I think Daniel makes a good case, although Seva is right that defining a new
variable is
less ambiguous, as you would probably not want to rely on the actual contents
of a
variable that is changing dynamically in a complex environment.
I firmly believe that the biggest mistake people make in computing is believing
in
determinism. Cfengine is all about coping with the unexpected, so we should
strive for
methods that are fault tolerant. Like Nakarin, I would like to see a specific
case where
this would be used to good effect.
These are interesting suggestions that can be made formally convergent. I'll
give it some
thought.
M
Daniel Dehennin wrote:
> Vasiliy G Tolstov <v.tols...@selfip.ru> writes:
>
>> is there are any plans to add support to append value to slist ?
>>
>> For example:
>>
>> "a" slist => { "1", "2" };
>> "a" slist => { "$(a)", "3" };
>
> +1
>
> I think the '"name" *list => {}' construct could be seen as: promise to
> have elements listed in '{}' in the list named 'name'.
>
> My point is:
> - a list construct define a list variable if it's not already defined
> - then add the listed elements if they are not presents
>
> Maybe an 'append=>' and 'prepend=>' could be used instead of overloading
> '*list=>' and changing its meaning?
>
> Regards.
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Help-cfengine mailing list
> Help-cfengine@cfengine.org
> https://cfengine.org/mailman/listinfo/help-cfengine
--
Mark Burgess
-------------------------------------------------
Professor of Network and System Administration
Oslo University College, Norway
Personal Web: http://www.iu.hio.no/~mark
Office Telf : +47 22453272
-------------------------------------------------
_______________________________________________
Help-cfengine mailing list
Help-cfengine@cfengine.org
https://cfengine.org/mailman/listinfo/help-cfengine
