Separate blacklist check

.gitignore

@@ -1,4 +1,5 @@
 *.pyc
+validate_email/lib
 build
 dist
 MANIFEST

CHANGELOG.txt

@@ -1,3 +1,6 @@
+0.1.12:
+- Blacklist/whitelist domains checking is now independent of regex checking.
+
 0.1.11:
 - Handling IDNA errors
 

requirements.txt

@@ -1,8 +1,2 @@
 dnspython==1.16.0
-entrypoints==0.3
-flake8==3.7.7
 idna==2.8
-isort==4.3.21
-mccabe==0.6.1
-pycodestyle==2.5.0
-pyflakes==2.1.1

setup.py

@@ -4,7 +4,7 @@ from urllib.request import urlopen
 from setuptools import find_packages, setup
 from setuptools.command.build_py import build_py
 
-blacklist_url = (
+BLACKLIST_URL = (
     'https://raw.githubusercontent.com/martenson/disposable-email-domains/'
     'master/disposable_email_blocklist.conf')
 
@@ -15,10 +15,10 @@ class PostBuildPyCommand(build_py):
     def run(self):
         if self.dry_run:
             return super().run()
-        with urlopen(url=blacklist_url) as fd:
+        with urlopen(url=BLACKLIST_URL) as fd:
             content = fd.read().decode('utf-8')
         target_dir = join(self.build_lib, 'validate_email/lib')
-        self.mkpath(target_dir)
+        self.mkpath(name=target_dir)
         with open(join(target_dir, 'blacklist.txt'), 'w') as fd:
             fd.write(content)
         super().run()
@@ -26,7 +26,7 @@ class PostBuildPyCommand(build_py):
 
 setup(
     name='py3-validate-email',
-    version='0.1.11',
+    version='0.1.12',
     packages=find_packages(exclude=['tests']),
     install_requires=['dnspython>=1.16.0', 'idna>=2.8'],
     author='László Károlyi',

tests/test_blacklist_check.py

@@ -0,0 +1,59 @@
+from os import makedirs
+from os.path import dirname, join
+from unittest.case import TestCase
+from urllib.request import urlopen
+
+from validate_email import validate_email
+from validate_email.domainlist_check import domainlist_check
+
+BLACKLIST_URL = (
+    'https://raw.githubusercontent.com/martenson/disposable-email-domains/'
+    'master/disposable_email_blocklist.conf')
+
+
+class DlBlacklist(object):
+    'Emulating downloading of blacklists on post-build command.'
+
+    def __init__(self):
+        from validate_email import domainlist_check
+        self.build_lib = dirname(dirname(domainlist_check.__file__))
+
+    def mkpath(self, name: str):
+        'Emulate mkpath.'
+        makedirs(name=name, exist_ok=True)
+
+    def run(self):
+        'Deploy function identical to the one in setup.py.'
+        with urlopen(url=BLACKLIST_URL) as fd:
+            content = fd.read().decode('utf-8')
+        target_dir = join(self.build_lib, 'validate_email/lib')
+        self.mkpath(name=target_dir)
+        with open(join(target_dir, 'blacklist.txt'), 'w') as fd:
+            fd.write(content)
+
+
+class BlacklistCheckTestCase(TestCase):
+    'Testing if the included blacklist filtering works.'
+
+    def test_blacklist_positive(self):
+        'Disallows blacklist item: mailinator.com.'
+        dl = DlBlacklist()
+        dl.run()
+        self.assertFalse(expr=domainlist_check(
+            email_address='pa2@mailinator.com'))
+        self.assertFalse(expr=validate_email(
+            email_address='pa2@mailinator.com', check_regex=False,
+            use_blacklist=True))
+        self.assertFalse(expr=validate_email(
+            email_address='pa2@mailinator.com', check_regex=True,
+            use_blacklist=True))
+
+    def test_blacklist_negative(self):
+        'Allows a domain not in the blacklist.'
+        self.assertTrue(expr=domainlist_check(
+            email_address='pa2@some-random-domain-thats-not-blacklisted.com'))
+
+    def test_erroneous_email(self):
+        'Will reject emails in erroneous format.'
+        self.assertFalse(expr=domainlist_check(
+            email_address='pa2-mailinator.com'))

validate_email/domainlist_check.py

@@ -0,0 +1,46 @@
+from os.path import dirname, join
+from typing import Optional
+
+SetOrNone = Optional[set]
+
+
+class DomainListValidator(object):
+    'Check the provided email against domain lists.'
+    domain_whitelist = frozenset()
+    domain_blacklist = frozenset()
+
+    def __init__(
+            self, whitelist: SetOrNone = None, blacklist: SetOrNone = None):
+        if whitelist:
+            self.domain_whitelist = set(x.lower() for x in whitelist)
+        if blacklist:
+            self.domain_blacklist = set(x.lower() for x in blacklist)
+        else:
+            self._load_builtin_blacklist()
+
+    def _load_builtin_blacklist(self):
+        'Load our built-in blacklist.'
+        path = join(dirname(__file__), 'lib', 'blacklist.txt')
+        try:
+            with open(path) as fd:
+                lines = fd.readlines()
+        except FileNotFoundError:
+            return
+        self.domain_blacklist = \
+            set(x.strip().lower() for x in lines if x.strip())
+
+    def __call__(self, email_address: str) -> bool:
+        'Do the checking here.'
+        if not email_address or '@' not in email_address:
+            return False
+
+        user_part, domain_part = email_address.rsplit('@', 1)
+
+        if domain_part in self.domain_whitelist:
+            return True
+        if domain_part in self.domain_blacklist:
+            return False
+        return True
+
+
+domainlist_check = DomainListValidator()

validate_email/regex_check.py

@@ -1,5 +1,4 @@
 from ipaddress import IPv4Address, IPv6Address
-from os.path import dirname, join
 from typing import Optional
 
 from .constants import HOST_REGEX, LITERAL_REGEX, USER_REGEX
@@ -32,44 +31,18 @@ def _validate_ipv46_address(value: str) -> bool:
     return _validate_ipv6_address(value)
 
 
-class EmailValidator(object):
+class RegexValidator(object):
     'Slightly adjusted email regex checker from the Django project.'
-    domain_whitelist = frozenset('localhost')
-    domain_blacklist = frozenset()
 
-    def __init__(
-            self, whitelist: SetOrNone = None, blacklist: SetOrNone = None):
-        self.domain_whitelist = set(whitelist) \
-            if whitelist else self.domain_whitelist
-        self._load_blacklist(blacklist=blacklist)
-
-    def _load_blacklist(self, blacklist: SetOrNone = None):
-        'Load our blacklist.'
-        self.domain_blacklist = set(blacklist) \
-            if blacklist else self.domain_blacklist
-        path = join(dirname(__file__), 'lib', 'blacklist.txt')
-        try:
-            with open(path) as fd:
-                lines = fd.readlines()
-        except FileNotFoundError:
-            return
-        self.domain_blacklist = self.domain_blacklist.union(
-            x.strip() for x in lines)
-
-    def __call__(self, value: str, use_blacklist: bool = True) -> bool:
-        if not value or '@' not in value:
+    def __call__(self, email_address: str, use_blacklist: bool = True) -> bool:
+        if not email_address or '@' not in email_address:
             return False
 
-        user_part, domain_part = value.rsplit('@', 1)
+        user_part, domain_part = email_address.rsplit('@', 1)
 
         if not USER_REGEX.match(user_part):
             return False
 
-        if domain_part in self.domain_whitelist:
-            return True
-        if domain_part in self.domain_blacklist:
-            return False
-
         if not self.validate_domain_part(domain_part):
             # Try for possible IDN domain-part
             try:
@@ -93,4 +66,4 @@ class EmailValidator(object):
         return False
 
 
-regex_check = EmailValidator()
+regex_check = RegexValidator()

validate_email/validate_email.py

@@ -1,5 +1,6 @@
 from typing import Optional
 
+from .domainlist_check import domainlist_check
 from .mx_check import mx_check
 from .regex_check import regex_check
 
@@ -15,8 +16,9 @@ def validate_email(
 
     Return `None` if the result is ambigious.
     """
-    if check_regex and not regex_check(
-            value=email_address, use_blacklist=use_blacklist):
+    if check_regex and not regex_check(email_address=email_address):
+        return False
+    if use_blacklist and not domainlist_check(email_address=email_address):
         return False
     if not check_mx:
         return True