| Age | Commit message (Collapse) | Author | Files |
|---|
|
|
|
This adds an optional S3 object info cache to the S3 storage driver backend (see [here](https://codeberg.org/gruf/go-storage/releases/tag/v0.4.0)) to reduce S3 calls largely during media cleanup operations, but it should also help in other situations cutting back on S3 calls when for example a key is already known to not exist.
Reviewed-on: https://codeberg.org/superseriousbusiness/gotosocial/pulls/4546
Co-authored-by: kim <grufwub@gmail.com>
Co-committed-by: kim <grufwub@gmail.com>
|
|
This is just a useful feature that it seemed more semantically correct to have in the storage library itself! Still, thank you to @vdyotte for the original change :)
Reviewed-on: https://codeberg.org/superseriousbusiness/gotosocial/pulls/4246
Co-authored-by: kim <grufwub@gmail.com>
Co-committed-by: kim <grufwub@gmail.com>
|
|
* update go-storage dependency, for S3Storage manually call PutObject() so we can set content-type
* update calls to PutFile() to include the contentType
|
|
available (#2924)
* update to use go-storage/ instead of go-store/v2/storage/
* pull in latest version from codeberg
* remove test output :innocent:
* add code comments
* set the exclusive bit when creating new files in disk config
* bump to actual release version
* bump to v0.1.1 (tis a simple no-logic change)
* update readme
* only use a temporary read seeker when decoding video if required (should only be S3 now)
* use fastcopy library to use memory pooled buffers when calling TempFileSeeker()
* update to use seek call in serveFileRange()
|
Terin Stock
kim
kim