asyncwhois

asyncwhois

asyncwhois | Python utility for WHOIS and RDAP queries.

Quickstart

import asyncio
from pprint import pprint

import asyncwhois

# pick a domain
domain = 'bitcoin.org'
# domain could also be a URL; asyncwhois uses tldextract to parse the URL
# domain = 'https://www.google.com?q=asyncwhois'

# basic example
query_string, parsed_dict = asyncwhois.whois(domain)
# query_string  # The semi-free text output from the whois server
# parsed_dict   # A dictionary of key:values extracted from `query_string`

# asyncio example
loop = asyncio.get_event_loop()
query_string, parsed_dict = loop.run_until_complete(asyncwhois.aio_whois(domain))

pprint(parsed_dict)
"""
{created: datetime.datetime(2008, 8, 18, 13, 19, 55),
 dnssec: 'unsigned',
 domain_name: 'bitcoin.org',
 expires: datetime.datetime(2029, 8, 18, 13, 19, 55),
 name_servers: ['dns1.registrar-servers.com', 'dns2.registrar-servers.com'],
 registrant_address: 'P.O. Box 0823-03411',
 registrant_city: 'Panama',
 registrant_country: 'PA',
 registrant_name: 'WhoisGuard Protected',
 registrant_organization: 'WhoisGuard, Inc.',
 registrant_state: 'Panama',
 registrant_zipcode: '',
 registrar: 'NAMECHEAP INC',
 status: ['clientTransferProhibited '
          'https://icann.org/epp#clientTransferProhibited'],
 updated: datetime.datetime(2019, 11, 24, 13, 58, 35, 940000)}
 ...
 """

# set `find_authoritative_server=False` to only query the TLD's whois server as specified in the IANA root db. 
# this will generally speed up execution times, but responses may not include complete information. 
query_string, parsed_dict = asyncwhois.whois("google.com", find_authoritative_server=False)

# support for IPv4, IPv6, and ASNs too
ipv4 = "8.8.8.8"
query_string, parsed_dict = asyncwhois.whois(ipv4)
pprint(parsed_dict)
"""
{abuse_address: None,
 abuse_email: '[e-mail]',
 abuse_handle: 'ABUSE5250-ARIN',
 abuse_name: 'Abuse',
 abuse_phone: '[phone]',
 abuse_rdap_ref: 'https://rdap.arin.net/registry/entity/ABUSE5250-ARIN',
 cidr: '8.8.8.0/24',
 net_handle: 'NET-8-8-8-0-2',
 net_name: 'GOGL',
 net_range: '8.8.8.0 - 8.8.8.255',
 net_type: 'Direct Allocation',
 org_address: '1600 Amphitheatre Parkway',
 org_city: 'Mountain View',
 org_country: 'US',
 org_id: 'GOGL',
 ...
"""

RDAP

The whodap project is used behind the scenes to perform RDAP queries.

domain = "https://google.com"
query_string, parsed_dict = asyncwhois.rdap(domain)
# OR with asyncio
query_string, parsed_dict = loop.run_until_complete(asyncwhois.aio_rdap(domain))

# Reusable client (caches the RDAP bootstrap server list, so it is faster for doing multiple calls)
client = asyncwhois.DomainClient()
for domain in ["google.com", "tesla.coffee", "bitcoin.org"]:
    query_string, parsed_dict = client.rdap(domain)
    # query_string, parsed_dict = await client.aio_rdap(domain)

# Using a proxy or need to configure something HTTP related? Try reconfiguring the client:
whodap_client = whodap.DNSClient.new_client(
    httpx_client=httpx.Client(proxies="https://proxy:8080")
)
# whodap_client = await whodap.DNSClient.new_aio_client(httpx_client=httpx.AsyncClient(proxies="https://proxy:8080"))
client = asyncwhois.DomainClient(whodap_client=whodap_client)

Proxies

SOCKS proxies are supported for WHOIS and RDAP queries.

import whodap

tor_host = "localhost"
tor_port = 9050

# WHOIS
query_string, parsed_dict = asyncwhois.whois(
    "8.8.8.8", proxy_url=f"socks5://{tor_host}:{tor_port}"
)

# RDAP
import httpx
from httpx_socks import SyncProxyTransport, AsyncProxyTransport  # EXTERNAL DEPENDENCY for SOCKS Proxies 

transport = SyncProxyTransport.from_url(f"socks5://{tor_host}:{tor_port}")
httpx_client = httpx.Client(transport=transport)
whodap_client = whodap.IPv6Client.new_client(httpx_client=httpx_client)
query_string, parsed_dict = asyncwhois.rdap('2001:4860:4860::8888', whodap_client=whodap_client)

transport = AsyncProxyTransport.from_url(f"socks5://{tor_user}:{tor_pw}@{tor_host}:{tor_port}")
async with httpx.AsyncClient(transport=transport) as httpx_client:
    whodap_client = await whodap.DNSClient.new_aio_client(httpx_client=httpx_client)
    query_string, parsed_dict = await asyncwhois.aio_rdap('bitcoin.org', whodap_client=whodap_client)

Exported Functions

Function/Object	Description
DomainClient	Reusable client for WHOIS or RDAP domain queries
NumberClient	Reusable client for WHOIS or RDAP ipv4/ipv6 queries
ASNClient	Reusable client for RDAP asn queries
whois	WHOIS entrypoint for domain, ipv4, or ipv6 queries
rdap	RDAP entrypoint for domain, ipv4, ipv6, or asn queries
aio_whois	async counterpart to whois
aio_rdap	async counterpart to rdap
whois_ipv4	[DEPRECATED] WHOIS lookup for ipv4 addresses
whois_ipv6	[DEPRECATED] WHOIS lookup for ipv6 addresses
rdap_domain	[DEPRECATED] RDAP lookup for domain names
rdap_ipv4	[DEPRECATED] RDAP lookup for ipv4 addresses
rdap_ipv6	[DEPRECATED] RDAP lookup for ipv6 addresses
rdap_asn	[DEPRECATED] RDAP lookup for Autonomous System Numbers
aio_whois_domain	[DEPRECATED] async counterpart to whois_domain
aio_whois_ipv4	[DEPRECATED] async counterpart to whois_ipv4
aio_whois_ipv6	[DEPRECATED] async counterpart to whois_ipv6
aio_rdap_domain	[DEPRECATED] async counterpart to rdap_domain
aio_rdap_ipv4	[DEPRECATED] async counterpart to rdap_ipv4
aio_rdap_ipv6	[DEPRECATED] async counterpart to rdap_ipv6
aio_rdap_asn	[DEPRECATED] async counterpart to rdap_asn

Contributions

Parsed output not what you expected? Unfortunately, "the format of responses [from a WHOIS server] follow a semi-free text format". Therefore, situations will arise where this module does not support parsing the output from a specific server, and you may find yourself needing more control over how parsing happens. Fortunately, you can create customized parsers to suit your needs.

Example: This is a snippet of the output from running the "whois google.be" command.

Domain:	google.be
Status:	NOT AVAILABLE
Registered:	Tue Dec 12 2000

Registrant:
    Not shown, please visit www.dnsbelgium.be for webbased whois.

Registrar Technical Contacts:
    Organisation:	MarkMonitor Inc.
    Language:	en
    Phone:	[phone]
    Fax:	[phone]

Registrar:
    Name:	 MarkMonitor Inc.
    Website: http://www.markmonitor.com

Nameservers:
    ns2.google.com
    ns1.google.com
    ns4.google.com
    ns3.google.com

Keys:

Flags:
    clientTransferProhibited
...

In this case, the "name servers" are listed on separate lines. The default BaseParser regexes won't find all of these server names. In order to accommodate this extra step, the "parse" method was overwritten within the parser subclass as seen below:

class RegexBE(BaseParser):
    _be_expressions = {  # the base class (BaseParser) will handle these regexes
        BaseKeys.CREATED: r'Registered: *(.+)',
        BaseKeys.REGISTRAR: r'Registrar:\n.+Name: *(.+)',
        BaseKeys.REGISTRANT_NAME: r'Registrant:\n *(.+)'
    }
    
    def __init__(self):
        super().__init__()
        self.update_reg_expressions(self._be_expressions)
    
    def parse(self, blob: str) -> Dict[str, Any]:
        # run base class parsing for other keys
        parsed_output = super().parse(blob)
        # custom parsing is needed to extract all the name servers
        ns_match = re.search(r"Name servers: *(.+)Keys: ", blob, re.DOTALL)
        if ns_match:
            parsed_output[BaseKeys.NAME_SERVERS] = [m.strip() for m in ns_match.group(1).split('\n') if m.strip()]
        return parsed_output