Add New Filtering Methods - 2.17 Beta 1

Adds new functions that can do multi-variable filtering to better detect more types of spam

Scripts/operations.py

@@ -840,6 +840,9 @@ def check_against_filter(current, filtersDict, miscData, config, currentCommentD
       languages = smartFilter['languages']
       sensitive =  smartFilter['sensitive']
       rootDomainRegex = smartFilter['rootDomainRegex']
+      accompanyingLinkSpamDict = smartFilter['accompanyingLinkSpamDict']
+      comboDict = smartFilter['comboDict']
+
       # Spam Lists
       spamListCombinedRegex = smartFilter['spamListCombinedRegex']
 
@@ -889,6 +892,27 @@ def check_if_only_link(string):
         else:
           return False
 
+      def find_accompanying_link_spam(string):
+        linkResult = re.search(accompanyingLinkSpamDict['videoLinkRegex'], string)
+        if not linkResult:
+          return False
+        else:
+          phrasesList = accompanyingLinkSpamDict['accompanyingLinkSpamPhrasesList']
+          notSpecialChars = accompanyingLinkSpamDict['notSpecial']
+          nonLinkString = string.replace(linkResult.group(0), '')
+          for char in notSpecialChars:
+            nonLinkString = nonLinkString.replace(char, '').replace('\n', '')
+          if any(phrase.lower().replace(' ', '') == nonLinkString for phrase in phrasesList):
+            return True
+          else:
+            return False
+
+      def multiVarDetect(text, username):
+        multiUsernameAllList = comboDict['multiUsernameAllList']
+        for checkList in multiUsernameAllList:
+          if all(word in username for word in checkList):
+            return True
+
       # ------------------------------------------------------------------------
 
       # Normalize usernames and text, remove multiple whitespace and invisible chars
@@ -946,6 +970,10 @@ def check_if_only_link(string):
         add_spam(current, config, miscData, currentCommentDict, videoID)
       elif config['detect_link_spam'] and check_if_only_link(commentTextNormalized.strip()):
         add_spam(current, config, miscData, currentCommentDict, videoID)
+      elif find_accompanying_link_spam(commentTextNormalized.lower()):
+        add_spam(current, config, miscData, currentCommentDict, videoID)
+      elif multiVarDetect(commentTextNormalized.lower(), authorChannelName.lower()):
+        add_spam(current, config, miscData, currentCommentDict, videoID)
       elif sensitive and re.search(smartFilter['usernameConfuseRegex'], authorChannelName):
         add_spam(current, config, miscData, currentCommentDict, videoID)
       elif not sensitive and (findObf(smartFilter['usernameConfuseRegex'], miscData.channelOwnerName, authorChannelName) or authorChannelName == miscData.channelOwnerName):

Scripts/prepare_modes.py

@@ -365,6 +365,12 @@ def prepare_filter_mode_smart(scanMode, config, miscData, sensitive=False):
     'cashRegex': cashRegex,
   }
 
+  accompanyingLinkSpamDict = {
+    'accompanyingLinkSpamPhrasesList': filter.accompanyingLinkSpamPhrasesList,
+    'notSpecial': filter.notSpecial,
+    'videoLinkRegex': re.compile(r"((?:https?:)?\/\/)?((?:www|m)\.)?((?:youtube\.com|youtu.be))(\/(?:[\w\-]+\?v=|embed\/|v\/)?)([\w\-]+)(\S+)?"),
+  }
+
   print("  Loading Filters  [======                        ]", end="\r")
 
   # Compile regex with upper case, otherwise many false positive character matches
@@ -438,7 +444,9 @@ def prepare_filter_mode_smart(scanMode, config, miscData, sensitive=False):
     'sensitiveRootDomainRegex': sensitiveRootDomainRegex,
     'unicodeCategoriesStrip': unicodeCategoriesStrip,
     'spamListCombinedRegex': spamListCombinedRegex,
-    'threadFiltersDict': threadFiltersDict
+    'threadFiltersDict': threadFiltersDict,
+    'accompanyingLinkSpamDict': accompanyingLinkSpamDict,
+    'comboDict': filter.comboDict
     }
   print("                                                                 ") # Erases line that says "loading filters"
 

YTSpammerPurge.py

@@ -36,8 +36,8 @@
 ### IMPORTANT:  I OFFER NO WARRANTY OR GUARANTEE FOR THIS SCRIPT. USE AT YOUR OWN RISK.
 ###             I tested it on my own and implemented some failsafes as best as I could,
 ###             but there could always be some kind of bug. You should inspect the code yourself.
-version = "2.17.0-Dev2"
-configVersion = 31
+version = "2.17.0-Beta1"
+configVersion = 32
 #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
 print("Importing Script Modules...")
 # Import other module files
@@ -430,6 +430,7 @@ def primaryInstance(miscData):
     # User selects scanning mode,  while Loop to get scanning mode, so if invalid input, it will keep asking until valid input
     print("\n{:<59}{:<18}{:>7}".format("> At any prompt, enter 'X' to return here", updateStringLabel, updateString))
     print("> Enter 'Q' now to quit")
+    print(f"{F.LIGHTYELLOW_EX}NOTE: This beta version uses a new method for checking updates. It is possible it will not work properly, so you should periodically check for new versions on the GitHub page yourself until the stable release.{S.R}")
 
     print(f"\n\n-------------------------------- {F.YELLOW}Scanning Options{S.R} --------------------------------")
     print(f"      1. Scan {F.LIGHTCYAN_EX}specific videos{S.R}")