[Feature]: Process for reading archives from Blobs

[AramZS]

Context

I'm very interested in reading a blob off a URL and reading an archive from that. I'm not sure the existing flow supports that? I wanted to check if it did before taking a whack at building that in as well. I'll add a sample blob URI shortly.

What change would you like to see?

As a user I'd like to put URLs into the archive that deliver Blobs.

Also how could we handle large archives that are contained in large blobs?

Requirements

1. Should be able to specify a Blob uri.
2. Have that loaded with the replay tool.
3. Have the replay tool present it to the user as normal.

Todo

No response

[ikreymer]

For clarity, do you mean loading from a blob: URL scheme that is created via some other method, eg.

1. On your site, you have a script that downloads a WACZ from some API and produces a blob: URL
2. Adds the replayweb.page embed <replay-web-page src="blob:..." ...> to the use the blob URL as the WACZ?

I don't think this works yet, but could be supported. The blob: URL are scoped to the origin, so you would have to load them on the same domain for them to work like this.

[AramZS]

Yeah, @ikreymer exactly that. I think it would be very useful for a number of use cases. Right now I'm looking at ATProto which stores files as Blobs and can produce URLs to them. But there are a variety of use cases I can imagine where a file might be blobbed in from local or into site memory where reading it out would be very useful if possible to do from a blob.