Ian Holsman wrote:
>
> Hi.
>
> I was looking through the code while I was attempting to patch the glib
> hash functions into the 1.x codebase, and was thinking that the services
> hash table wasn't ideal.
>
>
> wouldn't it be better if there was a linked list of services for each
> host (instead of either a global linked list/hash on host/service name)
>
>
> ie.. in the host structure you would have something like
>
> host_struct {
> ...
> service *service_head;
> ..
> }
>
> and in the service_struct something like
>
> service_struct {
> ..
> service *service_next;
> host*host
> }
>
I'm currently working on creating proper configuration structs for
nagios. If it doesn't get included in the main branch, I'll open up a
new project and use it there.
> this was you could navigate around instead of having to go through all
> the lists the matching host/service.
> I think the above would work better than having a hash on the combined
> service/host names. and you could use this in a lot of places in the
> nagios code.
>
Indeed. Much isn't properly done in the code, but I'm guessing that is
mostly for legacy reasons.
> oh.. a humble request..
> can someone please run indent on the code in CVS and check that in.
>
I agree completely. However, since Ethan is the main developer, his
coding standards are law in this project.
I suggest you get a nice .indent.pro (mine is attached) and run it
recursively on the nagios code as such;
cd nagios
for i in `find -type d` ; do indent $i/*.c $i/*.h ; done
indent(1) for those who do not know is a GNU program and can be
downloaded from http://www.gnu.org
It's most likely available as prepackaged binaries for your system.
>
> Regards
> Ian.
>
> --
> Ian Holsman
> blog: http://blog.holsman.net
> PH: 03-9857-3742 (oz)
--
Mvh / Best regards
Sourcerer / Andreas Ericsson
OP5 AB
+46 (0)733 709032
[email protected]
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]