[feature] Use ETag for robots.txt to prevent mishaps (#3829)

* [feature] Use ETag for robots.txt to prevent mishaps

* check incoming if-none-match header
This commit is contained in:
tobi 2025-02-24 11:17:18 +01:00 committed by GitHub
parent c9de6c9a1e
commit fd670c6a27
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 26 additions and 1 deletions

View file

@ -39,7 +39,7 @@ func (rb *Robots) Route(r *router.Router, m ...gin.HandlerFunc) {
// https://www.rfc-editor.org/rfc/rfc9309.html#section-2.4
robotsGroup.Use(
middleware.CacheControl(middleware.CacheControlConfig{
Directives: []string{"public", "max-age=86400"},
Directives: []string{"public", "no-cache"},
Vary: []string{"Accept-Encoding"},
}),
)

View file

@ -49,9 +49,29 @@ func (m *Module) Route(attachHandler func(method string, path string, f ...gin.H
}
func (m *Module) robotsGETHandler(c *gin.Context) {
const ETag = "\"" + apiutil.RobotsTxtETag + "\""
c.Header("ETag", ETag)
if c.Request.Header.Get("If-None-Match") == ETag {
// Cached.
c.AbortWithStatus(http.StatusNotModified)
return
}
// Not cached, serve.
c.String(http.StatusOK, apiutil.RobotsTxt)
}
func (m *Module) robotsGETHandlerDisallowNodeInfo(c *gin.Context) {
const ETag = "\"" + apiutil.RobotsTxtDisallowNodeInfoETag + "\""
c.Header("ETag", ETag)
if c.Request.Header.Get("If-None-Match") == ETag {
// Cached.
c.AbortWithStatus(http.StatusNotModified)
return
}
// Not cached, serve.
c.String(http.StatusOK, apiutil.RobotsTxtDisallowNodeInfo)
}

View file

@ -130,4 +130,9 @@ Disallow: /.well-known/webfinger
Disallow: /.well-known/nodeinfo
Disallow: /nodeinfo/
`
// MD5 hash of basic robots.txt.
RobotsTxtETag = `ce6729aacbb16fae3628210c04b462b7`
// MD5 hash of robots.txt with NodeInfo disallowed.
RobotsTxtDisallowNodeInfoETag = `a1e4ce6342978bc8d6c3e3dfab07cab4`
)