Backport #38108
- Enforce repository token scope on RSS/Atom feed endpoints so a PAT
without repo scope can no longer read private repo commit data.
- Block HTTP redirects during repository migration clones to prevent
SSRF reaching internal addresses via an attacker-controlled redirect.
- Redact the notification subject after repo access is revoked so
private issue/PR metadata is no longer leaked through the notification
API.
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Backport #38112
## Problem
`GET /repos/{owner}/{repo}/times` and `GET
/repos/{owner}/{repo}/issues/{index}/times` crash with a nil pointer
dereference when the `user` query filter names a user that does not
exist.
## Root cause
In `ListTrackedTimes` and `ListTrackedTimesByRepository`, the
`IsErrUserNotExist` branch sends the 404 but is missing a `return`, so
execution falls through to `opts.UserID = user.ID` with a nil `user`.
---------
Co-authored-by: Pycub <iamsokhandan@gmail.com>
Backport #38010 by @bircni
`ifNeedApproval` in `services/actions/notifier_helper.go` decided
whether a
fork PR's workflow run had to wait for maintainer approval. The bypass
clause
counted any prior `approved_by > 0` run for `(repo_id,
trigger_user_id)`, so
the very first Approve-and-run click on a contributor's fork PR
permanently
trusted that user for every future fork PR in the same repository —
including
PRs whose only change is the workflow YAML itself.
Approving a workflow *run* is not the same as merging *code*. This
change
aligns the gate with GitHub Actions' first-time-contributor model: trust
is
granted only after the user has had a pull request merged in the repo.
## Behavior change
- **Before**: one approval = permanent trust for that user in that repo.
- **After**: every fork PR is gated until the contributor has at least
one
merged PR in the repo.
Existing already-approved runs and merged PRs continue to work; only the
trust criterion for *future* fork PRs changes. Maintainers who rely on
the
implicit "approve once" trust will see the approval banner reappear
until
they merge a PR from that contributor.
---------
Signed-off-by: bircni <bircni@icloud.com>
Co-authored-by: bircni <bircni@icloud.com>
Backport #37660 by @jorgeortiz85
## Summary
Fixes#37528
This PR makes the workflow dispatch API reject workflows that do not
declare `workflow_dispatch`. Previously, `POST
/repos/{owner}/{repo}/actions/workflows/{workflow_id}/dispatches` could
create an `ActionRun` for a workflow that only declared another event
such as `push`.
The service now validates that the target workflow has a
`workflow_dispatch` trigger before inserting the run. The API maps that
validation failure to `422 Unprocessable Entity`, matching existing
validation failures in this handler.
The regression test creates a push-only workflow, dispatches it through
the public API, asserts the `workflow_dispatch` validation message, and
verifies that no run was inserted.
## Testing
- `go test ./services/actions`
- `TAGS="sqlite sqlite_unlock_notify" make
test-integration#TestWorkflowDispatchPublicApiRequiresWorkflowDispatchTrigger`
- `TAGS="sqlite sqlite_unlock_notify" make
test-integration#TestWorkflowDispatchPublicApi`
## Disclosure
Developed with assistance from OpenAI Codex.
Co-authored-by: Jorge Ortiz <jorge.ortiz@gmail.com>
Co-authored-by: Nicolas <bircni@icloud.com>
backport #37118
This PR closes remaining `public-only` token gaps in the API by making
the restriction apply consistently across repository, organization,
activity, notification, and authenticated `/api/v1/user/...` routes.
Previously, `public-only` tokens were still able to:
- receive private results from some list/search/self endpoints,
- access repository data through ID-based lookups,
- and reach several authenticated self routes that should remain
unavailable for public-only access.
This change treats `public-only` as a cross-cutting visibility boundary:
- list/search endpoints now filter private resources consistently,
- repository lookups enforce the same restriction even when addressed
indirectly,
- and self routes that inherently expose or mutate private account state
now reject `public-only` tokens.
---
Generated by a coding agent with Codex 5.2
Co-authored-by: silverwind <me@silverwind.io>
Co-authored-by: Claude (Opus 4.7) <noreply@anthropic.com>
Co-authored-by: Nicolas <bircni@icloud.com>
Backport #37657 by @bircni
Fixes an issue where users could not commit changes on a file which is
unprotected.
Fixes#37655
Co-authored-by: Nicolas <bircni@icloud.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
Backport #37704
This PR hardens OAuth token exchange validation by binding exchanged
credentials to the client and redirect URI that originally obtained
them.
What it changes:
- reject refresh token exchanges when the refresh token belongs to a
different OAuth application
- reject authorization code exchanges when the `redirect_uri` in the
token request differs from the `redirect_uri` stored with the
authorization code
- add integration coverage for:
- authorization code exchange with a mismatched redirect URI
- refresh token reuse across two different dynamically created OAuth
applications
Why:
OAuth authorization codes and refresh tokens must remain bound to the
client context that originally received them. Without those checks:
- a valid authorization code can be redeemed against a different
registered redirect URI of the same client
- a refresh token can be replayed by a different OAuth client
---------
Co-authored-by: Nicolas <bircni@icloud.com>
Backport #37588 by @pandareen
## Summary
Fixes
[go-gitea/gitea#37564](https://github.com/go-gitea/gitea/issues/37564):
when an OIDC provider returns a `picture` claim, Gitea is supposed to
download that image as the user's avatar (if `[oauth2_client]
UPDATE_AVATAR = true`). Two latent bugs prevented this from working
consistently:
1. **Default Go User-Agent rejected by some image hosts.**
`oauth2UpdateAvatarIfNeed` used `http.Get`, which sends `User-Agent:
Go-http-client/1.1`. Hosts like `upload.wikimedia.org` reject that UA
with `403`, and every error path silently returned, so the user was left
with an identicon and **no log line** to diagnose the issue.
2. **Link-account *register* path skipped avatar sync.** First-time OIDC
sign-ins where auto-registration is disabled (or required a
username/password retype) go through `LinkAccountPostRegister`, which
created the user but never called `oauth2SignInSync`. So the avatar /
full name / SSH keys from the IdP were dropped on the floor for those
users, even though the existing-account-link path (`oauth2LinkAccount`)
and the auto-register path (`handleOAuth2SignIn`) both already did the
sync.
## Changes
- `routers/web/auth/oauth.go` — `oauth2UpdateAvatarIfNeed` now uses
`http.NewRequest` + `http.DefaultClient.Do`, sets `User-Agent: Gitea
<version>`, and logs every failure path at `Warn` (invalid URL, fetch
error, non-200, body read error, oversize body, upload error). No silent
failures.
- `routers/web/auth/linkaccount.go` — `LinkAccountPostRegister` now
calls `oauth2SignInSync` after a successful user creation, mirroring the
auto-register and link-existing-account flows.
- `tests/integration/oauth_avatar_test.go` — new
`TestOAuth2AvatarFromPicture` integration test with five sub-cases:
- `AutoRegister_FetchesAvatarFromPictureWithGiteaUA` — happy path,
asserts `use_custom_avatar=true`, an avatar hash is set, exactly one
HTTP request was made, and the request carried a `Gitea ` UA. The mock
server enforces the UA prefix to mirror real-world hosts that reject
Go's default UA.
- `AutoRegister_NonOK_DoesNotUpdateAvatar` — server returns 403; user's
avatar must remain unset.
- `AutoRegister_EmptyPicture_NoFetch` — empty `picture` claim must not
trigger any HTTP request.
- `AutoRegister_UpdateAvatarFalse_NoFetch` — `UPDATE_AVATAR=false` must
not trigger any HTTP request.
- `LinkAccountRegister_FetchesAvatarFromPicture` — guards the
`linkaccount.go` fix; without the new `oauth2SignInSync` call this
assertion fails.
## Test plan
- [x] `go test -tags 'sqlite sqlite_unlock_notify' -run
'^TestOAuth2AvatarFromPicture$' ./tests/integration/ -v` — 5/5 sub-tests
pass.
- [x] Manual: log in as a Keycloak user with `picture` claim pointing at
`https://avatars.githubusercontent.com/u/9919?v=4` — Gitea avatar is
replaced with the GitHub picture.
- [x] Manual: same flow with `https://upload.wikimedia.org/...` —
request now succeeds (or returns a clearly logged `Warn` line if
rate-limited with `429`); previously it silently 403'd.
- [x] Manual: `UPDATE_AVATAR=false` — user keeps the identicon, no
outbound request in container logs.
- [ ] Reviewer: please double-check that no other call sites of
`oauth2UpdateAvatarIfNeed` rely on the old `http.Get` behaviour.
## Related
- Upstream issue: go-gitea/gitea#37564
--------------------------------------------
AI Editor was used in this PR
---------
Signed-off-by: silverwind <me@silverwind.io>
Co-authored-by: pandareen <7270563+pandareen@users.noreply.github.com>
Co-authored-by: silverwind <me@silverwind.io>
Co-authored-by: Claude (Opus 4.7) <noreply@anthropic.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: Nicolas <bircni@icloud.com>
Backport #37695 by @lunny
This PR fixes two permission-checking gaps in Git and LFS request
handling.
## What it changes
- keep wiki Git HTTP pushes on the normal write-permission path, even
when proc-receive support is enabled
- revalidate LFS bearer token requests against the current user state
and current repository permissions before allowing access
- add regression coverage for unauthorized wiki HTTP pushes
- add LFS tests for blocked users, revoked repository access, read-only
upload attempts, and valid write access
## Why
- wiki repositories should not inherit the relaxed refs/for handling
used for normal code repositories
- LFS authorization tokens should not remain usable after a user is
disabled or loses repository access
---------
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #37631 by @silverwind
`UpdateLog` short-circuits on `len(Rows)==0` before honoring `NoMore`,
so a final empty `UpdateLog{NoMore:true}` never runs `TransferLogs`. The
task's `dbfs_data` rows are then never moved to log storage and never
deleted.
The bug has been latent since the original Actions implementation,
`act_runner` versions after
[runner#819](https://gitea.com/gitea/runner/pulls/819) trip it
deterministically.
Fix: let `NoMore=true` with no new rows fall through to `TransferLogs`.
Bail when the runner has outrun the server (`Index > ack`) even with
`NoMore`, since archiving a log with a gap is worse than retrying.
Always call `WriteLogs` so `offset==0` bootstraps an empty DBFS file in
the no-output case (otherwise `TransferLogs` would fail at `dbfs.Open`).
Fixes: https://github.com/go-gitea/gitea/issues/37623
Ref: [runner#952](https://gitea.com/gitea/runner/pulls/952)
Ref: [runner#950](https://gitea.com/gitea/runner/pulls/950)
---
This PR was written with the help of Claude Opus 4.7
Co-authored-by: silverwind <me@silverwind.io>
Co-authored-by: Claude (Opus 4.7) <noreply@anthropic.com>
Backport of #37662.
---
This PR was written with the help of Claude Opus 4.7
---------
Co-authored-by: Giteabot <teabot@gitea.io>
Co-authored-by: Claude (Opus 4.7) <noreply@anthropic.com>
## Summary
- handle compare requests where base and head refs have no common merge
base without returning 500
- keep the compare branch selectors usable and show a clear warning
message
- add regression coverage for unrelated-history compare selection and
merge-base error detection
Fixes#37469
Manuel Backport of: https://github.com/go-gitea/gitea/pull/37470
---------
Co-authored-by: Codex <codex@openai.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport of #37403 to `release/v1.26`.
The `events › logout propagation` e2e test was racing the SSE connection
setup: if page2's SharedWorker had not finished registering its
messenger by the time page1 triggered logout, the event was silently
dropped and page2 stayed on the authenticated page.
Wait 500ms after verifying page2 is signed in, before triggering the
logout from page1, so the SharedWorker has time to register. Comment
points at a cleaner future fix (expose a ready attribute on the page)
that will also work for the planned WebSocket SharedWorker.
---
This PR was written with the help of Claude Opus 4.7
Co-authored-by: Claude (Opus 4.7) <noreply@anthropic.com>
Backport #37288 by @KalashThakare
## Summary
Fixes#37252
The `/api/v1/repos/{owner}/{repo}/actions/runs` endpoint was returning
`event: "push"` for workflow runs triggered by `schedule:` (cron),
instead
of `event: "schedule"`.
## Root Cause
`ActionRun` has two separate fields:
- `Event` — the workflow registration event (e.g. `push`, set when the
workflow file was first pushed)
- `TriggerEvent` — the actual event that triggered the run (e.g.
`schedule`)
`ToActionWorkflowRun` in `services/convert/action.go` was serializing
`run.Event` into the API response instead of `run.TriggerEvent`, causing
scheduled runs to be indistinguishable from push events via the API.
This was already asymmetric — the tasks/jobs API correctly used
`TriggerEvent`.
## Fix
Changed `ToActionWorkflowRun` to use `run.TriggerEvent` for the `event`
field in the API response, consistent with how the jobs API works.
## Before
`event: "push"` returned for all scheduled runs:
<img width="1112" height="191" alt="Screenshot 2026-04-19 115642"
src="https://github.com/user-attachments/assets/c0a169f5-bbd9-4f5d-9474-e4c3795110e4"
/>
## After
`event: "schedule"` correctly returned for scheduled runs:
<img width="890" height="166" alt="Screenshot 2026-04-19 121723"
src="https://github.com/user-attachments/assets/860e99ac-0935-4a43-86a1-7b60f8113480"
/>
## Testing
- Added unit test `TestToActionWorkflowRun_UsesTriggerEvent` in
`services/convert/action_test.go` that explicitly verifies the API
returns `TriggerEvent` and not `Event` for a scheduled run.
- Manually verified via the API against a live Gitea instance with a
`cron: "* * * * *"` workflow.
Co-authored-by: Kalash Thakare ☯︎ <kalashthakare898@gmail.com>
Co-authored-by: Nicolas <bircni@icloud.com>
Backport #37327 by @prettysunflower
Nyallo~
In pull request #36901, a change is made so that the link to
authentication sources is now escaped with the QueryEscape filter.
https://github.com/go-gitea/gitea/pull/36901/changes#diff-34c39c9736a8b62e293c0c0b24c4b5b8c1c792790018c5809f9ff2cbc12b16b1R4
The problem is that [QueryEscape replace spaces with the `+`
character](https://cs.opensource.google/go/go/+/refs/tags/go1.26.2:src/net/url/url.go;l=234;drc=917949cc1d16c652cb09ba369718f45e5d814d8f),
and this is not unescaped when a user tries to log in with an
authentication source that contains a space, which throws an error.
This commit fixes that by unescaping the provider name in the URL.
---
Example of the error, on my instance, when I try to log in with
`prettysunflower's auth`
```
2026/04/21 00:11:41 routers/web/auth/oauth.go:42:SignInOAuth() [E] SignIn: oauth2 source not found, name: "prettysunflower's+auth"
/go/src/code.gitea.io/gitea/routers/web/auth/oauth.go:42 (0x2cfa5c5)
/usr/local/go/src/reflect/value.go:586 (0x51e245)
/usr/local/go/src/reflect/value.go:369 (0x51d0f8)
/go/src/code.gitea.io/gitea/modules/web/handler.go:181 (0x1a6aaf6)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:188 (0x1a6ab65)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:188 (0x1a6ab65)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:188 (0x1a6ab65)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/services/context/context.go:217 (0x2df1b23)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/gitea.com/go-chi/session@v0.0.0-20251124165456-68e0254e989e/session.go:258 (0x197eb82)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/chain.go:31 (0x1a61d05)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:479 (0x1a64fae)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:73 (0x1a628c2)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:321 (0x1a6421a)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/chain.go:31 (0x1a61d05)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:479 (0x1a64fae)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/middleware/get_head.go:37 (0x2c33a67)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:73 (0x1a628c2)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:321 (0x1a6421a)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/routers/common/maintenancemode.go:50 (0x2b752da)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/chain.go:31 (0x1a61d05)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:479 (0x1a64fae)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/routing/logger_manager.go:124 (0x127d1ec)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/chi-middleware/proxy@v1.1.1/middleware.go:37 (0x2b76acf)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/routers/common/middleware.go:89 (0x2b78cd6)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/routers/common/middleware.go:104 (0x2b7890f)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/src/code.gitea.io/gitea/modules/web/handler.go:145 (0x1a6afb5)
/usr/local/go/src/net/http/server.go:2286 (0x94dc88)
/go/pkg/mod/github.com/go-chi/chi/v5@v5.2.5/mux.go:90 (0x1a62881)
/go/src/code.gitea.io/gitea/modules/web/router.go:286 (0x1a6d2a2)
/go/src/code.gitea.io/gitea/modules/web/router.go:221 (0x1a6cbc6)
/usr/local/go/src/net/http/server.go:3311 (0x96e36d)
/usr/local/go/src/net/http/server.go:2073 (0x94bd6f)
/usr/local/go/src/runtime/asm_amd64.s:1771 (0x49af20)
```
Signed-off-by: prettysunflower <me@prettysunflower.moe>
Signed-off-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: prettysunflower <me@prettysunflower.moe>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: silverwind <me@silverwind.io>
Backport #37290 by wxiaoguang
Fix#37289
Don't tell container client that the instance needs basic auth if the
public access is available.
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #37256 by wxiaoguang
1. Make sure OmitEmail won't panic
2. SSH principal keys are not for signing or authentication
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #37185 by @Mohit25022005
Fix 500 error when comparing branches across fork repositories
## Problem
The compare API returns a 500 Internal Server Error when comparing
branches where the head commit exists only in the fork repository.
## Cause
The API was using the base repository's GitRepo and repository context
when converting commits. This fails when the commit does not exist in
the base repository, resulting in a "fatal: bad object" error.
## Solution
Use the head repository and HeadGitRepo when available to ensure commits
are resolved in the correct repository context.
## Result
* Fixes "fatal: bad object" error
* Enables proper comparison between base and fork repositories
* Prevents 500 Internal Server Error
Fixes#37168
Co-authored-by: Mohit Swarnkar <mohitswarnkar13@gmail.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #36085 by @eliroca
When authentication is handled externally by a reverse proxy SSO
provider, users can be redirected to an external logout URL or relative
path defined on the reverse proxy.
Co-authored-by: Elisei Roca <eroca@suse.de>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Backport #37130. Only one merge conflict in lockfile.
---
This PR was written with the help of Claude Opus 4.6
Co-authored-by: Claude (Opus 4.6) <noreply@anthropic.com>
Unties settings page from package version and adds button to delete the
package version
Settings page now allows for deletion of entire package and it's
versions as opposed to a single version
Adds an API endpoint to delete the entire package with all versions from
registry
fixes: https://github.com/go-gitea/gitea/issues/36904
Co-Authored-By: gemini-3-flash
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: Giteabot <teabot@gitea.io>
Keep `swagger` and `external-render-helper` as a standalone entries for
external render.
- Move `devtest.ts` to `modules/` as init functions
- Make external renders correctly load its helper JS and Gitea's current theme
- Make external render iframe inherit Gitea's iframe's background color to avoid flicker
- Add e2e tests for external render and OpenAPI iframe
---------
Co-authored-by: Claude (Opus 4.6) <noreply@anthropic.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Add a new e2e test for toggling issue reactions via the reaction picker
dropdown.
Add `aria-label` attributes to improve reaction accessibility:
- Add `aria-label="Reaction"` to the reaction picker dropdown
- Add `role="group"` with `aria-label="Reactions"` to the reactions
container, giving it a semantic identity for screen readers
- Include the reaction key in each reaction button's `aria-label` (e.g.
`+1: user1, user2`) so screen readers announce which reaction a button
represents
E2e test improvements:
- Simplify `randomString` to use `Math.random` instead of `node:crypto`
- Replace `generatePassword` with a static password, remove unused
`clickDropdownItem`
- Enable `fullyParallel: true` and `workers: '50%'` in Playwright config
- Run both chromium and firefox in all environments (not just CI)
- Parallelize `login` and `apiCreateRepo` setup where possible
- Use dedicated test user in `user-settings` test for concurrency safety
Co-authored-by: Claude (Opus 4.6) <noreply@anthropic.com>
Follow up #36842
Migration `326` can be prohibitively slow on large instances because it
scans and rewrites all commit status target URLs generated by Gitea
Actions in the database. This PR refactors migration `326` to perform a
partial update instead of rewriting every legacy target URL. The reason
for this partial rewrite is that **smaller legacy run/job indexes are
the most likely to be ambiguous with run/job ID-based URLs** during
runtime resolution, so this change prioritizes that subset while
avoiding the cost of rewriting all legacy records.
To preserve access to old links, this PR introduces
`resolveCurrentRunForView` to handle both ID-based URLs and index-based
URLs:
- For job pages (`/actions/runs/{run}/jobs/{job}`), it first tries to
confirm that the URL is ID-based. It does so by checking whether `{job}`
can be treated as an existing job ID in the repository and whether that
job belongs to `{run}`. If that match cannot be confirmed, it falls back
to treating the URL as legacy `run index + job index`, resolves the
corresponding run and job, and redirects to the correct ID-based URL.
- When both ID-based and index-based interpretations are valid at the
same time, the resolver **prefers the ID-based interpretation by
default**. For example, if a repository contains one run-job pair
(`run_id=3, run_index=2, job_id=4`), and also another run-job pair
(`run_id=1100, run_index=3, job_id=1200, job_index=4`), then
`/actions/runs/3/jobs/4` is ambiguous. In that case, the resolver treats
it as the ID-based URL by default and shows the page for `run_id=3,
job_id=4`. Users can still explicitly force the legacy index-based
interpretation with `?by_index=1`, which would resolve the same URL to
`/actions/runs/1100/jobs/1200`.
- For run summary pages (`/actions/runs/{run}`), it uses a best-effort
strategy: by default it first treats `{run}` as a run ID, and if no such
run exists in the repository, it falls back to treating `{run}` as a
legacy run index and redirects to the ID-based URL. Users can also
explicitly force the legacy interpretation with `?by_index=1`.
- This summary-page compatibility is best-effort, not a strict ambiguity
check. For example, if a repository contains two runs: runA (`id=7,
index=3`) and runB (`id=99, index=7`), then `/actions/runs/7` will
resolve to runA by default, even though the old index-based URL
originally referred to runB.
The table below shows how valid legacy index-based target URLs are
handled before and after migration `326`. Lower-range legacy URLs are
rewritten to ID-based URLs, while higher-range legacy URLs remain
unchanged in the database but are still handled correctly by
`resolveCurrentRunForView` at runtime.
| run_id | run_index | job_id | job_index | old target URL | updated by
migration 326 | current target URL | can be resolved correctly |
|---|---|---|---|---|---|---|---|
| 3 | 2 | 4 | 1 | `/user2/repo2/actions/runs/2/jobs/1` | true |
`/user2/repo2/actions/runs/3/jobs/4` | true |
| 4 | 3 | 8 | 4 | `/user2/repo2/actions/runs/3/jobs/4` | true |
`/user2/repo2/actions/runs/4/jobs/8` | true (without migration 326, this
URL will resolve to run(`id=3`)) |
| 80 | 20 | 170 | 0 | `/user2/repo2/actions/runs/20/jobs/0` | true |
`/user2/repo2/actions/runs/80/jobs/170` | true |
| 1500 | 900 | 1600 | 0 | `/user2/repo2/actions/runs/900/jobs/0` | false
| `/user2/repo2/actions/runs/900/jobs/0` | true |
| 2400 | 1500 | 2600 | 0 | `/user2/repo2/actions/runs/1500/jobs/0` |
false | `/user2/repo2/actions/runs/1500/jobs/0` | true |
| 2400 | 1500 | 2601 | 1 | `/user2/repo2/actions/runs/1500/jobs/1` |
false | `/user2/repo2/actions/runs/1500/jobs/1` | true |
For users who already ran the old migration `326`, this change has no
functional impact. Their historical URLs are already stored in the
ID-based form, and ID-based URLs continue to resolve correctly.
For users who have not run the old migration `326`, only a subset of
legacy target URLs will now be rewritten during upgrade. This avoids the
extreme runtime cost of the previous full migration, while all remaining
legacy target URLs continue to work through the web-layer compatibility
logic.
Many thanks to @wxiaoguang for the suggestions.
