From: Vsevolod Stakhov <vsevolod@rspamd.com>
Date: Mon, 13 Apr 2026 14:30:06 +0000 (+0100)
Subject: [Feature] selectors: add fuzzy_digest, fuzzy_shingles, authenticated, received_count... 
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=58bdc9f5385b5de5c79091dbf6b2ffc55d6fed7e;p=thirdparty%2Frspamd.git

[Feature] selectors: add fuzzy_digest, fuzzy_shingles, authenticated, received_count (#5981)

* [Feature] selectors: add fuzzy_digest, fuzzy_shingles, authenticated, received_count

* [Minor] selectors: deduplicate helpers and reuse cheaper task APIs

Extract the "find largest text part" loop into common.largest_text_part
shared by fuzzy_digest and fuzzy_shingles. Drop the pcall wrappers
around get_fuzzy_hashes since the C API does not throw. Use
get_header_count('Received') instead of allocating the full received
headers table just to take its length. Add selector smoke tests for
authenticated and received_count.

* [Minor] selectors: use lua_mime.get_displayed_text_part

Switch fuzzy_digest and fuzzy_shingles to lua_mime.get_displayed_text_part,
which already handles the right MUA-display semantics: prefer non-attachment
HTML, fall back to plain text, then to attachment HTML/text within size
limits, with a minimum-words threshold. Drops the largest_text_part helper
added in the previous commit since it picked the wrong part (e.g. a verbose
text/plain alternative or a .txt attachment).
---

diff --git a/lualib/lua_selectors/common.lua b/lualib/lua_selectors/common.lua
index c375dd34e3..2caa77a326 100644
--- a/lualib/lua_selectors/common.lua
+++ b/lualib/lua_selectors/common.lua
@@ -94,4 +94,4 @@ exports.create_raw_digest = create_raw_digest
 exports.get_cached_or_raw_digest = get_cached_or_raw_digest
 exports.encode_digest = encode_digest
 
-return exports
\ No newline at end of file
+return exports
diff --git a/lualib/lua_selectors/extractors.lua b/lualib/lua_selectors/extractors.lua
index efd37558c6..14810a8568 100644
--- a/lualib/lua_selectors/extractors.lua
+++ b/lualib/lua_selectors/extractors.lua
@@ -17,6 +17,7 @@ limitations under the License.
 local fun = require 'fun'
 local meta_functions = require "lua_meta"
 local lua_util = require "lua_util"
+local lua_mime = require "lua_mime"
 local rspamd_util = require "rspamd_util"
 local rspamd_url = require "rspamd_url"
 local common = require "lua_selectors/common"
@@ -593,6 +594,63 @@ The first argument must be header name.]],
     end,
     ['description'] = 'Get hostname of the filter server',
   },
+  -- Get strong fuzzy digest of the displayed text part
+  ['fuzzy_digest'] = {
+    ['get_value'] = function(task)
+      local best = lua_mime.get_displayed_text_part(task)
+      if not best then
+        return nil
+      end
+      local digest = best:get_fuzzy_hashes(task:get_mempool())
+      if not digest then
+        return nil
+      end
+      return digest, 'string'
+    end,
+    ['description'] = [[Get strong fuzzy digest (hex string) of the part an MUA would display
+(see lua_mime.get_displayed_text_part). Returns nil if the message has no usable text part.]],
+  },
+  -- Get fuzzy shingles of the displayed text part
+  ['fuzzy_shingles'] = {
+    ['get_value'] = function(task)
+      local best = lua_mime.get_displayed_text_part(task)
+      if not best then
+        return {}, 'string_list'
+      end
+      local _, shingles = best:get_fuzzy_hashes(task:get_mempool())
+      if type(shingles) ~= 'table' then
+        return {}, 'string_list'
+      end
+      local res = {}
+      for _, s in ipairs(shingles) do
+        if type(s) == 'table' and s[1] then
+          table.insert(res, tostring(s[1]))
+        end
+      end
+      return res, 'string_list'
+    end,
+    ['description'] = [[Get list of fuzzy shingle hashes (as strings) for the part an MUA would display
+(see lua_mime.get_displayed_text_part). Returns an empty list if no usable text part exists
+or shingles cannot be computed.]],
+  },
+  -- Check if the message was submitted by an authenticated user
+  ['authenticated'] = {
+    ['get_value'] = function(task)
+      if task:get_user() then
+        return 'true', 'string'
+      end
+      return 'false', 'string'
+    end,
+    ['description'] = [[Returns the string 'true' if the task has an authenticated user,
+otherwise 'false'. Useful as a cheap proxy for outbound/submission traffic.]],
+  },
+  -- Number of Received headers (hop count)
+  ['received_count'] = {
+    ['get_value'] = function(task)
+      return tostring(task:get_header_count('Received')), 'string'
+    end,
+    ['description'] = [[Get the number of Received headers as a string (hop count for the message).]],
+  },
 }
 
 return extractors
diff --git a/test/lua/unit/selectors.lua b/test/lua/unit/selectors.lua
index 437014fe2f..72876339bc 100644
--- a/test/lua/unit/selectors.lua
+++ b/test/lua/unit/selectors.lua
@@ -384,6 +384,15 @@ context("Selectors test", function()
       selector = "task_cache('cachevar2')",
       expect = {{"hello", "world"}}
     },
+
+    ["authenticated true"] = {
+      selector = "authenticated",
+      expect = {"true"}
+    },
+    ["received_count"] = {
+      selector = "received_count",
+      expect = {"2"}
+    },
   }
 
   for case_name, case in lua_util.spairs(cases_plain) do