Rework fuzzy check module.

- Now all checks are organized to rules.
- Allow to specify read_only rules to avoid problems on learning.
- Use better normalizer for fuzzy module and it now returns values
  from 0 to 1.0 (like bayes does).
- Update configuration accordingly.
- Drop legacy configuration support.
- Detect tanh as well and provide some reasonable (linear) fallback.

CMakeLists.txt

@@ -10,10 +10,10 @@ PROJECT(rspamd C)
 
 SET(RSPAMD_VERSION_MAJOR 0)
 SET(RSPAMD_VERSION_MINOR 6)
-SET(RSPAMD_VERSION_PATCH 1)
+SET(RSPAMD_VERSION_PATCH 2)
 
 SET(RSPAMD_VERSION         "${RSPAMD_VERSION_MAJOR}.${RSPAMD_VERSION_MINOR}.${RSPAMD_VERSION_PATCH}")
-SET(RSPAMD_MASTER_SITE_URL "http://bitbucket.org/vstakhov/rspamd")
+SET(RSPAMD_MASTER_SITE_URL "https://rspamd.com")
 
 IF(NOT RSPAMD_USER)
 	SET(RSPAMD_USER "nobody")
@@ -862,6 +862,7 @@ CHECK_FUNCTION_EXISTS(wait4 HAVE_WAIT4)
 CHECK_FUNCTION_EXISTS(waitpid HAVE_WAITPID)
 CHECK_FUNCTION_EXISTS(flock HAVE_FLOCK)
 CHECK_FUNCTION_EXISTS(tanhl HAVE_TANHL)
+CHECK_FUNCTION_EXISTS(tanh HAVE_TANH)
 CHECK_FUNCTION_EXISTS(expl HAVE_EXPL)
 CHECK_FUNCTION_EXISTS(exp2l HAVE_EXP2L)
 CHECK_FUNCTION_EXISTS(sendfile HAVE_SENDFILE)

conf/metrics.conf

@@ -408,22 +408,22 @@ metric {
         name = "BAYES_HAM";
     }
     symbol {
-        weight = 1.0;
+        weight = 10.0;
         description = "Generic fuzzy hash match";
         name = "R_FUZZY";
     }
     symbol {
-        weight = 1.0;
+        weight = 10.0;
         description = "Denied fuzzy hash";
         name = "FUZZY_DENIED";
     }
     symbol {
-        weight = 1.0;
+        weight = 5.0;
         description = "Probable fuzzy hash";
         name = "FUZZY_PROB";
     }
     symbol {
-        weight = 1.0;
+        weight = -2.1;
         description = "Whitelisted fuzzy hash";
         name = "FUZZY_WHITE";
     }

conf/modules.conf

@@ -1,24 +1,26 @@
 # Rspamd modules configuration
 fuzzy_check {
-    servers = "highsecure.ru:11335";
-    symbol = "R_FUZZY";
     min_bytes = 300;
-    max_score = 10;
-    mime_types = "application/pdf";
-    fuzzy_map = {
-        FUZZY_DENIED {
-            weight = 10.0;
-            flag = 1
-        }
-        FUZZY_PROB {
-            weight = 5.0;
-            flag = 2
-        }
-        FUZZY_WHITE {
-            weight = -2.1;
-            flag = 3
-        }
-    }
+	rule {
+		servers = "highsecure.ru:11335";
+		symbol = "R_FUZZY";
+		mime_types = "application/pdf";
+    	max_score = 10;
+		fuzzy_map = {
+			FUZZY_DENIED {
+				max_score = 10.0;
+				flag = 1
+			}
+			FUZZY_PROB {
+				max_score = 5.0;
+				flag = 2
+			}
+			FUZZY_WHITE {
+				max_score = 5.0;
+				flag = 3
+			}
+		}
+	}
 }
 forged_recipients {
     symbol_sender = "FORGED_SENDER";

config.h.in

@@ -145,6 +145,7 @@
 #cmakedefine HAVE_FLOCK          1
 
 #cmakedefine HAVE_TANHL          1
+#cmakedefine HAVE_TANH           1
 
 #cmakedefine HAVE_EXPL           1
 #cmakedefine HAVE_EXP2L          1

src/cfg_utils.c

@@ -635,12 +635,14 @@ internal_normalizer_func (struct config_file *cfg, long double score, void *data
     }
 #ifdef HAVE_TANHL
     return max * tanhl (score / max);
-#else
+#elif defined(HAVE_TANHL)
     /*
      * As some implementations of libm does not support tanhl, try to use
      * tanh
      */
     return max * tanh ((double) (score / max));
+#else
+    return score < max ? score / max : max;
 #endif
 }