Fixes#2325.
This introduces a way to download downsized versions of the user and repository avatars:
* `/avatars/123abcd` still serves the full-size avatar
* `/avatars/123abcd?size=64` serves it at size 64x64 px
Those downsized versions are computed on demand when requested for the first time and cached. The caching is done in a storage location configurable in the instance settings, just like the storage locations for the full-sized avatars are. The sizes of the downsized images are restricted to a fixed set of sizes, so that the cache doesn't grow too big. The caching and resizing logic is exposed in a way that could potentially be reused for other types of images (such as user uploads in issue discussions).
Luckily, the Go templates already specify in many places which size those avatars should be rendered, even if this information was only used for external avatar providers (such as Gravatar) until now.
The range of sizes requested by the HTML templates is rather wide: the table below lists all the sizes I could find, and the corresponding size served by the backend with the logic I implemented. The scaling factor of 2 was already used for requesting resized external avatars, and likely exists to make sure that users with display scaling enabled get a sharper picture.
| Size requested in the template | After scaling (x2) | Size of the image served |
|---------|---------|---------|
| 256 px | 512 px | original (512 px) |
| 140 px | 280 px | original (512 px) |
| 48 px | 96 px | 128 px |
| 40 px | 80 px | 128 px |
| 32 px | 64 px | 64 px |
| 28 px | 56 px | 64 px |
| 24 px | 48 px | 64 px |
| 20 px | 40 px | 64 px |
Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/11242
Reviewed-by: Gusted <gusted@noreply.codeberg.org>
This PR fixes a number of typos throughout the entire repository. Running https://github.com/crate-ci/typos and then changing all occurrences that I naively deemed "safe enough".
Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/10753
Reviewed-by: Gusted <gusted@noreply.codeberg.org>
Co-authored-by: Christoph Mewes <christoph@kubermatic.com>
Co-committed-by: Christoph Mewes <christoph@kubermatic.com>
- Remove options that currently aren't set
on `GarbageCollectLFSMetaObjectsOptions` and
`IterateLFSMetaObjectsForRepoOptions`.
- Simplify `IterateRepositoryIDsWithLFSMetaObjects` and
`IterateLFSMetaObjectsForRepo`.
- `IterateLFSMetaObjectsForRepo` was previously able to get in a
loop (`gc-lfs` doctor check was able to reproduce this) because the code
expected that the records would be updated to not match the SQL query,
but that wasn't the case. Simply enforce that only records higher than
the latest `id` from the previous iteration are allowed.
- For `gc-lfs` doctor check this was because `UpdatedLessRecentlyThan`
option was not set, which caused that records just marked as active in
the iteration weren't being filtered.
- Add unit tests
- Most likely a regression from 2cc3a6381c.
- The bug with `gc-lfs` was found on Codeberg.
