Merge pull request #1 from richardpenman/master

Pull from origin
richardpenman · Jan 14, 2020 · a4794c4 · a4794c4
2 parents a10e26b + c6b441a
commit a4794c4
Show file tree

Hide file tree

Showing 11 changed files with 230 additions and 68 deletions.
diff --git a/setup.py b/setup.py
@@ -1,7 +1,7 @@
 import sys, os
 import setuptools
 
-version = '0.7.0'
+version = '0.7.2'
 
 setuptools.setup(
     name='python-whois',

diff --git a/test/samples/expected/google.de b/test/samples/expected/google.de
@@ -1 +1 @@
-{"domain_name": "google.de", "status": "connect", "updated_date": "2018-03-12T21:44:25+01:00", "name": null, "org": null, "address": null, "zipcode": null, "city": null, "country_code": null, "phone": null, "fax": null, "name_servers": ["ns1.google.com", "ns2.google.com", "ns3.google.com", "ns4.google.com"], "emails": null}
+{"domain_name": "google.de", "status": "connect", "updated_date": "2018-03-12 21:44:25+01:00", "name": null, "org": null, "address": null, "zipcode": null, "city": null, "country_code": null, "phone": null, "fax": null, "name_servers": ["ns1.google.com", "ns2.google.com", "ns3.google.com", "ns4.google.com"], "emails": null}
diff --git a/test/samples/expected/sapo.pt b/test/samples/expected/sapo.pt
@@ -1 +1 @@
-{"domain_name": "sapo.pt", "expiration_date": "2018-11-02 00:00:00", "updated_date": null, "creation_date": "2002-10-30 00:00:00", "status": "ACTIVE"}
+{"domain_name": "sapo.pt", "expiration_date": "2019-11-02 23:59:00", "updated_date": null, "creation_date": "2002-10-30 00:00:00", "status": "Registered"}
diff --git a/test/samples/expected/web.de b/test/samples/expected/web.de
@@ -1 +1 @@
-{"domain_name": "web.de", "expiration_date": null, "updated_date": ["2016-04-11T11:09:54+02:00", "2011-08-10T17:09:10+02:00"], "creation_date": null, "status": "connect"}
+{"domain_name": "web.de", "expiration_date": null, "updated_date": ["2016-04-11 11:09:54+02:00", "2011-08-10 17:09:10+02:00"], "creation_date": null, "status": "connect"}
diff --git a/test/samples/whois/sapo.pt b/test/samples/whois/sapo.pt
@@ -1,31 +1,20 @@
-Nome de domínio / Domain Name: sapo.pt
-Data de registo / Creation Date (dd/mm/yyyy): 30/10/2002
-Data de expiração / Expiration Date (dd/mm/yyyy): 02/11/2018
-Estado / Status: ACTIVE
-
-Titular / Registrant
-   MEO - SERVIÇOS DE COMUNICAÇÕES E MULTIMÉDIA S.A.
-   A/C Direção de Tecnologias de Informação
-   Av. Fontes Pereira de Melo, 40
-   1069-300 Lisboa
-   Email: gestao.dominios@telecom.pt
-
-Entidade Gestora / Billing Contact
-   MEO - SERVIÇOS DE COMUNICAÇÕES E MULTIMÉDIA S.A.
-   Email: gestao.dominios@telecom.pt
-
-Responsável Técnico / Tech Contact
-   MEO - SERVIÇOS DE COMUNICAÇÕES E MULTIMÉDIA S.A.
-   Email: gestao.dominios@telecom.pt
-
-Nameserver Information
-   Nameserver: sapo.pt  NS      ns2.sapo.pt.
-   Nameserver: sapo.pt  NS      dns02.sapo.pt.
-   Nameserver: sapo.pt  NS      ns.sapo.pt.
-   Nameserver: sapo.pt  NS      dns01.sapo.pt.
-   Nameserver: ns2.sapo.pt.     A       212.55.154.194
-   Nameserver: dns02.sapo.pt.   AAAA    2001:8a0:2206:4:213:13:30:116
-   Nameserver: ns.sapo.pt.      A       212.55.154.202
-   Nameserver: dns01.sapo.pt.   AAAA    2001:8a0:2106:4:213:13:28:116
-   Nameserver: dns01.sapo.pt.   A       213.13.28.116
-   Nameserver: dns02.sapo.pt.   A       213.13.30.116
+Domain: sapo.pt
+Domain Status: Registered
+Creation Date: 30/10/2002 00:00:00
+Expiration Date: 02/11/2019 23:59:00
+Owner Name: MEO - SERVI?OS DE COMUNICA??ES E MULTIM?DIA S.A.
+Owner Address: A/C Dire??o de Tecnologias de Informa??o
+Owner Locality: Av. Fontes Pereira de Melo, 40
+Owner ZipCode: 1069-300
+Owner Locality ZipCode: Av. Fontes Pereira de Melo, 40
+Owner Email: gestao.dominios@telecom.pt
+Admin Name: MEO - SERVI?OS DE COMUNICA??ES E MULTIM?DIA S.A.
+Admin Address: A/C Dire??o de Tecnologias de Informa??o
+Admin Locality: Av. Fontes Pereira de Melo, 40
+Admin ZipCode: 1069-300
+Admin Locality ZipCode: Av. Fontes Pereira de Melo, 40
+Admin Email: gestao.dominios@telecom.pt
+Name Server: ns2.sapo.pt | IPv4: 212.55.154.194 and IPv6:
+Name Server: dns01.sapo.pt | IPv4: 213.13.28.116 and IPv6: 2001:8a0:2106:4:213:13:28:116
+Name Server: dns02.sapo.pt | IPv4: 213.13.30.116 and IPv6: 2001:8a0:2206:4:213:13:30:116
+Name Server: ns.sapo.pt | IPv4: 212.55.154.202 and IPv6:
diff --git a/test/test_main.py b/test/test_main.py
@@ -31,3 +31,17 @@ def test_unicode_domain_and_tld(self):
         url = 'http://россия.рф/'
         domain = 'россия.рф'
         self.assertEqual(domain, extract_domain(url))
+
+    def test_ipv6(self):
+        """ Verify that ipv6 addresses work """
+        url = '2607:f8b0:4006:802::200e'
+        domain = '1e100.net'
+        # double extract_domain() so we avoid possibly changing hostnames like lga34s12-in-x0e.1e100.net
+        self.assertEqual(domain, extract_domain(extract_domain(url)))
+
+    def test_ipv4(self):
+        """ Verify that ipv4 addresses work """
+        url = '172.217.3.110'
+        domain = '1e100.net'
+        # double extract_domain() so we avoid possibly changing hostnames like lga34s18-in-f14.1e100.net
+        self.assertEqual(domain, extract_domain(extract_domain(url)))
diff --git a/test/test_parser.py b/test/test_parser.py
@@ -1,3 +1,4 @@
+# -*- coding: utf-8 -*-
 from __future__ import print_function
 from __future__ import unicode_literals
 from __future__ import division
@@ -97,7 +98,7 @@ def date2str4json(obj):
                     print("%s \t(%s):\t Missing in results" % (domain, key,))
                     fail += 1
                     continue
-                
+
                 result = results.get(key)
                 if isinstance(result, list):
                     result = [str(element) for element in result]
@@ -125,7 +126,7 @@ def test_ca_parse(self):
         Registry Registrant ID: 70
 
         Registrant Name:       Test Industries
-        Registrant Organization: 
+        Registrant Organization:
 
         Admin Name:            Test Person1
         Admin Street:          Test Address
@@ -169,6 +170,39 @@ def test_ca_parse(self):
         }
         self._parse_and_compare('testcompany.ca', data, expected_results,
                                 whois_entry=WhoisCa)
+
+    def test_cn_parse(self):
+        data = """
+            Domain Name: cnnic.com.cn
+            ROID: 20021209s10011s00047242-cn
+            Domain Status: serverDeleteProhibited
+            Domain Status: serverUpdateProhibited
+            Domain Status: serverTransferProhibited
+            Registrant ID: s1255673574881
+            Registrant: 中国互联网络信息中心
+            Registrant Contact Email: servicei@cnnic.cn
+            Sponsoring Registrar: 北京新网数码信息技术有限公司
+            Name Server: a.cnnic.cn
+            Name Server: b.cnnic.cn
+            Name Server: c.cnnic.cn
+            Name Server: d.cnnic.cn
+            Name Server: e.cnnic.cn
+            Registration Time: 2000-09-14 00:00:00
+            Expiration Time: 2023-08-16 16:26:39
+            DNSSEC: unsigned
+        """
+        expected_results = {
+            "domain_name": "cnnic.com.cn",
+            "registrar": "北京新网数码信息技术有限公司",
+            "creation_date": "2000-09-14 00:00:00",
+            "expiration_date": "2023-08-16 16:26:39",
+            "name_servers": ["a.cnnic.cn", "b.cnnic.cn", "c.cnnic.cn", "d.cnnic.cn", "e.cnnic.cn"],
+            "status": ["serverDeleteProhibited", "serverUpdateProhibited", "serverTransferProhibited"],
+            "emails": "servicei@cnnic.cn",
+            "dnssec": "unsigned",
+            "name": "中国互联网络信息中心"
+        }
+        self._parse_and_compare('cnnic.com.cn', data, expected_results)
 
     def test_il_parse(self):
         data = """

diff --git a/test/test_query.py b/test/test_query.py
@@ -23,3 +23,27 @@ def test_simple_unicode_domain(self):
     def test_unicode_domain_and_tld(self):
         domain = 'россия.рф'
         whois(domain)
+
+    def test_ipv4(self):
+        """ Verify ipv4 addresses. """
+        domain = '172.217.3.110'
+        whois_results = whois(domain)
+        if isinstance(whois_results['domain_name'], list):
+            domain_names = [_.lower() for _ in whois_results['domain_name']]
+        else:
+            domain_names = [whois_results['domain_name'].lower()]
+
+        self.assertIn('1e100.net', domain_names)
+        self.assertIn('ns1.google.com', [_.lower() for _ in whois_results['name_servers']])
+
+    def test_ipv6(self):
+        """ Verify ipv6 addresses. """
+        domain = '2607:f8b0:4006:802::200e'
+        whois_results = whois(domain)
+        if isinstance(whois_results['domain_name'], list):
+            domain_names = [_.lower() for _ in whois_results['domain_name']]
+        else:
+            domain_names = [whois_results['domain_name'].lower()]
+
+        self.assertIn('1e100.net', domain_names)
+        self.assertIn('ns1.google.com', [_.lower() for _ in whois_results['name_servers']])
diff --git a/whois/__init__.py b/whois/__init__.py
@@ -16,10 +16,13 @@
 from .whois import NICClient
 
 
+# thanks to https://www.regextester.com/104038
+IPV4_OR_V6 = re.compile(r"((^\s*((([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])\.){3}([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5]))\s*$)|(^\s*((([0-9A-Fa-f]{1,4}:){7}([0-9A-Fa-f]{1,4}|:))|(([0-9A-Fa-f]{1,4}:){6}(:[0-9A-Fa-f]{1,4}|((25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)(\.(25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)){3})|:))|(([0-9A-Fa-f]{1,4}:){5}(((:[0-9A-Fa-f]{1,4}){1,2})|:((25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)(\.(25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)){3})|:))|(([0-9A-Fa-f]{1,4}:){4}(((:[0-9A-Fa-f]{1,4}){1,3})|((:[0-9A-Fa-f]{1,4})?:((25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)(\.(25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)){3}))|:))|(([0-9A-Fa-f]{1,4}:){3}(((:[0-9A-Fa-f]{1,4}){1,4})|((:[0-9A-Fa-f]{1,4}){0,2}:((25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)(\.(25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)){3}))|:))|(([0-9A-Fa-f]{1,4}:){2}(((:[0-9A-Fa-f]{1,4}){1,5})|((:[0-9A-Fa-f]{1,4}){0,3}:((25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)(\.(25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)){3}))|:))|(([0-9A-Fa-f]{1,4}:){1}(((:[0-9A-Fa-f]{1,4}){1,6})|((:[0-9A-Fa-f]{1,4}){0,4}:((25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)(\.(25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)){3}))|:))|(:(((:[0-9A-Fa-f]{1,4}){1,7})|((:[0-9A-Fa-f]{1,4}){0,5}:((25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)(\.(25[0-5]|2[0-4]\d|1\d\d|[1-9]?\d)){3}))|:)))(%.+)?\s*$))")
+
 
 def whois(url, command=False, flags=0):
     # clean domain to expose netloc
-    ip_match = re.match(r"^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}$", url)
+    ip_match = IPV4_OR_V6.match(url)
     if ip_match:
         domain = url
         try:
@@ -63,8 +66,12 @@ def extract_domain(url):
     globo.com
     >>> print(extract_domain('1-0-1-1-1-0-1-1-1-1-1-1-1-.0-0-0-0-0-0-0-0-0-0-0-0-0-10-0-0-0-0-0-0-0-0-0-0-0-0-0.info'))
     0-0-0-0-0-0-0-0-0-0-0-0-0-10-0-0-0-0-0-0-0-0-0-0-0-0-0.info
+    >>> print(extract_domain('2607:f8b0:4006:802::200e'))
+    1e100.net
+    >>> print(extract_domain('172.217.3.110'))
+    1e100.net
     """
-    if re.match(r'\d+\.\d+\.\d+\.\d+', url):
+    if IPV4_OR_V6.match(url):
         # this is an IP address
         return socket.gethostbyaddr(url)[0]