Have you tried split function instead of any2array?
El 18/09/15 a las 16:59, Sean escribió:
My woes were definitely due to the data in the parameters being treated
as strings throughout. Basically, for whatever reason even when using
any2array(), the issue came down to all the desired individual elements
were being treated as a single string that became element 0 of an array...
Effectively:
$whitelist = [ 'foo::a, foo::b, foo::c, foo::d, foo::e', ]
$blacklist = [ 'foo::c, foo::e', ]
It's possible that whitelist was not, but I can confirm that blacklist
(sent to puppet via ENC) was indeed a string, that any2array converted
to a single element array. I have coerced the datatype in the ENC using
Foreman's SmartClass parameter override, and am now getting the correct
results.
I also converted the inline template into a custom function to avoid
having to use any2array.
On Thursday, September 17, 2015 at 4:16:16 PM UTC-4, Sean wrote:
Hi Peter,
Thanks for that! I must have lost the = in translation from the old
post, I appreciate you 2nd set of eyes. Now that I'm passed that
and have data in the result, I've only graduated to having incorrect
results. I had used the any2array function to ensure the array
status of the template output and the blacklist param. The
blacklist param will be delivered from and ENC (specifically
Foreman), and while it should work, I've had issues in the past with
arrays and hashes being taken as strings. That said, here's what I
see now:
$whitelist = [ foo::a, foo::b, foo::c, foo::d, foo::e ]
$blacklist = [ foo::c, foo::e ]
< apply corrected code, sending notifies for $whitelist, $blacklist,
and $include_list >
The notify outputs show that $include_list is identical to
$whitelist. If I actually attempt the include $include_list
statement (which is usually commented) I get duplicate resource
definition errors since the blacklisted classes create resources
like package { 'apache': ensure => absent } and the node's purpose
might include being a webserver so we have another class which
manages the package resource for apache.
I've experimented with the parameter from the ENC, and how I define
whitelist in params.pp. I have tried several experiments with
quoting the array elements, with singles, with doubles, not quoting
the elements, and even defining them as strings, and allowing
any2array to convert them before the template. The outputs are
different each time, but the result is the same. The blacklist
items still appear in the end result.
FWIW, I'm running puppet 3.6.2 on the agent and 3.8.2 on the master,
I assume you refer to current as the version 4 tree.
On Thursday, September 17, 2015 at 3:15:37 PM UTC-4, Peter Huene wrote:
Hi Sean,
On Thu, Sep 17, 2015 at 12:01 PM, Sean <[email protected]> wrote:
Hello,
I have been working on trying to drive an include statement
with an array parameter.
The idea looks like so (in pseudo code) :
|
classfoo($whitelist =$::foo::params::whitelist,$blacklist =[],){
$include_list =inline_template("<% @whitelist -
@blacklist %>")
validate_array($include_list)
include $include_list
}
|
I picked up that inline template from a message in the group
dating back to 2011.
The template is missing a '=' character to write the result,
like so:
<%= @whitelist - @blacklist %>
However, this will return a string-ified version of the array
and is not what you want as include will treat it as a single
class name.
In the current version of the Puppet language, two arrays can
simply be subtracted from one another:
$whitelist = [foo, bar, baz]
$blacklist = [bar]
include $whitelist - $blacklist
This would include "foo" and "baz", but not "bar".
Basically, class foo has a ton of subclasses who's names
populate the default value for $whitelist. Normally all
these get applied, but I need to allow for deviations for
specific cases, thus the blacklist.
The issue seems to be that no matter what I put in whitelist
or blacklist, include_list is always empty. What am I
missing? I have also tried writing a Custom Function to do
the same as the inline template but the result was not
anymore successful, though the debugging was more difficult.
Thanks for your thoughts on this!
--
You received this message because you are subscribed to the
Google Groups "Puppet Users" group.
To unsubscribe from this group and stop receiving emails
from it, send an email to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/puppet-users/6ac9c22b-f5fc-409b-a75e-7d286862220b%40googlegroups.com
<https://groups.google.com/d/msgid/puppet-users/6ac9c22b-f5fc-409b-a75e-7d286862220b%40googlegroups.com?utm_medium=email&utm_source=footer>.
For more options, visit https://groups.google.com/d/optout
<https://groups.google.com/d/optout>.
--
--
Peter Huene
Software Engineer, Puppet Labs
Puppet Open Source Team
-------------------------------------------
*
*
*PuppetConf 2015 <http://2015.puppetconf.com/>
** is right around the corner! Join us October 5-9 in Portland,
OR. **Register now <https://puppetconf2015.eventbrite.com/>
**.*
--
You received this message because you are subscribed to the Google
Groups "Puppet Users" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to [email protected]
<mailto:[email protected]>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/puppet-users/8c70cabb-71c2-40a4-9081-ac0e10eb4ced%40googlegroups.com
<https://groups.google.com/d/msgid/puppet-users/8c70cabb-71c2-40a4-9081-ac0e10eb4ced%40googlegroups.com?utm_medium=email&utm_source=footer>.
For more options, visit https://groups.google.com/d/optout.
--
Angel L. Mateo Martínez
Sección de Telemática
Área de Tecnologías de la Información
y las Comunicaciones Aplicadas (ATICA)
http://www.um.es/atica
Tfo: 868887590
Fax: 868888337
--
You received this message because you are subscribed to the Google Groups "Puppet
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/puppet-users/55FFA6E2.4080901%40um.es.
For more options, visit https://groups.google.com/d/optout.
