Running git update-index for every individual file is slow, so add and
remove everything with a single git command.
When such a big commit lands in the default branch, it could cause PR
creation and patch checking for all open PRs to be slow, or time out
entirely. For example, a commit that removes 1383 files was measured to
take more than 60 seconds and timed out. With this change checking took
about a second.
This is related to #27967, though this will not help with commits that
change many lines in few files.
(cherry picked from commit b88e5fc72d99e9d4a0aa9c13f70e0a9e967fe057)
- Remove a unused dependency. This dependency was added to handle YAML
'frontmatter' meta, parsing them and converting them to a table or
details in the resulting HTML. As can be read in the issue that reported
the behavior of YAML frontmatter being rendered literally,
https://github.com/go-gitea/gitea/issues/5377.
- It's an unused dependency as the codebase since then moved on to do this YAML
parsing and rendering on their own, this was implemented in
812cfd0ad9.
- Adds unit tests that was related to this functionality, to proof the
codebase already handles this and to prevent regressions.
- Don't make checkpoints or radio inputs full width on a small screen,
these obviously shouldn't try to take up the whole width of a container.
- Wrap the label for organisation permission box inside a `<span
class="inline field">`, so it gets a left-margin from the `.inline.field
> :first-child` selector. This make the checkboxes and radio buttons
groups look indented from the left.
- Resolves#4361
- It was noticed [in the Forgejo matrix channel](https://matrix.to/#/!qjPHwFPdxhpLkXMkyP:matrix.org/$vk78UR0eFCwQMDMTZ7-DWjMVB_LIAwHW6SkjhEcGkQQ?via=matrix.org) that the generation of the Forgejo project contributor stats was taking quite a while on codeberg.org. This was continued with the fact that a few moments later it was once again generating them again; it seemed like they weren't being cached while they were.
- The problem was that the cache TTL is hardcoded to ten minutes and not to the configured TTL. This patch changes that by using the configured TLL for the contributor cache, as this is a computationally heavy operation and should be cached for as long as possible for a good user experience. This doesn't impact the accuracy of this feature because the commit ID of the default branch is used as a cache key.
- Also changed in this patch, is that errors aren't cached and are instead being logged, this is more helpful to the administrator. For the user essentially nothing changed on this side, the contributor stats just looks like it's loading indefinitely.
- Realistically, testing this isn't possible, as the cache library Forgejo currently uses doesn't expose the TTL or expiration time of a key. Manually testing this behavior is quite lengthy, as one of the steps would need to be "wait for ten minutes" and describe how you can notice the data was cached or was just generated, and because you could use different types of cache, it will be quite hard to write down how you could check the TTL of a key for a particular cache (I'm not even sure it's even possible for some).
- Update the `github.com/buildkite/terminal-to-html/v3` dependency from
version v3.10.1 to v3.13.0.
- Version v3.12.0 introduced an incompatible change, the return type of
`AsHTML` changed from `[]byte` to `string`. That same version also
introduced streaming mode
https://github.com/buildkite/terminal-to-html/pull/126, which allows us
to avoid reading the whole input into memory.
- Closes#4313
This padding causes visual bug, because it only applies to the first line and not to the rest in case the line is broken into multiple.
I don't think it's actually needed here for good look, so I decided to remove it. In case decreased padding looks worse, the padding can be grown back via other element so it wouldn't cause this bug.
Preview: https://codeberg.org/attachments/56fd2ee8-4955-409d-998f-1feba987b9af
Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/4318
Reviewed-by: Gusted <gusted@noreply.codeberg.org>
A test must not fail depending on the performance of the machine it
runs on. It creates false negative and serves no purpose. These are
not benchmark tests for the hardware running them.
When a new go version is published, it takes about 24h for
https://github.com/actions/go-versions to be updated (see
https://github.com/actions/go-versions/pull/102 for example).
In the meantime the setup-go action that depends on it will install a
version of go that fails golang.org/x/vuln/cmd/govulncheck.
Move the security check to be the last step of the test job instead of
the first. It will still block the PRs from being merged but it will
allow the PR authors to keep working and look at the test results in
the meantime.
Fixes: https://codeberg.org/forgejo/forgejo/issues/4294
