From 84b57c53f4f6e9b3915d57b4a5c83570fea50dc3 Mon Sep 17 00:00:00 2001
From: Vsevolod Stakhov <vsevolod@highsecure.ru>
Date: Fri, 16 Nov 2018 17:16:25 +0000
Subject: [Minor] Allow to have bulk learn in the default config

---
 conf/statistic.conf                     | 32 ++++++++++++++++++--------------
 src/libstat/learn_cache/sqlite3_cache.c |  2 ++
 2 files changed, 20 insertions(+), 14 deletions(-)

diff --git a/conf/statistic.conf b/conf/statistic.conf
index 26e73c4d2..d9b9a1b72 100644
--- a/conf/statistic.conf
+++ b/conf/statistic.conf
@@ -43,22 +43,26 @@ classifier "bayes" {
   }
   learn_condition =<<EOD
 return function(task, is_spam, is_unlearn)
-  local prob = task:get_mempool():get_variable('bayes_prob', 'double')
+  local learn_type = task:get_request_header('Learn-Type')
 
-  if prob then
-    local in_class = false
-    local cl
-    if is_spam then
-      cl = 'spam'
-      in_class = prob >= 0.95
-    else
-      cl = 'ham'
-      in_class = prob <= 0.05
-    end
+  if not (learn_type and tostring(learn_type) == 'bulk') then
+    local prob = task:get_mempool():get_variable('bayes_prob', 'double')
+
+    if prob then
+      local in_class = false
+      local cl
+      if is_spam then
+        cl = 'spam'
+        in_class = prob >= 0.95
+      else
+        cl = 'ham'
+        in_class = prob <= 0.05
+      end
 
-    if in_class then
-      return false,string.format('already in class %s; probability %.2f%%',
-        cl, math.abs((prob - 0.5) * 200.0))
+      if in_class then
+        return false,string.format('already in class %s; probability %.2f%%',
+          cl, math.abs((prob - 0.5) * 200.0))
+      end
     end
   end
 
diff --git a/src/libstat/learn_cache/sqlite3_cache.c b/src/libstat/learn_cache/sqlite3_cache.c
index 255c835bb..52921326d 100644
--- a/src/libstat/learn_cache/sqlite3_cache.c
+++ b/src/libstat/learn_cache/sqlite3_cache.c
@@ -221,6 +221,8 @@ rspamd_stat_cache_sqlite3_check (struct rspamd_task *task,
 			/* We have some existing record in the table */
 			if (!!flag == !!is_spam) {
 				/* Already learned */
+				msg_warn_task ("already seen stat hash: %*bs",
+						rspamd_cryptobox_HASHBYTES, out);
 				return RSPAMD_LEARN_INGORE;
 			}
 			else {
-- 
cgit v1.2.3