Thanks to David and Geert for their pointers and explanation. It was
indeed simply the xdmp:host-name function I was looking for (doh!).


cheers,
Jakob.



On Fri, Mar 4, 2011 at 15:46, Geert Josten <[email protected]> wrote:
> Hi Jakob,
>
> There is of course xdmp:host and xdmp:host-name, but that won't return the 
> host-in-the-url entered by the end-user. The host-name could even be very 
> unusefull to the end-user, as it is the value entered in the admin interface, 
> and doesn't need to have any reference to reality as far as I know. You can 
> change it to anything you like at least..
>
> If you just want to show the host name as known to the end user, you could 
> alternatively use javascript within the html returned to the user.
>
> Clustering might be involved indeed, but I expect xdmp:host to return the 
> host ID of the evaluator node, e.g. the host that received the actual 
> request, not all other hosts that might be involved in data retrieval. If you 
> have load balancing, then different host names would be shown..
>
> Kind regards,
> Geert
>
> -----Oorspronkelijk bericht-----
> Van: [email protected] 
> [mailto:[email protected]] Namens Jakob Fix
> Verzonden: vrijdag 4 maart 2011 15:30
> Aan: General Mark Logic Developer Discussion
> Onderwerp: [MarkLogic Dev General] how to retrieve the hostname, easily
>
> Hi,
>
> I'm looking for a not-so-complicated way to retrieve the host name
> (i.e. the name that is displayed on top of each page in the MarkLogic
> admin interface).
> Why is there a xdmp:get-request-path, xdmp:get-request-protocol,
> xdmp:get-request-url, but no xdmp:get-host-name?
>
> Of course, I've ploughed through the Admin/ code, and finally found
> the function get-session-host-node() in session.xqy, but it also
> depends on an $hosts.xml variable which in turn uses yet another
> library.  In the end it's very involved if all you want to do is to
> display the hostname the current application is running on (I guess
> the fact that it's not easy probably has its reasons in the fact that
> more complicated scenarios involving clustering exist that need to be
> accounted for).
>
> cheers,
> Jakob.
> _______________________________________________
> General mailing list
> [email protected]
> http://developer.marklogic.com/mailman/listinfo/general
> _______________________________________________
> General mailing list
> [email protected]
> http://developer.marklogic.com/mailman/listinfo/general
>
_______________________________________________
General mailing list
[email protected]
http://developer.marklogic.com/mailman/listinfo/general

Reply via email to