Issue #8040 has been updated by Melissa Stone.
Released in Puppet 3.4.0-rc1
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-100406
* Author: Jeff McCune
*
Issue #8040 has been updated by Melissa Stone.
Released in Puppet 3.4.0-rc1
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-100487
* Author: Jeff McCune
*
Issue #8040 has been updated by Melissa Stone.
Released in Puppet 3.4.0-rc1
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-100590
* Author: Jeff McCune
*
Issue #8040 has been updated by Melissa Stone.
Released in Puppet 3.4.0-rc1
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-100690
* Author: Jeff McCune
*
Issue #8040 has been updated by Melissa Stone.
Released in Puppet 3.4.0-rc1
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-100838
* Author: Jeff McCune
*
Issue #8040 has been updated by Patrick Carlisle.
Target version changed from 3.x to 3.4.0
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-100161
* Author:
Issue #8040 has been updated by Andre Nathan.
In code like this:
node foo {
class { 'base_server':
...
}
class { 'web_server':
...
}
}
If I want to make sure all code in base_server runs before the code in
web_server, what's the preferred way
Issue #8040 has been updated by John Bollinger.
Andre Nathan wrote:
In code like this:
node foo {
class { 'base_server':
...
}
class { 'web_server':
...
}
}
If I want to make sure all code in base_server runs before the code in
Issue #8040 has been updated by Roman Esq.
Is it expected that the fix will be included in 2.6.19 version? Or is there any
other possibility to use contain fix with 2.6 branch?
Bug #8040: Classes should be able to contain other classes to provide a
Issue #8040 has been updated by eric sorenson.
Roman Esq wrote:
Is it expected that the fix will be included in 2.6.19 version? Or is there
any other possibility to use contain fix with 2.6 branch?
No guarantees whatsoever, because 2.6 is past end of life. But the 'contain'
function is
Issue #8040 has been updated by Rob Terhaar.
Two years to discuss it, and two weeks to fix it ;)
I am really happy this is finally fixed. Thank you everyone!
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
Issue #8040 has been updated by John Bollinger.
Ruslan Lutsenko wrote:
Will all classes have optional boolean attribute contained, or
parameterized classes are not meant to be contained?
It is already accounted for:
class container {
class { 'contained':
param = value
Issue #8040 has been updated by Luke Kanies.
(As sent to the list.)
As mentioned in my other email, the solution to this problem should not in any
way require changes to containment semantics, and certainly shouldn't require
class evaluation to indicate class containment. As I said, it used
Issue #8040 has been updated by John Bollinger.
Nick Fagerlund wrote:
A proposed solution has come up a few times in convo w/ Dan Bode, and I want
to make sure it's captured here:
Why not just say that resource-like declarations (`class {'myclass':}`)
always cause containment? This
Issue #8040 has been updated by Andre Nathan.
Completely agree with John. A contains built-in function would avoid breaking
current semantics and allow us to have this very much needed feature sooner.
Bug #8040: Classes should be able to contain other
Issue #8040 has been updated by Patrick Carlisle.
Assignee changed from Charlie Sharpsteen to Patrick Carlisle
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
Issue #8040 has been updated by eric sorenson.
Support Urls deleted (https://support.puppetlabs.com/tickets/772
https://support.puppetlabs.com/tickets/1659)
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
Issue #8040 has been updated by Dennis Jacobfeuerborn.
Support Urls deleted (https://support.puppetlabs.com/tickets/772
https://support.puppetlabs.com/tickets/1659)
This bug is two years old now so is this actually looked into?
AFAIK most modules on Puppet Forge don't actually use the proposed
Issue #8040 has been updated by Andre Nathan.
Support Urls deleted (https://support.puppetlabs.com/tickets/772
https://support.puppetlabs.com/tickets/1659)
Completely agree with Dennis. It's a pain to deal with this, it should be given
a higher priority.
Issue #8040 has been updated by Markus Uckelmann.
Support Urls deleted (https://support.puppetlabs.com/tickets/772
https://support.puppetlabs.com/tickets/1659)
I just stumbled over this bug and it took me two days to figure it out. Could
you please solve this issue?
Issue #8040 has been updated by Charlie Sharpsteen.
Keywords changed from anchor containment contain graph modules module
self-contained dependency reuse usability forge backlog to anchor containment
contain graph modules module self-contained dependency reuse usability forge
backlog customer
Issue #8040 has been updated by Charlie Sharpsteen.
Assignee set to Charlie Sharpsteen
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-87335
* Author: Jeff
Issue #8040 has been updated by eric sorenson.
Priority changed from High to Immediate
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-82528
Author: Jeff
Issue #8040 has been updated by eric sorenson.
Priority changed from Normal to High
Target version set to 3.x
Keywords changed from anchor containment contain graph modules module
self-contained dependency reuse usability forge to anchor containment contain
graph modules module self-contained
Issue #8040 has been updated by Jeff Goldschrafe.
Nick Fagerlund wrote:
(This would almost definitely reveal currently hidden dependency cycles in
peoples' code. That means we should implement it in a major version with a
temporary pref to turn the behavior off, just so people can upgrade
Issue #8040 has been updated by Nick Fagerlund.
A proposed solution has come up a few times in convo w/ Dan Bode, and I want to
make sure it's captured here:
Why not just say that resource-like declarations (`class {'myclass':}`) always
cause containment? This should be possible without
Issue #8040 has been updated by John Bollinger.
Chris Price wrote:
We've had some discussions around this lately, so I'll try to update the
ticket for posterity:
* There seems to be general consensus that we should be able to get by with
only two types of inclusions; a simple one-level
Issue #8040 has been updated by Chris Price.
John, thanks for the comments. Rest assured that backwards compatibility is a
huge concern for us. The last comment that you replied to was speaking more to
the idea of how would the language look if we were designing it from scratch?
It is
Issue #8040 has been updated by Chris Price.
Assignee deleted (Chris Price)
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
https://projects.puppetlabs.com/issues/8040#change-65076
Author: Jeff McCune
Issue #8040 has been updated by Chris Price.
We've had some discussions around this lately, so I'll try to update the ticket
for posterity:
* There seems to be general consensus that we should be able to get by with
only two types of inclusions; a simple one-level dependency--ala existing
Issue #8040 has been updated by Chris Price.
Assignee changed from Daniel Pittman to Chris Price
Andy and I are both working on this. I'm currently scouting for the original
tickets where the whits-related changes were introduced, and will link them to
this ticket accordingly.
#3382 seems
Issue #8040 has been updated by Chris Price.
Also related: ticket #5200 commit:
https://github.com/puppetlabs/puppet/commit/2a6c6cb8fabf82d2f2127c90db670c1a856427c5
Bug #8040: Classes should be able to contain other classes to provide a self
Issue #8040 has been updated by Chris Price.
Jeff--starting to look at this and would love to pick your brain about it for a
few minutes sometime soon. My most pressing question at the moment is this:
Your updated graph shows a relationship between Anchor[java::*] and
Issue #8040 has been updated by Jeff McCune.
Chris Price wrote:
Jeff--starting to look at this and would love to pick your brain about it for
a few minutes sometime soon. My most pressing question at the moment is this:
Sure, come on over to my desk anytime. Even if I have my headphones
On Fri, Mar 9, 2012 at 9:00 AM, tick...@puppetlabs.com wrote:
Issue #8040 has been updated by Nigel Kersten.
Can someone file the issue around a contains function then?
Would that function automatically include a class if it hasn’t been
declared? Or just specify the relationship?
IIRC,
Issue #8040 has been updated by Jeff McCune.
Nigel Kersten wrote:
Can someone file the issue around a contains function then?
Would that function automatically include a class if it hasn't been declared?
Or just specify the relationship?
Filed as #13045
I commented on this ticket via
Issue #8040 has been updated by Nick Lewis.
I believe the decision we came to was to add a `contains` function to the
stdlib module, but I think the actual work slipped through the cracks.
Bug #8040: Classes should be able to contain other classes to
Issue #8040 has been updated by Nan Liu.
A decision regarding class containment will also impacts stage behavior. See
#8263
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
Issue #8040 has been updated by Daniel Pittman.
Following Daniel’s suggestion I’m setting to Needs Decision and assigning to
Nigel for placement on the road map. I’d like to see this fixed in 2.7.x but
I think it can certainly wait until Telly. I think it must be solved in
Telly though.
To
Issue #8040 has been updated by Garrett Honeycutt.
This is a very clever solution to a problem that adds extra complexity to one's
manifests. Puppet already has ways to describe dependency relationships. While
this appears to work, I do not think it is a pattern that we should propagate.
Issue #8040 has been updated by Jeff McCune.
Garrett Honeycutt wrote:
This is a very clever solution to a problem that adds extra complexity to
one's manifests. Puppet already has ways to describe dependency
relationships. While this appears to work, I do not think it is a pattern
that
Issue #8040 has been updated by Jeff McCune.
File Anchor_no_relationships.png added
File Anchor_both_anchors.png added
Bug #8040: Classes should be able to contain other classes to provide a self
contained module
Issue #8040 has been updated by Jeff McCune.
# Graph View #
To visualize the problem. Please consider the graph produced when the work
around is not used.
pre
class activemq {
class { 'activemq::pacakge': }
-
class { 'activemq::config': }
-
class { 'activemq::service': }
}
/pre
43 matches
Mail list logo
