Partial Backport of #18415
Instead of using an asynchronous goroutine to push to disk on shutdown
just close the datachan and immediately push to the disk.
Prevents messages of incompletely flushed queues.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Backport #18823
Currently Gitea will wait for HammerTime or nice shutdown if kill -1 or kill -2
is sent. We should just immediately hammer if there is a second kill.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Backport #18820
There is a potential panic due to a mistaken resetting of the length parameter when
multibyte characters go over a read boundary.
Signed-off-by: Andrew Thornton <art27@cantab.net>
There is no need to call UpdateRepoStats in the InsertIssues and
InsertPullRequests function. They are only called during migration by
the CreateIssues and CreateReviews methods of the gitea uploader.
The UpdateRepoStats function will be called by the Finish method of
the gitea uploader after all reviews and issues are inserted. Calling
it before is therefore redundant and the associated SQL requests are
not cheap.
The statistics tests done after inserting an issue or a pull request
are also removed. They predate the implementation of UpdateRepoStats,
back when the calculation of the statistics was an integral part of
the migration function. The UpdateRepoStats is now tested
independantly and these tests are no longer necessary.
Signed-off-by: singuliere <singuliere@autistici.org>
Co-authored-by: zeripath <art27@cantab.net>
Co-authored-by: 6543 <6543@obermui.de>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #18770
v208.go is seriously broken as it misses an ID() check. We need to no-op and remigrate all of the u2f keys.
See #18756
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Various Mermaid improvments
- Render into iframe for improved security
- Use built-in dark theme instead of color inversion
- Remove flexbox attributes, resulting in more consistent size rendering
- Update API usage and update to latest version
* restart ci
* misc tweaks
* remove unneccesary declaration
* make it work without allow-same-origin, add loading=lazy
* remove loading attribute, does not seem to work
* rename variable
* skip roundtrip to DOM for rendering
* don't guess chart height
* update comment to make it clear it's intentional
* tweak
* replace deprecated 'scrolling' property
* remove unused css file
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Backport #18447
* Ensure git tag tests and other create test repos in tmpdir
There are a few places where tests appear to reuse testing repos which
causes random CI failures.
This PR simply changes these tests to ensure that cloning always happens
into new temporary directories.
Fix#18444
* Change log root for integration tests to use the REPO_TEST_DIR
There is a potential race in the drone integration tests whereby test-mysql etc
will start writing to log files causing make test-check fail.
Fix#18077
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: 6543 <6543@obermui.de>
* Increase the size of the webauthn_credential credential_id field (#18739)
Backport #18739
Unfortunately credentialIDs in u2f are 255 bytes long which with base32 encoding
becomes 408 bytes. The default size of a xorm string field is only a VARCHAR(255)
This problem is not apparent on SQLite because strings get mapped to TEXT there.
Fix#18727
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Ignore the migrate if u2f_registration is not exist (#18760)
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
* Prevent dangling GetAttribute calls
It appears possible that there could be a hang due to unread data from the
repo-attribute command pipes. This PR simply closes these during the defer.
Signed-off-by: Andrew Thornton <art27@cantab.net>
* move close into the defer
Signed-off-by: Andrew Thornton <art27@cantab.net>
* lets try again
Signed-off-by: Andrew Thornton <art27@cantab.net>
Backport #18707
Addresses #17892, where emails notifications are not sent to assignees (issue and PR) and reviewers (PR) when they have the email setting Only email on mention enabled.
From the user experience perspective, when a user gets a issue/PR assigned or a PR review request, he/she would expect to be implicitly mentioned since the assignment or request is personal and targeting a single person only. Thus I see #17892 as a bug. Could we therefore mark this ticket as such?
The changed code just explicitly checks for the EmailNotificationsOnMention setting beside the existing EmailNotificationsEnabled check. Too rude?
@lunny mentioned a mock mail server for tests, is there something ready. How could I make use of it?
See #12774 (comment)
Fix#17892
Co-authored-by: flozzone <flozzone@gmail.com>
Backport #18714
* Prevent double encoding of branch names in delete branch
There is a double encoding issue in branch template whereby the branch name
ends up double encoded.
Fix#18709
Signed-off-by: Andrew Thornton <art27@cantab.net>
* and tag name
Signed-off-by: Andrew Thornton <art27@cantab.net>
* And fix#18704
Signed-off-by: Andrew Thornton <art27@cantab.net>
Backport #18713
Move setting PullRequestWorkInProgressPrefixes to the start of PrepareViewPullInfo.
Fix#18706
Signed-off-by: Andrew Thornton <art27@cantab.net>
Backport #18729
When the issues repo unit is disabled, or an external issue tracker is used, there is no "a.ref-in-new-issue".
Fixes#18721
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #18708
Backport #18475
Backport #18443
The CI currently downloads all go modules in each pipeline step because go modules reside outside the project directory. Fix this by introducing a volume for the `/go` directory [1] so modules are only downloaded once per pipeline using a new `deps-backend` make target.
For completeness, I also included new `deps` and `deps-frontend` targets and the frontend one is also triggered explicitly on CI where needed.
[1] https://docs.drone.io/pipeline/kubernetes/examples/language/golang/#dependencies
* Also backports #18475 and #18443 so that is was able to merge cleanly.
Co-authored-by: 6543 <6543@obermui.de>
Co-authored-by: techknowlogick <techknowlogick@gitea.io>
When migrating a repository (from GitHub) using the API (**POST** `repos/migrate`), the Code Indexer is not updated. Searching in the user interface will not return any results.
When migrating the same repository using **+/New Migration** in the web interface, the search index is updated and searching works as expected.
Caused by the fact that object `repo` is never updated with the migrated repo so `setting.Indexer.RepoIndexerEnabled && !repo.IsEmpty` in `modules/notification/indexer/indexer.go:NotifyMigrateRepository` always evaluates to `false`.
Tested with gitea:1.16.1, MariaDB:10, Breve in `Run Mode: Dev`.
Co-authored-by: Hugo Hoitink <10838836+hoitih@users.noreply.github.com>
Backport #18678
WebAuthn may cause a security exception if the provided APP_ID is not allowed for the
current origin. Therefore we should reattempt authentication without the appid
extension.
Also we should allow [u2f] as-well as [U2F] sections.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Backport #18690
There is a problem with the current shimming script in that it will double quote the
provided GITEA_APP_INI due to a mistake in the bash. Here we change this to use a bash array.
Fix https://gitea.com/gitea/helm-chart/issues/287
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Restart zero worker if there is still work to do (#18658)
Backport #18658
It is possible for the zero worker to timeout before all the work is finished.
This may mean that work may take a long time to complete because a worker will only
be induced on repushing.
Also ensure that requested count is reset after pulls and push mirror sync requests and add some more trace logging to the queue push.
Fix#18607
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Update modules/queue/workerpool.go
Backport #18642
When a net.OpError occurs during rendering the underlying connection is essentially
dead and therefore attempting to render further data will only cause further errors.
Therefore in serverErrorInternal detect if the passed in error is an OpError and
if so do not attempt any further rendering.
Fix#18629
Signed-off-by: Andrew Thornton <art27@cantab.net>
Instead of the issue ID which is not a valid anchor.
Signed-off-by: singuliere <singuliere@autistici.org>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
* Only attempt to flush queue if the underlying worker pool is not finished (#18593)
Backport #18593
There is a possible race whereby a worker pool could be cancelled but yet the
underlying queue is not empty. This will lead to flush-all cycling because it
cannot empty the pool.
* On shutdown of Persistant Channel Queues close datachan and empty
Partial Backport #18415
Although we attempt to empty the datachan in queues - due to
races we are better off just closing the channel and forcibly emptying
it in shutdown.
Fix#18618
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Move zero workers warning to debug
Fix#18617
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Update modules/queue/manager.go
Co-authored-by: Gusted <williamzijl7@hotmail.com>
* Update modules/queue/manager.go
Co-authored-by: Gusted <williamzijl7@hotmail.com>
Co-authored-by: Gusted <williamzijl7@hotmail.com>
Backport #18587
It appears that the blob-excerpt links do not work on the wiki - likely since their
introduction.
This PR adds support for the wiki on these links.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Backport #18451
One of the repeated intermittent failures we see in testing is a failure due to
branches not being ready to merge.
Prior to the immediate queue implementation we would attempt to flush all the queues
and this would prevent the issue. However, the immediate queue is not flushable so
the flushall is not successful at preventing this.
This PR proposes an alternative solution - wait some time and try again up to 5 times.
If this fails then there is a genuine issue and we should fail.
Related #17719
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Ensure commit-statuses box is sized correctly in headers (#18538)
Backport #18538
Backport #18605
* Ensure commit-statuses box is sized correctly in headers
When viewing commits as commits the commit-status box will be fixed at 30px in height
due to being forced to be this size by a fomantic selector. This PR simply adds a
few more selectors to force this to have height auto.
Fix#18498
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
* Remove the spurious space in the .ui.right additional selector
Somehow a spurious space sneaked in to #18538
this PR simply removes it.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #18566
Gitea will currrently resend the same message-id for the closed/merged/reopened
messages for issues. This will cause the merged message to leap to the top of an
email chain and become out of sync.
This PR adds specific suffices for these actions.
Fix#18560
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Fix pushing to 1-x-dev docker tag
It appears that #18551 and #18573 have a mistake in that raymond does not have
an {{else}} on {{#equal}}. This PR notes that Sprig has a hasPrefix function
and so we use this with another if.
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Fix pushing to 1-x-dev docker tag (part 2)
Although we now have the manifest working, we need to create the images.
Here we adjust the .drone.yml to force building of the images
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Fix pushing to 1-x-dev docker tag
OK now we have the images building we should make sure that the main ones stays
dev and the release/v* ones become *-dev-*
Signed-off-by: Andrew Thornton <art27@cantab.net>
Backport #18539
There was an unintended regression in #17917 which leads to only
repository admin commits being trusted. This PR restores the old logic.
Fix#18501
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Backport #18562
There was an unfortunate regression in #17962 where following detection of the
UserProhibitLogin error the err is cast to a pointer by mistake.
This causes a panic due to an interface error.
Fix#18561
Signed-off-by: Andrew Thornton <art27@cantab.net>
Backport #18551
(Backporting this will enable this target to create 1.16-dev)
One of the problems with our current docker tagging is that although we
have strict version tags, latest and dev we do not have a way for docker
users to track the current release branch. This PR simply suggests that
we use the 1.x-dev tag for these and we build and push these. This will
give users who want or need unreleased bug fixes the option of tracking
the pre-release version instead of simply jumping to dev.
(Also contains backport for #18573)
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: KN4CK3R <admin@oldschoolhack.me>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Backport #18536
Unforunately git apply --3way reports conflicts differently than standard patches
resulting in conflicts being missed.
Adjust the conflict detection code to account for this different error reporting.
Fix#18514
Signed-off-by: Andrew Thornton <art27@cantab.net>
- Update all JS dependencies, including a security issue in mermaid
- Fix new linter errors related to value-keyword-case
- Tested Mermaid and Swagger
Co-authored-by: zeripath <art27@cantab.net>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: techknowlogick <techknowlogick@gitea.io>
