cmd/anubis: add rule hashes for admin-configured denials (#696)

* cmd/anubis: add rule hashes for admin-configured denials

Closes #695

Signed-off-by: Xe Iaso <me@xeiaso.net>

* cmd/anubis: remove theoretical nil pointer deference panic

This won't actually happen in real life, but the code paths might change so we should be somewhat defensive.

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

---------

Signed-off-by: Xe Iaso <me@xeiaso.net>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

5 files changed, 57 insertions, 7 deletions

diff --git a/cmd/anubis/index_templ.go b/cmd/anubis/index_templ.go
index 50d65ef..4b400d1 100644
--- a/cmd/anubis/index_templ.go
+++ b/cmd/anubis/index_templ.go
@@ -1,6 +1,6 @@
 // Code generated by templ - DO NOT EDIT.
 
-// templ: version: v0.3.819
+// templ: version: v0.3.833
 package main
 
 //lint:file-ignore SA4006 This context is only used if a nested component is present.
diff --git a/cmd/anubis/main.go b/cmd/anubis/main.go
index 69be826..de0a592 100644
--- a/cmd/anubis/main.go
+++ b/cmd/anubis/main.go
@@ -114,6 +114,21 @@ func main() {
 		log.Fatal(err)
 	}
 
+	fmt.Println("Rule error IDs:")
+	for _, rule := range s.policy.Bots {
+		if rule.Action != config.RuleDeny {
+			continue
+		}
+
+		hash, err := rule.Hash()
+		if err != nil {
+			log.Fatalf("can't calculate checksum of rule %s: %v", rule.Name, err)
+		}
+
+		fmt.Printf("* %s: %s\n", rule.Name, hash)
+	}
+	fmt.Println()
+
 	mux := http.NewServeMux()
 	xess.Mount(mux)
 
@@ -229,7 +244,7 @@ type Server struct {
 }
 
 func (s *Server) maybeReverseProxy(w http.ResponseWriter, r *http.Request) {
-	cr := s.check(r)
+	cr, rule := s.check(r)
 	r.Header.Add("X-Anubis-Rule", cr.Name)
 	r.Header.Add("X-Anubis-Action", string(cr.Rule))
 	lg := slog.With(
@@ -272,7 +287,19 @@ func (s *Server) maybeReverseProxy(w http.ResponseWriter, r *http.Request) {
 	case config.RuleDeny:
 		clearCookie(w)
 		lg.Info("explicit deny")
-		templ.Handler(base("Oh noes!", errorPage("Access Denied")), templ.WithStatus(http.StatusOK)).ServeHTTP(w, r)
+		if rule == nil {
+			lg.Error("rule is nil, cannot calculate checksum")
+			templ.Handler(base("Oh noes!", errorPage("Other internal server error (contact the admin)")), templ.WithStatus(http.StatusInternalServerError)).ServeHTTP(w, r)
+			return
+		}
+		hash, err := rule.Hash()
+		if err != nil {
+			lg.Error("can't calculate checksum of rule", "err", err)
+			templ.Handler(base("Oh noes!", errorPage("Other internal server error (contact the admin)")), templ.WithStatus(http.StatusInternalServerError)).ServeHTTP(w, r)
+			return
+		}
+		lg.Debug("rule hash", "hash", hash)
+		templ.Handler(base("Oh noes!", errorPage(fmt.Sprintf("Access Denied: error code %s", hash))), templ.WithStatus(http.StatusOK)).ServeHTTP(w, r)
 		return
 	case config.RuleChallenge:
 		lg.Debug("challenge requested")
diff --git a/cmd/anubis/policy.go b/cmd/anubis/policy.go
index 6e2241e..421bfff 100644
--- a/cmd/anubis/policy.go
+++ b/cmd/anubis/policy.go
@@ -35,6 +35,19 @@ type Bot struct {
 	Action    config.Rule `json:"action"`
 }
 
+func (b Bot) Hash() (string, error) {
+	var pathRex string
+	if b.Path != nil {
+		pathRex = b.Path.String()
+	}
+	var userAgentRex string
+	if b.UserAgent != nil {
+		userAgentRex = b.UserAgent.String()
+	}
+
+	return sha256sum(fmt.Sprintf("%s::%s::%s", b.Name, pathRex, userAgentRex))
+}
+
 func parseConfig(fin io.Reader, fname string) (*ParsedConfig, error) {
 	var c config.Config
 	if err := json.NewDecoder(fin).Decode(&c); err != nil {
@@ -110,20 +123,20 @@ func cr(name string, rule config.Rule) CheckResult {
 }
 
 // Check evaluates the list of rules, and returns the result
-func (s *Server) check(r *http.Request) CheckResult {
+func (s *Server) check(r *http.Request) (CheckResult, *Bot) {
 	for _, b := range s.policy.Bots {
 		if b.UserAgent != nil {
 			if b.UserAgent.MatchString(r.UserAgent()) {
-				return cr("bot/"+b.Name, b.Action)
+				return cr("bot/"+b.Name, b.Action), &b
 			}
 		}
 
 		if b.Path != nil {
 			if b.Path.MatchString(r.URL.Path) {
-				return cr("bot/"+b.Name, b.Action)
+				return cr("bot/"+b.Name, b.Action), &b
 			}
 		}
 	}
 
-	return cr("default/allow", config.RuleAllow)
+	return cr("default/allow", config.RuleAllow), nil
 }
diff --git a/cmd/anubis/static/js/main.mjs.gz b/cmd/anubis/static/js/main.mjs.gz
index 868c622..c104f77 100644
--- a/cmd/anubis/static/js/main.mjs.gz
+++ b/cmd/anubis/static/js/main.mjs.gz
diff --git a/cmd/anubis/testdata/everything_blocked.json b/cmd/anubis/testdata/everything_blocked.json
new file mode 100644
index 0000000..e1763e4
--- /dev/null
+++ b/cmd/anubis/testdata/everything_blocked.json
@@ -0,0 +1,10 @@
+{
+  "bots": [
+    {
+      "name": "everything",
+      "user_agent_regex": ".*",
+      "action": "DENY"
+    }
+  ],
+  "dnsbl": false
+}
\ No newline at end of file