ray = ['a','b','c']
$var = myfunction('first arg', $array)
$empty_array = []
$other_var = myfunction($empty_array)
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
You received this message becaus
On 22 Apr 2011, at 11:26 AM, Jacob Helwig wrote:
> Thanks for doing this, Ian! Josh and I have merged the following additional
> tests into the patch.
That looks good; thanks!
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
You receive
promotion to stable.
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
You received this message because you are subscribed to the Google Groups
"Puppet Developers" group.
To post to this group, send email to puppet-dev@googlegroups.com.
To un
ysconfdir}/tmpfiles.d
Typo (repeated mkdir)?
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
You received this message because you are subscribed to the Google Groups
"Puppet Developers" group.
To post to this group, send email to puppet
ill only autorequire its
parent directory, if the latter is managed. If my patch attached to #6368 makes
it into Statler, then the file will autorequire its nearest managed ancestor
directory, but only the nearest.
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford
of the
relationship graph.
Signed-off-by: Ian Ward Comfort
---
On 18 Feb 2011, at 2:20 PM, Luke Kanies wrote:
> autorequire is a bit of a weird beast, so I know why you're having problems.
>
> My code was just about changing the behavior of existing autorequire, rather
> than addin
On 18 Feb 2011, at 2:08 PM, Luke Kanies wrote:
> On Feb 18, 2011, at 1:20 PM, Ian Ward Comfort wrote:
>> On 18 Feb 2011, at 12:29 PM, Luke Kanies wrote:
>>> It's actually pretty straightforward to do the trimming, I think - we
>>> currently just return the par
On 18 Feb 2011, at 12:29 PM, Luke Kanies wrote:
> On Feb 18, 2011, at 11:37 AM, Ian Ward Comfort wrote:
>> Heh, right. I was trying to think of a clean way to give types a way to say,
>> "autorequire the first of these dependencies that's found in the catalog,
>>
mething it would almost certainly break the existing
API.
Shall I code up a weather-balloon patch for autorequirefirst?
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
You received this message because you are subscribed to the Google Groups
"Puppet D
On 18 Feb 2011, at 11:19 AM, Daniel Pittman wrote:
> On Fri, Feb 18, 2011 at 10:48, Ian Ward Comfort wrote:
>> Yeah, requiring just the first managed parent, as suggested, would be
>> sufficient. I wasn't sure how to implement this, though, from a quick glance
>> t
On 18 Feb 2011, at 10:34 AM, Daniel Pittman wrote:
> On Feb 18, 2011 10:30 AM, "Luke Kanies" wrote:
>> On Feb 17, 2011, at 10:48 PM, Ian Ward Comfort wrote:
>>>
>>> I've just put an RFC feature request into the tracker, and I'm curious what
>>
n ancestor directory are implicitly added
> to a directory node, and nothing more?
>
> You do not expect this to implicitly create directories on disk, or anything
> that would be externally observable, just to influence dependency execution
> order?
Correct on all counts.
--
com/issues/6368
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
You received this message because you are subscribed to the Google Groups
"Puppet Developers" group.
To post to this group, send email to puppet-dev@googlegroups.com.
To unsubscr
email to puppet-dev@googlegroups.com.
> To unsubscribe from this group, send email to
> puppet-dev+unsubscr...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/puppet-dev?hl=en.
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanf
On 3 Feb 2011, at 1:38 PM, Luke Kanies wrote:
> On Feb 3, 2011, at 12:12 PM, Ian Ward Comfort wrote:
>> On 3 Feb 2011, at 9:59 AM, Luke Kanies wrote:
>>> On Feb 3, 2011, at 1:40 AM, Thomas Bellman wrote:
>>>> What about the already existing syntax:
>
lysis without compiling all of the
> classes, and even then you wouldn't know if a given class always has that
> list of dependencies, or only for that one host.
>
> That make sense?
The require function also includes the named classes without parameters,
correct?
--
Ian
On 1 Feb 2011, at 3:24 PM, Nigel Kersten wrote:
> On Tue, Feb 1, 2011 at 2:30 PM, Ian Ward Comfort
> wrote:
>> Since the longhand syntax is already permitted, providing "Self" as
>> shorthand doesn't seem too dangerous -- modulo keyword namespace concerns, I
as about parameterized
classes and how I'd like to use them are still purely theoretical.) Since the
longhand syntax is already permitted, providing "Self" as shorthand doesn't
seem too dangerous -- modulo keyword namespace concerns, I guess.
I'm curious to hear Nige
or some
other class, which is itself a descendant of defaults). This saves typing,
compacts our code a *lot*, and somewhat reduces errors due to forgetfulness.
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
You received this message because you are
layed with them
enough, since we're still a little ways off from migrating to 2.6.1+, but I
agree that efforts should be focused on making those work sensibly.
> If a class includes another class and the child needs access to a variable in
> the default, if we must allow this beha
et Dashboard release.
>
> I've created a ticket for tracking this task at
> http://projects.puppetlabs.com/issues/4524 -- can you please add yourself as
> a watcher so we can continue the discussion there?
Sure, done.
--
Ian Ward Comfort
Systems Team Lead, Academic Computing S
Now that the sidebar shows "Ever succeeded" and "Ever failed" statistics, most
page loads generate SQL of the form:
SELECT `nodes`.* FROM `nodes` INNER JOIN `reports`
ON reports.node_id = nodes.id WHERE (reports.success = 1)
GROUP BY nodes.id;
Adjusting the index to reports(node_id,success)
erformance
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
app/controllers/pages_controller.rb|1 -
.../20100811204545_add_success_to_report_index.rb | 11 +++
db/schema.rb |4 ++-
The @reports variable is not used, and (in at least some versions of Rails)
the :order => 'time ASC' used to generate it causes an expensive database
query which sorts by time first in ascending, then in descending order (due to
the ordering on Report's default_scope). Remove this query entirely to
On 12 Mar 2010, at 7:20 PM, Rein Henrichs wrote:
Ian, thanks for another useful patch. I've already pulled this into
master.
No problem. Glad I won't have to maintain 'em myself.
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
--
Yo
Relativizing URLs in JavaScript is slightly trickier than CSS. The strategy
here is to save our relative_url_root in a global JS variable in the topmost
application view template, then use that variable in all our included scripts.
---
app/views/layouts/application.html.haml |2 ++
public/java
First, the low-hanging fruit: change all (two) absolute URLs in the CSS to use
relative paths. This restores the Dashboard logo when the app is run at a
non-root URL.
---
public/stylesheets/application.css |2 +-
public/stylesheets/fcbkcomplete.css |2 +-
2 files changed, 2 insertions(+),
node.js
autocompletion code is used.)
These patches can also be pulled from:
git://github.com/icomfort/puppet-dashboard.git relative-url-root
--
Ian Ward Comfort
Systems Team Lead, Academic Computing Services, Stanford University
app/views/layouts/application.html.haml |2 ++
then I'm fine with this.
On Mar 11, 2010, at 5:35 PM, Ian Ward Comfort wrote:
A dependency on rake >= 0.8.4 was added to the Dashboard to fix
issue #2897,
caused by a missing reenable method on Rake::Task. Rake's release
notes [1]
indicate that this method was added in 0.8.2. Th
On 11 Mar 2010, at 5:35 PM, Ian Ward Comfort wrote:
A dependency on rake >= 0.8.4 was added to the Dashboard to fix
issue #2897,
caused by a missing reenable method on Rake::Task. Rake's release
notes [1]
indicate that this method was added in 0.8.2. The Rails frozen in
vendor/
requ
A dependency on rake >= 0.8.4 was added to the Dashboard to fix issue #2897,
caused by a missing reenable method on Rake::Task. Rake's release notes [1]
indicate that this method was added in 0.8.2. The Rails frozen in vendor/
require rake >= 0.8.3. Since EPEL currently ships Rake 0.8.3, it would b
r.class.stubs(:curl)
> +@provider.class.stubs(:installpkg)
> +File.stubs(:exist?).returns true
> +File.expects(:unlink).with("/tmp/
> #...@resource[:name]}").once
> +@provider.install
> +end
> end
> end
> --
, but it can choose to let the client decide.
> That seems like the best overall approach, but it still leaves
> unsolved the overall complexity.
I'll second (or third) this suggestion -- it's exactly how I use
environments in 0.24.8 today.
--
Ian Ward Comfort
Systems
33 matches
Mail list logo
