On Fri, Apr 28, 2006 at 03:05:04PM +0100 or thereabouts, Kostas Georgiou wrote:
> On Fri, Apr 28, 2006 at 01:34:21PM +0100, Steve Traylen wrote:
>
> > Checking through the output it looks okay other than possibly lowering
> > the case of the attributes which may in fact be okay as far as ldif is
> > concerned but I expect something will blow up somewhere. But I know
> > you said it was alpha version. Looking through it it looks like you have
> > lowered the case on purpose?
>
> Since ldif attributes are case insensitive there is little point to
> complicate the data structures to store the original value (easilly
> done though). In my opinion if any grid tool depends on the case of
> ldap attributes it is horribly broken and should be fixed.
You are of course correct , i guess we could try it. I suspect it will
just break some monitoring rather than any middleware.
>
> On a different note in the original script if an attribute doesn't
> appear in the static list the dynamic value is ignored is this the
> "right" behaviour? I would have thought that missing values in the
> dynamic value are substituted by the static ones. Does anyone have
> an idea on what the correct way to handle this is?
If there was some motivation for doing it that way then it does
force you to set sensible defaults for any dynamic value that can't
be retrieved for some transient reason.
Steve
>
> Kostas
--
Steve Traylen
[log in to unmask]
http://www.gridpp.ac.uk/
|