This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author leonn
Recipients eric.smith, leonn, ncoghlan, pitrou, pmoody
Date 2014-02-02.14:02:08
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <20140202140208.GC10572@dirac.q-ix.net>
In-reply-to <1391347974.2292.9.camel@fsol>
Content
> According to apt-cache, the version I tried with is
> 1:9.9.3.dfsg.P2-4ubuntu1

Ok, then it depends on something else apparently. It’s not so relevant anyway because even with the trailing dot, “host” is inconsistent between IPv4 and IPv6, and we should choose a consistent variant.

I’m still slightly in favour of having the trailing dot, because it is the correct representation of the DNS name. RFC1034 section 3.1 explicitly states that “Since a complete domain name ends with the root label, this leads to a printed form which ends in a dot.”. It goes on to explain that relative names (names not ending in a dot) “appear mostly at the user interface, where their interpretation varies from implementation to implementation” and that the most common interpretation is relative to the root zone.

So leaving the dot off is merely a convention, which is good enough for Wikipedia and RIPE documents; however I think in an API we should implement the more strict option. Also, I don’t see anything breaking with the dot, while without the dot, a user needs to take extra care when feeding the name to dnspython’s dns.resolver.query(), which would do undesired things without the trailing dot (append a search domain if there is one in /etc/resolv.conf).

Any other opinions?
History
Date User Action Args
2014-02-02 14:02:09leonnsetrecipients: + leonn, ncoghlan, pitrou, eric.smith, pmoody
2014-02-02 14:02:09leonnlinkissue20480 messages
2014-02-02 14:02:08leonncreate