Fixes#38456
## Problem
The Alpine package registry serves one `APKINDEX.tar.gz` per
architecture. When a repository contains only `noarch` packages (no
architecture-specific packages), only the `noarch` index is built.
Because `apk` substitutes `$ARCH` with the host architecture and
requests e.g. `x86_64/APKINDEX.tar.gz`, such a repository returned HTTP
404 and was unusable, matching the report in #38456.
## Fix
`GetRepositoryFile` now falls back to the `noarch` index when the
requested architecture has no index of its own, mirroring the fallback
already present in the sibling `DownloadPackageFile` handler. `noarch`
packages are installable on every architecture, so serving them for any
requested architecture is correct. The index-build side is unchanged;
only the serving path gains the fallback, so mixed repositories (which
already merge `noarch` into each per-architecture index) are unaffected.
## AI assistance disclosure
This change was implemented with the help of an AI coding assistant,
which gitea's CONTRIBUTING.md explicitly welcomes when disclosed. I have
reviewed the change, understand it, and can explain and defend it.
## Tests
Added a `NoArchOnly` subtest to `TestPackageAlpine` that publishes only
a `noarch` package to a fresh repository and asserts that `GET
.../x86_64/APKINDEX.tar.gz` now returns `200` (previously `404`) and
that the served index lists the noarch package. Verified locally with
`go build`/`go vet` on the changed package and a compile of the
integration test package (`go test -c`); the full integration run relies
on CI.
---------
Signed-off-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Addresses a batch of privately reported security issues, grouped by
area:
- **SSRF** - migration PR-patch/asset fetches, OAuth2 avatar & OpenID
discovery, pull-mirror URL re-validation, and the outbound proxy path.
- **Access-token scope** - prevent scope escalation on token creation;
keep public-only tokens confined (feeds, packages, Actions listings,
star/watch lists, limited/private owners).
- **Access control / disclosure** - go-get default-branch leak, webhook
authorization-header leak, watch clearing on private transitions,
label/attachment scoping.
- **Denial of service** - input bounds for npm dist-tags, Debian control
files, Arch file lists, and SSH keys.
### 📌 Attention for site admins
Not breaking - existing configs keep working - but two changes are worth
a look:
- **New SSRF protection** Outbound requests (migrations, OAuth2 avatars,
OpenID discovery, pull mirrors, proxy path) are now validated against
the allow/block host lists. If your instance legitimately reaches
internal hosts, you may need to add them to
`[security].ALLOWED_HOST_LIST` (and the relevant `ALLOW_LOCALNETWORKS`
settings).
- **Deprecation** `[webhook].ALLOWED_HOST_LIST` is deprecated and will
be removed in a future release. Use `[security].ALLOWED_HOST_LIST`
instead; the old key still works for now.
---------
Co-authored-by: TheFox0x7 <thefox0x7@gmail.com>
Co-authored-by: techknowlogick <techknowlogick@gitea.io>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: Zettat123 <zettat123@gmail.com>
**Newline injection into the Debian Release and Packages indices**
The `distribution` and `component` come straight from the request path
and are written line by line into the generated `Release` and `Packages`
files (the `Suite`/`Codename`/`Components` lines and the `Filename:
pool/<distribution>/<component>/...` line), but `UploadPackageFile` only
checked they were non-empty. `ctx.PathParam` url-decodes the segment, so
an encoded newline such as `main%0AInjected-Field: x` is accepted,
stored and then re-emitted for that distribution, which lets an
authenticated uploader forge extra fields in the index apt consumes.
Restricted both values to a conservative name pattern in the handler,
since that is the layer that accepts them; this should also keep the
pool paths well formed.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
### Description
Replaces all remaining direct `gopkg.in/yaml.v3` imports with
`go.yaml.in/yaml/v4` across models, modules, routers, services, and
integration tests. `gopkg.in/yaml.v3` moves from a direct to an indirect
dependency in `go.mod`.
#### API compatibility
The yaml.Node type, node.Kind/node.Content traversal style
(modules/markup/markdown/convertyaml.go), and the
UnmarshalYAML(*yaml.Node) interface signature
(modules/optional/serialization.go) are all preserved in v4 — no
call-site changes were required beyond the import path.
**Related:**
- https://github.com/go-gitea/gitea/pull/36564#issuecomment-4526536805
---------
Co-authored-by: silverwind <me@silverwind.io>
Co-authored-by: Claude (Opus 4.8) <noreply@anthropic.com>
Fixes#37086, fix the bug in MatchPath, and swap the order of
overlapping routes in api.go to make it look better.
---------
Signed-off-by: Rohan Guliani <rohansguliani@google.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Use shared repo permission resolution for Actions task users in issue
label remove and clear paths, and add a regression test for deleting
issue labels with a Gitea Actions token.
This fixes issue label deletion when the request is authenticated with a
Gitea Actions token.
Fixes#37011
The bug was that the delete path re-resolved repository permissions
using the normal user permission helper, which does not handle Actions
task users. As a result, `DELETE
/api/v1/repos/{owner}/{repo}/issues/{index}/labels/{id}` could return
`500` for Actions tokens even though label listing and label addition
worked.
---------
Co-authored-by: Codex <codex@openai.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: Giteabot <teabot@gitea.io>
Pass `ServeHeaderOptions` by value instead of pointer across all call
sites — no nil-check semantics are needed and the struct is small enough
that copying is fine.
## Changes
- **`services/context/base.go`**: `SetServeHeaders` and `ServeContent`
accept `ServeHeaderOptions` (value, not pointer); internal unsafe
pointer cast replaced with a clean type conversion
- **`routers/api/packages/helper/helper.go`**: `ServePackageFile`
variadic changed from `...*context.ServeHeaderOptions` to
`...context.ServeHeaderOptions`; internal variable is now a value type
- **All call sites** (13 files): `&context.ServeHeaderOptions{...}` →
`context.ServeHeaderOptions{...}`
Before/after at the definition level:
```go
// Before
func (b *Base) SetServeHeaders(opt *ServeHeaderOptions) { ... }
func (b *Base) ServeContent(r io.ReadSeeker, opts *ServeHeaderOptions) { ... }
func ServePackageFile(..., forceOpts ...*context.ServeHeaderOptions) { ... }
// After
func (b *Base) SetServeHeaders(opts ServeHeaderOptions) { ... }
func (b *Base) ServeContent(r io.ReadSeeker, opts ServeHeaderOptions) { ... }
func ServePackageFile(..., forceOpts ...context.ServeHeaderOptions) { ... }
```
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: wxiaoguang <2114189+wxiaoguang@users.noreply.github.com>
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Principles: let the caller decide what it needs, but not let the
framework (middleware) guess what it should do.
Then a lot of hacky code can be removed. And some FIXMEs can be fixed.
This PR introduces a new kind of middleware: "PreMiddleware", it will be
executed before all other middlewares on the same routing level, then a
route can declare its options for other middlewares.
By the way, allow the workflow badge to be accessed by Basic or OAuth2
auth.
Fixes: https://github.com/go-gitea/gitea/pull/36830
Fixes: https://github.com/go-gitea/gitea/issues/36859
I was testing typos-cli and fixed some misspelled wording here.
All changes are internal — no public API fields, database columns,
locale keys, or migration names are affected.
Fixes: https://github.com/go-gitea/gitea/issues/36152
Enable the `nilnil` linter while adding `//nolint` comments to existing
violations. This will ensure no new issues enter the code base while we
can fix existing issues gradually.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* Implicitly handle Actions Task Token for Nuget Api Keys
* Support same tokens as Basic Auth in Nuget Api Key Header
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Fixes#35159
Swift Package Manager expects an 'author.name' field in package
metadata, but Gitea was only providing schema.org format fields
(givenName, middleName, familyName). This caused SPM to fail with
keyNotFound error when fetching package metadata.
Changes:
- Add 'name' field to Person struct (inherited from
https://schema.org/Thing)
- Populate 'name' field in API response using existing String() method
- Maintains backward compatibility with existing schema.org fields
- Provides both formats for maximum compatibility
The fix ensures Swift Package Manager can successfully resolve packages
while preserving full schema.org compliance.
Fixes#21291, allowing icons and other missing attributes to appear for
NuGet packages from inside Visual Studio like they do with GitHub Nuget
packages.
Adds additional NuGet package information, particularly `IconURL`, to
bring the Gitea NuGet API more in-line with GitHub's NuGet API.
ref: https://learn.microsoft.com/en-us/nuget/api/search-query-service-resource
1. make our "route group pattern match" also update chi's RoutePattern
2. fix incorrect "NotFound" call in conda package
3. make ".flex-item .flex-item-main" has a general gap, then no need to
use `tw` tricks
4. improve the "test webhook" UI
1. Use `OpenXxx` instead of `GetXxx` because the returned readers should
be correctly closed, and clarify the behaviors of the functions: they
increase the download counter
2. Use `packages-content` styles instead of `issue-content`
1. Add some missing (optional) fields for nuget v2, and sort the fields
to make it easier to maintain
2. Add missing "platform" for rubygems: `VERSION-PLATFORM` and
`VERSION_PLATFORM`
Co-authored-by: Giteabot <teabot@gitea.io>
1. Using existing "content" variable in `swift.go`
2. Do not report 500 server error in `GetPullDiffStats` middleware,
otherwise a PR missing ref won't be able to view.
3. Fix the abused "label button" when listing commits, there was too
much padding space, see the screenshot below.
