The code introduced by #18185 gets the error from response after it was processed by goth.
That is incorrect, as goth (and golang.org/x/oauth) doesn't really care about the error, and it sends a token request with an empty authorization code to the server anyway, which always results in a `oauth2: cannot fetch token: 400 Bad Request` error from goth.
It means that unless the "state" parameter is omitted from the error response (which is required to be present, according to [RFC 6749, Section 4.1.2.1](https://datatracker.ietf.org/doc/html/rfc6749#section-4.1.2.1)) or the page is reloaded (makes the session invalid), a 500 Internal Server Error page will be displayed.
This fixes it by handling the error before the request is passed to goth.
If the user-agent starts with git and user must change password but
hasn't return a 401 with the message.
It must be a 401, git doesn't seem to show the contents of the error message
when we return a 403
Fixes#19090
There appears to be a strange bug whereby the comment_id index can sometimes be missed
or missing from the action table despite the sync2 that should create it in the earlier
part of this migration. However, looking through the code for Sync2 there is no need
for this pre-code to exist and Sync2 should drop/create the indices as necessary.
I think therefore we should simplify the migration to simply be Sync2.
Signed-off-by: Andrew Thornton <art27@cantab.net>
* go.mod: add go-fed/{httpsig,activity/pub,activity/streams} dependency
go get github.com/go-fed/activity/streams@master
go get github.com/go-fed/activity/pub@master
go get github.com/go-fed/httpsig@master
* activitypub: implement /api/v1/activitypub/user/{username} (#14186)
Return informations regarding a Person (as defined in ActivityStreams
https://www.w3.org/TR/activitystreams-vocabulary/#dfn-person).
Refs: https://github.com/go-gitea/gitea/issues/14186
Signed-off-by: Loïc Dachary <loic@dachary.org>
* activitypub: add the public key to Person (#14186)
Refs: https://github.com/go-gitea/gitea/issues/14186
Signed-off-by: Loïc Dachary <loic@dachary.org>
* activitypub: go-fed conformant Clock instance
Signed-off-by: Loïc Dachary <loic@dachary.org>
* activitypub: signing http client
Signed-off-by: Loïc Dachary <loic@dachary.org>
* activitypub: implement the ReqSignature middleware
Signed-off-by: Loïc Dachary <loic@dachary.org>
* activitypub: hack_16834
Signed-off-by: Loïc Dachary <loic@dachary.org>
* Fix CI checks-backend errors with go mod tidy
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Change 2021 to 2022, properly format package imports
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Run make fmt and make generate-swagger
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Use Gitea JSON library, add assert for pkp
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Run make fmt again, fix err var redeclaration
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Remove LogSQL from ActivityPub person test
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Assert if json.Unmarshal succeeds
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Cleanup, handle invalid usernames for ActivityPub person GET request
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Rename hack_16834 to user_settings
Signed-off-by: Anthony Wang <ta180m@pm.me>
* Use the httplib module instead of http for GET requests
* Clean up whitespace with make fmt
* Use time.RFC1123 and make the http.Client proxy-aware
* Check if digest algo is supported in setting module
* Clean up some variable declarations
* Remove unneeded copy
* Use system timezone instead of setting.DefaultUILocation
* Use named constant for httpsigExpirationTime
* Make pubKey IRI #main-key instead of /#main-key
* Move /#main-key to #main-key in tests
* Implemented Webfinger endpoint.
* Add visible check.
* Add user profile as alias.
* Add actor IRI and remote interaction URL to WebFinger response
* fmt
* Fix lint errors
* Use go-ap instead of go-fed
* Run go mod tidy to fix missing modules in go.mod and go.sum
* make fmt
* Convert remaining code to go-ap
* Clean up go.sum
* Fix JSON unmarshall error
* Fix CI errors by adding @context to Person() and making sure types match
* Correctly decode JSON in api_activitypub_person_test.go
* Force CI rerun
* Fix TestActivityPubPersonInbox segfault
* Fix lint error
* Use @mariusor's suggestions for idiomatic go-ap usage
* Correctly add inbox/outbox IRIs to person
* Code cleanup
* Remove another LogSQL from ActivityPub person test
* Move httpsig algos slice to an init() function
* Add actor IRI and remote interaction URL to WebFinger response
* Update TestWebFinger to check for ActivityPub IRI in aliases
* make fmt
* Force CI rerun
* WebFinger: Add CORS header and fix Href -> Template for remote interactions
The CORS header is needed due to https://datatracker.ietf.org/doc/html/rfc7033#section-5 and fixes some Peertube <-> Gitea federation issues
* make lint-backend
* Make sure Person endpoint has Content-Type application/activity+json and includes PreferredUsername, URL, and Icon
Setting the correct Content-Type is essential for federating with Mastodon
* Use UTC instead of GMT
* Rename pkey to pubKey
* Make sure HTTP request Date in GMT
* make fmt
* dont drop err
* Make sure API responses always refer to username in original case
Copied from what I wrote on #19133 discussion: Handling username case is a very tricky issue and I've already encountered a Mastodon <-> Gitea federation bug due to Gitea considering Ta180m and ta180m to be the same user while Mastodon thinks they are two different users. I think the best way forward is for Gitea to only use the original case version of the username for federation so other AP software don't get confused.
* Move httpsig algs constant slice to modules/setting/federation.go
* Add new federation settings to app.example.ini and config-cheat-sheet
* Return if marshalling error
* Make sure Person IRIs are generated correctly
This commit ensures that if the setting.AppURL is something like "http://127.0.0.1:42567" (like in the integration tests), a trailing slash will be added after that URL.
* If httpsig verification fails, fix Host header and try again
This fixes a very rare bug when Gitea and another AP server (confirmed to happen with Mastodon) are running on the same machine, Gitea fails to verify incoming HTTP signatures. This is because the other AP server creates the sig with the public Gitea domain as the Host. However, when Gitea receives the request, the Host header is instead localhost, so the signature verification fails. Manually changing the host header to the correct value and trying the veification again fixes the bug.
* Revert "If httpsig verification fails, fix Host header and try again"
This reverts commit f53e46c721.
The bug was actually caused by nginx messing up the Host header when reverse-proxying since I didn't have the line `proxy_set_header Host $host;` in my nginx config for Gitea.
* Go back to using ap.IRI to generate inbox and outbox IRIs
* use const for key values
* Update routers/web/webfinger.go
* Use ctx.JSON in Person response to make code cleaner
* Revert "Use ctx.JSON in Person response to make code cleaner"
This doesn't work because the ctx.JSON() function already sends the response out and it's too late to edit the headers.
This reverts commit 95aad98897.
* Use activitypub.ActivityStreamsContentType for Person response Content Type
* Limit maximum ActivityPub request and response sizes to a configurable setting
* Move setting key constants to models/user/setting_keys.go
* Fix failing ActivityPubPerson integration test by checking the correct field for username
* Add a warning about changing settings that can break federation
* Add better comments
* Don't multiply Federation.MaxSize by 1<<20 twice
* Add more better comments
* Fix failing ActivityPubMissingPerson test
We now use ctx.ContextUser so the message printed out when a user does not exist is slightly different
* make generate-swagger
For some reason I didn't realize that /templates/swagger/v1_json.tmpl was machine-generated by make generate-swagger... I've been editing it by hand for three months! 🤦
* Move getting the RFC 2616 time to a separate function
* More code cleanup
* Update go-ap to fix empty liked collection and removed unneeded HTTP headers
* go mod tidy
* Add ed25519 to httpsig algorithms
* Use go-ap/jsonld to add @context and marshal JSON
* Change Gitea user agent from the default to Gitea/Version
* Use ctx.ServerError and remove all remote interaction code from webfinger.go
fgprof is a sampling Go profiler that allows you to analyze On-CPU as
well as Off-CPU (e.g. I/O) time together.
Go's builtin sampling CPU profiler can only show On-CPU time, but it's
better than fgprof at that. Go also includes tracing profilers that can
analyze I/O, but they can't be combined with the CPU profiler.
fgprof is designed for analyzing applications with mixed I/O and CPU
workloads. This kind of profiling is also known as wall-clock profiling.
Whilst fgprof can cause significant STW latencies in applications with a
lot of goroutines (> 1-10k), these latencies only occur if the profile
is requested - it doesn't cause a delay by simply being available.
The fgprof profile is mounted on
`http://localhost:6060/debug/fgprof?seconds=3`
Signed-off-by: Andrew Thornton <art27@cantab.net>
It appears possible that stopwatches can become orphaned or have been orphaned in
the past.
This PR adds Orphan checks for Stopwatches.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
* Empty log queue on flush and close
It is possible for log events to remain in the buffer off the multichannelledlog
and thus not be logged despite close or flush.
This PR simply adds a function to empty the queue before closing or flushing.
(Except when the logger is paused.)
Reference #19982
Signed-off-by: Andrew Thornton <art27@cantab.net>
* and do similar for ChannelledLog
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
If there are dangling stopwatches with missing issues there will be repeated
logging of Unable to APIFormat stopwatches. These are unhelpful and instead
we should only log if the error is not an issue not exist error.
And we should also prevent an error on missing issue in GetActiveStopwatch too
Signed-off-by: Andrew Thornton <art27@cantab.net>
gitea doctor --run check-db-consistency is currently broken due to an incorrect
and old use of Count() with a string.
Signed-off-by: Andrew Thornton <art27@cantab.net>
When migrating git repositories we should ensure that the commit-graph is written.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: delvh <dev.lh@web.de>
Add fetch.writeCommitGraph to gitconfig to ensure that a commit-graph will be written
on git fetch calls.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Allow git push to work when networked file systems with mixed
ownership are used with Gitea docker images >= 1.16.6 or Gitea
binaries running alongside git versions published after 04/2022.
There are circumstances independent of Gitea (networked file systems
with various permission systems) by which the git repositories managed
by Gitea may have mixed owners. It is not a behavior that Gitea have
control over nor is it a problem as long as the permissions for Gitea to
operate are correct. Gitea instances have been operating under these
conditions for a number of years.
It is detected as a potential security risk ( see
GHSA-vw2c-22j4-2fh2
) by the most recent git versions. However, Gitea always runs git
commands with a current directory matching the repository on
which it operates. That makes Gitea immune from this security problem
and it is safe to ignore the mixed owner permission check.
This gitconfig modification is done on a file dedicated to the user
exclusively used by Gitea.
Fixes: #19455
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: zeripath <art27@cantab.net>
* clean git support for ver < 2.0
* fine tune tests for markup (which requires git module)
* remove unnecessary comments
* try to fix tests
* try test again
* use const for GitVersionRequired instead of var
* try to fix integration test
* Refactor CheckAttributeReader to make a *git.Repository version
* update document for commit signing with Gitea's internal gitconfig
* update document for commit signing with Gitea's internal gitconfig
Co-authored-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
- Fomantic tries to prevent overflowing on the `y/x`-as by default on
stackable menu's on mobile screens. We already solve this issue by
forcing overflow on x as and hide it on y as(due to some issues with
other menu's), since https://github.com/go-gitea/gitea/pull/19486.
- However this edge case does require a y-overflow to show the dropdown,
because you cannot easily adjust this with CSS, once you're fiddling
with overflow's (https://stackoverflow.com/a/6433475). However
interesting behavior is noted
https://css-tricks.com/popping-hidden-overflow/ when you remove the
position: relative, it will suddenly work again. Well because this is
the only solution without redesigning dropdowns, I think we can live
with the side-effect of the dropdown items being full-width instead
"relative" width to their parent.
- Resolves#19976
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
- We don't need to buffer the logger with a thousand capacity. It's not
a high-throughput logger, this also caused issue whereby the logger
can't keep up with repeated messages being send(somehow they are lost in
the queue?).
- Resolves#19969
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
* Allow render HTML with css/js external links
* Fix bug because of filename escape chars
* Fix lint
* Update docs about new configuration item
* Fix bug of render HTML in sub directory
* Add CSP head for displaying iframe in rendering file
* Fix test
* Apply suggestions from code review
Co-authored-by: delvh <dev.lh@web.de>
* Some improvements
* some improvement
* revert change in SanitizerDisabled of external renderer
* Add sandbox for iframe and support allow-scripts and allow-same-origin
* refactor
* fix
* fix lint
* fine tune
* use single option RENDER_CONTENT_MODE, use sandbox=allow-scripts
* fine tune CSP
* Apply suggestions from code review
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: delvh <dev.lh@web.de>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
- Don't specify the field in `Count` instead use `Cols` for this.
- Call `log.Error` when a error occur.
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
* When non-admin users use code search, get code unit accessible repos in one main query
* Modified some comments to match the changes
* Removed unnecessary check for Access Mode in Collaboration table
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: Lauris BH <lauris@nix.lv>
- Always give a best-effort to fetching the repositories, if even that
fails indeed give a disconnected mirror found error.
- *Partially* resolves#19928
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
* Fix cli command restore-repo: "units" should be parsed as StringSlice because after #15790 it's read by c.StringSlice("units"). Before, the "units" were processed by strings.Split
* Add checking for invalid unit names
Co-authored-by: 6543 <6543@obermui.de>
