Open max-arnold opened 10 months ago
The use case is to detect whether a file is compressed (gzip, brotli) and serve it via a Cloud Function using appropriate http headers
Hi, it seems S3 ListObjects responses miss Content-Type for objects. ListObjects also missed UserMetadata. And GeeseFS tries to not perform HEAD requests for individual objects because it's really slow. Because of that, Yandex now has ListObjects list-type=ext-v1 which includes UserMetadata, but it seems it still misses ContentType :) so it should be implemented on the server first.
Am I correct that currently there is no way to access the Content-Type attribute even with an extra HEAD request through GeeseFS?
If yes, do you recommend to keep this issue open, or it is better to file a feature request against YC?
With an extra HEAD request it's possible but I don't really want to implement it this way
Ok, submitted a feature request to YC. Let's keep this issue open for a while
Is it possible to get content-encoding (and content-type) via xattrs?
Right now in Yandex Cloud Functions mounted bucket I only see
s3.etag
anduser.mtime
(as retrieved by Python'sos.listxattr
)