Refs: https://codeberg.org/forgejo/forgejo/issues/8
Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/153
Refs: https://codeberg.org/forgejo/forgejo/issues/123
Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/299
(cherry picked from commit 08dcef0c8c)
[DOCS] CONTRIBUTING/RELEASE: https://forgejo.org/docs/admin
Because the version is not displayed on the landing page of Forgejo,
there cannot be a link to a versionned documentation. There must exist
a link that points to the latest version on the website for the
forgejo instance to display.
Better but more complicated approaches could be to:
* Embed the documentation in Forgejo
* Allow the admin to not display the help
* Allow the admin to display a versionned help or not
(cherry picked from commit 83cc389239)
(cherry picked from commit 5df52b8a4f)
(cherry picked from commit 9a66b3d70b)
[DOCS] CONTRIBUTING/RELEASE: reminder to update FORGEJO_VERSION
(cherry picked from commit 2a4d0bd164)
(cherry picked from commit d68576ba67)
[DOCS] CONTRIBUTING: updates
* Remove obsolete description of the well being & moderation team and
replace them with a link to the moderation email.
* Remove description of the governance process and replace with a link
to the governance readme that did not exist at the time.
* Add links to the Forgejo documentation
(cherry picked from commit df749da272)
(cherry picked from commit 4da39128c5)
(cherry picked from commit f442ca6f40)
(cherry picked from commit 031928c447)
(cherry picked from commit 1e6ba47216)
(cherry picked from commit 8916a1f8c6)
(cherry picked from commit 663e28c412)
(cherry picked from commit d56b130baa)
[DOCS] CONTRIBUTING (squash)
(cherry picked from commit 0dc4b6e5c9)
(cherry picked from commit 52f2163807)
Although it would be possible to modify these files, it would create
conflicts when rebasing. Instead, this commit removes them entirely
and another commit can start from scratch, borrowing content from the
original files.
The drawback of this approach is that some content updates from Gitea
that also need updating in Forgejo will have to be copy/pasted
instead of being merged.
(cherry picked from commit eb85782115)
(cherry picked from commit 34401f2004)
(cherry picked from commit ef43b1c691)
(cherry picked from commit d17fe25e2f)
(cherry picked from commit 3f65dea3e7)
(cherry picked from commit 3cacb64a1b)
(cherry picked from commit b531b014b7)
(cherry picked from commit adf6e7a840)
(cherry picked from commit 68c19d977d)
(cherry picked from commit 31aa8647ec)
(cherry picked from commit be70dd7208)
(cherry picked from commit 7f44bb60b1)
(cherry picked from commit 5098f1224b)
(cherry picked from commit dc7a859144)
f92e0a4018 added minio to the tests/pgsql.ini.tmpl and Forgejo CI does
not run a minio server. It will if there are external storage related
changes at some point but it is not the case now so it is not worth
the burden.
(cherry picked from commit 4cfbf4718d)
(cherry picked from commit 9bd644d601)
(cherry picked from commit 604636c7c4)
(cherry picked from commit 341cdb8540)
(cherry picked from commit 57bdc91de5)
(cherry picked from commit c6ec6517b4)
(cherry picked from commit 2f03fc1b29)
(cherry picked from commit aaecd10966)
(cherry picked from commit 7025ff7b06)
(cherry picked from commit 15e6485ae1)
- Merge the file filters into `files-changed.yml`
- Remove unused yaml anchors like `&backend`
- Merge the `compliance-docs` workflow into `compliance`
- Add actions linting
- Misc cleanups for whitespace and step names
The admin config page has been broken for many many times, a little
refactoring would make this page panic.
So, add a test for it, and add another test to cover the 500 error page.
Co-authored-by: Giteabot <teabot@gitea.io>
Changes:
1. Use uniform links types relative to doc folder (start with `doc/`)
2. According to [docusaurus
links](https://docusaurus.io/docs/markdown-features/links), if `<a>` is
used, the `href` is resolved as URL location, but not file location. So
need to use `[text]({{< relref "path" >}})` instead.
This PR creates an API endpoint for creating/updating/deleting multiple
files in one API call similar to the solution provided by
[GitLab](https://docs.gitlab.com/ee/api/commits.html#create-a-commit-with-multiple-files-and-actions).
To archive this, the CreateOrUpdateRepoFile and DeleteRepoFIle functions
in files service are unified into one function supporting multiple files
and actions.
Resolves#14619
gitea.com experienced the corrupted LevelQueue bug again.
I think the problem is clear now: if the keys in LevelDB went
out-of-sync, the LevelQueue itself doesn't have the ability to recover,
eg:
* LevelQueue.Len() reports 100
* LevelQueue.LPop() reports ErrNotFound = errors.New("no key found")
So it needs to dive into the LevelDB to remove all keys to recover the
corrupted LevelQueue.
More comments are in TestCorruptedLevelQueue.
As discussed in
https://github.com/go-gitea/gitea/pull/24953#issuecomment-1565630156.
## ⚠️ BREAKING ⚠️
1. The `ui.THEME_COLOR_META_TAG` setting has been removed. If you still
need to set the `theme-color` meta tag, add it via
`$GITEA_CUSTOM/templates/custom/header.tmpl` instead.
2. The non-standard `default-theme` meta-tag added in
https://github.com/go-gitea/gitea/pull/13809 has been removed. Third
party code that needs to obtain the currently loaded theme should use
the `theme-<name>` class on the `<html>` node instead, which reflect the
currently active theme.
Replace the `reset` module with a modern version based on
[modern-normalize](https://github.com/sindresorhus/modern-normalize).
The only things I removed from that module are the `font-family` rules
we don't need. Otherwise, it's similar to Fomantic's reset, but with the
legacy IE stuff removed.
I documented every change done to the module.
Also this introduces a new `--tab-size` variable but it has no real
effect on code yet.
The old EventWriter's Run does:
```go
for {
handlePause()
select {
case event <- Queue:
write the log event ...
}
}
```
So, if an event writer is started before the logger is paused, there is
a chance that the logger isn't paused for the first message.
The new logic is:
```go
for {
select {
case event <- Queue:
handlePause()
write the log event ...
}
}
```
Then the event writer can be correctly paused
The old process manager's `Trace` function by default calls `log.Printf`
to output "trace" level logs. That's not ideal because by default the
trace level logs should not be outputted. In history it didn't cause
problems because there was no other call to the process manager before
the logger system's initialization.
But if there is any package using the process manager before the "Trace"
function gets assigned to the logger system's trace function, the
process manager will outputs unexpected verbose messages, this behavior
is not expected in most cases.
Now, the logger system also uses process manager to manage its goroutine
contexts, so it's the time to fix the old "trace" behavior: by default,
do not output the trace level messages. Fix#24951
The 500 page was not loading theme CSS, so always appeared in light
theme. This extracts the stylesheets into a shared template used by all
pages.
<img width="1262" alt="image"
src="https://github.com/go-gitea/gitea/assets/115237/72222b8d-a523-40d0-aa91-b9db32e5625d">
Small related rant: I think there should only be one `<head>` in all of
the templates, but I see it was deliberatly done that the 500 page has
its own `<head>` because "it should only depend the minimal template
functions", but I disagree because we are missing a lot of things that
are in the regular `<head>`.
Don't really know a better name for this. I've gone through some Forms
and added missing HTML attributes (mostly `maxlength`). I tried to fill
the Forms with dummy Data and see if Gitea throws a Error (e.g. maximum
length). If yes, I added the missing HTML attribute.
While working on this, I discovered that the Form to add OAuth2 Apps
just silently fails when filled with invalid data, so I fixed that too.
Before there was a "graceful function": RunWithShutdownFns, it's mainly
for some modules which doesn't support context.
The old queue system doesn't work well with context, so the old queues
need it.
After the queue refactoring, the new queue works with context well, so,
use Golang context as much as possible, the `RunWithShutdownFns` could
be removed (replaced by RunWithCancel for context cancel mechanism), the
related code could be simplified.
This PR also fixes some legacy queue-init problems, eg:
* typo : archiver: "unable to create codes indexer queue" => "unable to
create repo-archive queue"
* no nil check for failed queues, which causes unfriendly panic
After this PR, many goroutines could have better display name:
