As long as you don't try to get all 10,000 of your domains (and info) at
once, we probably won't notice. The logic being, you haven't kept track
until now - if it takes a week to parse, let it do so...
We're painfully aware of the lack of "data dumps" our system currently has
for our resellers. We WILL get to it, but it's a back burner issue. If the
data is THAT important, it should be tracked locally - it's when you decide
to start tracking (after some time) that becomes tricky. It'll happen...
until then, Tom's tricks will help (thanks Tom!) - as long as you remember
to be patient ;)
Charles Daminato
Product Manager (ccTLDs)
Tucows Inc. - [EMAIL PROTECTED]
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On
> Behalf Of Tom Brown
> Sent: March 7, 2001 1:55 PM
> To: [EMAIL PROTECTED]
> Subject: RE: How to retrive a list of all our customers domains
>
>
>
> did you guys break the sslbot again? (doesn't look like it. :) I had to
> write another one for a different supplier and came up with a much more
> robust page parsing algorithm... :-) [split out the tables, then split out
> the rows... then parse the rows] Almost looking forward to rewriting that
> sslbot script :-)
>
> Considering that forcing clients to write robot scripts to pound your
> servers into the dirt is a just generally a bad idea, and a simple text
> output screen to dump everything would be _trivial_ to write... you guys
> don't make this easy... (I know we've discussed this before, and I don't
> hold you responsible Chuck.... :)
>
> That said, I have no real desire to grab all the customer fields, and I
> would probably do that via a whois lookup instead anyway... much easier to
> parse text output than HTML, and it should be faster than using an SSL
> connection.
>
> Taking it from that point of view, the sslbot link shown below is a fine
> starting point, parsing whois output is trivial compared to handling the
> RWI authentication and parsing HTML ... e.g. use the sslbot to get the
> list of domains, then write a short script to hit the whois server for
> each domain listed by the sslbot.
>
> the only gotcha there is if OpenSRS gets upset because you pounded on
> their whois server too many times.
>
> -Tom
>
> (p.s. anyone who wants to respond to me, please send To: me, and cc: the
> list if you want... I don't generally read the list except for OpenSRS
> staff postings, but the filters will pickup directly address mail...)
>
> On Wed, 7 Mar 2001, Charles Daminato wrote:
>
> > You can use a variation of a script found here:
> >
> > http://www.opensrs.org/archives/dev-list/0102/0156.html
> >
> > or an older version (not using curl) here:
> >
> > http://www.opensrs.org/archives/discuss-list/0012/0147.html
> >
> > The second script doesn't work as is - it'll take some tweaking, but the
> > core pieces are there.
> >
> > Charles Daminato
> > Product Manager (ccTLDs)
> > Tucows Inc. - [EMAIL PROTECTED]
> >
> > > -----Original Message-----
> > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On
> > > Behalf Of Guy Baconnière
> > > Sent: March 7, 2001 12:33 PM
> > > To: [EMAIL PROTECTED]
> > > Subject: How to retrive a list of all our customers domains
> > >
> > > Hi,
> > >
> > > We want to import into our SQL database all contacts fields
> and billing
> > > informations of all customers domains registered.
> > >
> > > We cannot have access to the list of all domains by the API without
> > > having the login/password of each customer. We need a "reseller" API
> > > that can give us access to all informations available manually on
> > > the reseller web
> > > interface (http://resellers.opensrs.net/) like contacts fields,
> > > billing status, etc.
> > >
> > > Is there any way to do that directly by perl script without
> > > having our customer
> > > user and password ?
>
>
