whois client was using the first whois server in list rather than the whois server associated with the domain we are querying issue #74

README.rst edited online with Bitbucket

added support for native client

added pretty formatting to string rendering

return None when an attribute does not exist for a TLD

added rf class issue #72

fixed recent domain implementations #72

converted README from markdown to RST for pypi

merged

[#71] missing .mobi, .io, .kg, .su, .biz added classes with inheritance of existing ones, because of same registrar. .kg is trickier. it's registrar is unique, and i haven't figured out how to parse name_servers properly

add socket timeout exception

added manifest to add missing tlds.txt file

simplified EU name servers

README.md edited online with Bitbucket

changed base class to dictionary interface

added entries for DE, NL, CA, BE

Always import datetime no matter if dateutil is present

Handle experimental lookup method for bad domains Attempting to use NICClient with bad domains could hit a bad logical path that leads to an attempt to reference a variable that hasn't been defined. So in case we don't manage to find a nichost for a domain, just set `result` to `None`

committed support for new .org and .info format #60

[#59] Add support for dateutil.parser dayfirst and yearfirst arguments The WhoisEntry object should expose a dayfirst and yearfist argument to signify how the datetime format for the TLD should be parsed by dateutil. The base WhoisEntry class gives both those fields a value of False. Those are also the default values dateutil assumes.

Use timezone naive datetime objects Otherwise comparisons don't always work

Fix bad import in parser.py

Stylistic improvements on parser.py

Better date parsing support if python-dateutil is available The `strptime` and `strftime` functions can't deal with timezone names not known to the current system when there's a %Z in the format string.

Ensure lookups work with both unicode and bytes objects

Add .рф TLD expiration_date regex

Add support for punycode TLDs

Fix extract_domain to work with unicode domains

Pass all domains through encode('idna') in NICClient This makes sure that non-ascii domains will be properly encoded in punnycode before sending the whois query to the server.

Allow explicit usage of NICClient even if whois binary is available

Stylistic fixes on the NICClient class

Move tlds.txt to a data/ folder and add it to the package

Raise AttributeError instead of KeyError This way it's possible to use `getattr`

Read in all TLDs from a file File taken from https://data.iana.org/TLD/tlds-alpha-by-domain.txt

added dns check to WhoisEntry

Add .bg expiry date format withou mseconds

Add expiry_date regex for .bg domains

avoid DNS lookup of IP addresses

added support for pickling the results

added support for .org expiry date

added IP test cases

added support for PT domains

added eu domain

added support for kr domain

support polish date format

added support for AU domain

avoid duplicate values

applied Brazilian patch

fixed tuple string patch

applied fi patch

fixed test cases

added support for .pl domain

added support for org.uk

removed debugging

added support for IP addresses

automatically parse dates

fixed builtin whois parser for pwhois.rotld.ro

added support for japanese domains added support for automatically parsing datetime strings

applied French domain patch

UK name servers patch

applied Finish parser patch

updated setup

initial commit to mercurial