Faster checking for blocked domains

blocking.py

@@ -8,6 +8,7 @@ __status__="Production"
 
 import os
 from utils import isEvil
+from utils import evilIncarnate
 
 def addGlobalBlock(baseDir: str, \
                    blockNickname: str,blockDomain: str) -> bool:
@@ -113,6 +114,23 @@ def isBlockedHashtag(baseDir: str,hashtag: str) -> bool:
             return True
     return False
 
+def getDomainBlocklist(baseDir: str) -> str:
+    """Returns all globally blocked domains as a string
+    This can be used for fast matching to mitigate flooding
+    """
+    blockedStr=''
+
+    evilDomains=evilIncarnate()
+    for evil in evilDomains:
+        blockedStr+=evil+'\n'
+
+    globalBlockingFilename=baseDir+'/accounts/blocking.txt'
+    if not os.path.isfile(globalBlockingFilename):
+        return blockedStr
+    with open(globalBlockingFilename, 'r') as file:
+        blockedStr += file.read()
+    return blockedStr
+
 def isBlockedDomain(baseDir: str,domain: str) -> bool:
     """Is the given domain blocked?
     """

daemon.py

@@ -107,6 +107,7 @@ from blocking import addGlobalBlock
 from blocking import removeGlobalBlock
 from blocking import isBlockedHashtag
 from blocking import isBlockedDomain
+from blocking import getDomainBlocklist
 from config import setConfigParam
 from config import getConfigParam
 from roles import outboxDelegate
@@ -936,7 +937,14 @@ class PubServer(BaseHTTPRequestHandler):
         callingDomain=None
         if self.headers.get('Host'):
             callingDomain=self.headers['Host']
-            if isBlockedDomain(self.server.baseDir,callingDomain):
+
+            if self.server.blocklistUpdateCtr<=0:
+                self.server.blocklistUpdateCtr=self.server.blocklistUpdateInterval
+                self.server.domainBlocklist=getDomainBlocklist(self.server.baseDir)
+
+            self.server.blocklistUpdateCtr-=1
+
+            if callingDomain in self.server.domainBlocklist:
                 print('GET domain blocked: '+callingDomain)
                 self._400()
                 return
@@ -4381,7 +4389,14 @@ class PubServer(BaseHTTPRequestHandler):
         callingDomain=None
         if self.headers.get('Host'):
             callingDomain=self.headers['Host']
-            if isBlockedDomain(self.server.baseDir,callingDomain):
+
+            if self.server.blocklistUpdateCtr<=0:
+                self.server.blocklistUpdateCtr=self.server.blocklistUpdateInterval
+                self.server.domainBlocklist=getDomainBlocklist(self.server.baseDir)
+
+            self.server.blocklistUpdateCtr-=1
+
+            if callingDomain in self.server.domainBlocklist:
                 print('POST domain blocked: '+callingDomain)
                 self._400()
                 return
@@ -6194,6 +6209,12 @@ def runDaemon(blogsInstance: bool,mediaInstance: bool, \
         print('ERROR: HTTP server failed to start. '+str(e))
         return False
 
+    # This counter is used to update the list of blocked domains in memory.
+    # It helps to avoid touching the disk and so improves flooding resistance
+    httpd.blocklistUpdateCtr=0
+    httpd.blocklistUpdateInterval=100
+    httpd.domainBlocklist=getDomainBlocklist(baseDir)
+
     httpd.onionDomain=onionDomain
     httpd.useBlurHash=useBlurHash
     httpd.mediaInstance=mediaInstance

utils.py

@@ -93,12 +93,15 @@ def getStatusNumber() -> (str,str):
     published=currTime.strftime("%Y-%m-%dT%H:%M:%SZ")
     return statusNumber,published
 
+def evilIncarnate() -> []:
+    return ('gab.com','gabfed.com','spinster.xyz','kiwifarms.cc','djitter.com')
+
 def isEvil(domain: str) -> bool:
     if not isinstance(domain, str):
         print('WARN: Malformed domain '+str(domain))
         return True
     # https://www.youtube.com/watch?v=5qw1hcevmdU
-    evilDomains=('gab.com','gabfed.com','spinster.xyz','kiwifarms.cc','djitter.com')
+    evilDomains=evilIncarnate()
     for concentratedEvil in evilDomains:
         if domain.endswith(concentratedEvil):
             return True