David Parks

You have my vote on streaming. We house very large files on S3 and a common use case is to read small bits like headers from them. I do this...

I agree, I had a similar problem with TarFS previously. When walking a .tar file, in order to be strictly compliant, TarFS has to double scan the archive to guard...

I also have this issue, I am working with a shared bucket where creating extra meta information objects in the bucket would be an unfortunate complication, and one certainly not...

> Unfortunately specifying `--endpoint-url` like suggested by @skunkworker does not work for me confused I think you just need `https://ams3.digitaloceanspaces.com`, I guess you missed the `https://` part.

I also see this and many seemingly related issues very often. I came to git issues to look for related tiling reports like this. I'm seeing fewer than I expected....

I've been looking through the `smart_open` code today to find a solution to this problem. I notice that the range is set by default in `Reader.__init__()` where a `seek(0)` is...

Here's the effect on SeaweedFS, just to elucidate the effect this issue has. At about 04:00 I was running a continuous loop reading 100 bytes from a 4 GB file,...

So I only succeeded in solving the problem for a single read less than the buffer size (~132k by default). There's a more fundamental issue at play. `smart_open` is opening...

After sleeping on this here's the solution I propose implementing (I'm working towards this and a pull request). We have two use cases that are at odds here: **Use case...

This works as long as the first read request still sets the first byte range to exactly what is requested in that first read. This covers the use case of...