[prev in list] [next in list] [prev in thread] [next in thread] 

List:       rwhois
Subject:    Re: scripts
From:       David Blacka <davidb () rwhois ! net>
Date:       1997-03-12 19:25:46
Message-ID: kovi6whpz9.fsf () shaker ! internic ! net
[Download RAW message or body]

>>>>> "Paul" == Paul Roberts <proberts@edge.edge.net> writes:

 Paul> On Thu, 6 Mar 1997, Jim Bacon wrote:
 >> Does anyone have scripts for maintaining the databse files that
 >> they would share?

 Paul> (tap-tap-tap I this thing on?  It's dreadfully quiet in here.)

Yes.  Everything seems to be working from here :-)  I guess no one has
had much to say recently.

[ some details of nifty script snipped ]

Paul, this sounds like an excellent tool.  I wouldn't mind putting
this in the rwhoisd distribution in the future.

Some things that I would like you (and everyone on the list) to be
aware of, though: we are planning to migrate the rwhois root server
(root.rwhois.net or rwhois.internic.net) to the 1.5 server, which
implements the 1.5 protocol.  The upshot of this is that when this
happens, your script is likely to stop working.  The rwhoisd v.1.5
conforms to what is outlined in the 1.5 spec that was just announced.
The 1.5 protocol is a bit different (although it looks subtantially
the same).

 Paul> The second part opens up the domainz file and then repeatedly
 Paul> hits rwhois.internic.net and tries to grab the record for each
 Paul> domain, taking advantage of the fact that '%domain.edu' pretty
 Paul> much spools you the raw data that's stored in the server, along
 Paul> with a pretty inconsequential (and therefore ignored) error
 Paul> message.

This is what we call "dump" format.  It is the only machine parsable
format that was defined for v1.0.  In spec version 1.5, this is the
*only* display type supported.  The "human readable" whois style
formatting is left up to the client.

In v1.0, you could have accomplished the same thing with 'dump
domain.edu', or better yet, 'dump domain domain.edu'

In v1.5, you would just use 'domain.edu' or 'domain domain.edu'.

 Paul> This is pretty much a kludge, because there are a number of
 Paul> things which need to be changed, but there is virtually NO
 Paul> documentation I can find at the moment about what each token
 Paul> means.

Yeah.  I don't think there is any documentation for '%'.  I don't even
know what it does, and I've been maintaining the server :-).  In any
case, it was a whois hack that was carried over.  It won't work in the
v1.5 server, because strict whois compatibility was dropped.

 Paul> Things left to do...

 Paul> 	o Grab all those nifty little user entries, and vacuum them out,
 Paul> 	  too.

In the near future, with the v.1.5 server, the mechanism for doing
this will be different, in that a new data 'type' is introduced, ID.
All of these user entries or whatever that are associated with a
domain are done just by listing the object ID.

 Paul> Note: I'm quite sure there's going to be more than one person
 Paul> here who thinks that rwhois is quite self-explanatory, and that
 Paul> my questions are silly.  Flames will only be accepted with
 Paul> reasonable explanations.  Thanks.  ;)

No; my feeling is that most people find using the server considerably
less than self explanitory.  Feel free to send questions to this list,
or, if you think it isn't appropriate for the whole list, send it to
bugs@rwhois.net.

Since you are working on a script, I might suggest subscribing to
rwhois-workers@rwhois.net, too, although that list has been equally
quiet.

-- 
David Blacka			  Software Engineer		   Directory Services/
davidb@rwhois.net		  Network Solutions, Inc.  RWhois Development



[prev in list] [next in list] [prev in thread] [next in thread]