Dear stdlib''ers...
I have just discovered the wonders of the parser functions, and got
impressed with the tens of functions that come with stdlib. First things
first... good work!!! Thanks!!
And now the issue. It seems like the writer of the range() function did
not think about ranges with more than one digit that need leading zeros
in the first items, like "01..99", when you usually want to have 01,
02,
and so on. Ruby allows you to do ("01".."99") and that will
do the right
thing, but the range() function provided with stdlib does some type
conversion (detects if it''s a number, and changes the type to integer)
which converts "01" to 1 breaking this possibility. I tried to submit
a
bug report, but I just can list the open ones, can''t make one myself.
Is
this intentional? How do I properly address this request?
So, I tried to change that myself, but no matter what I do to the
range.rb file, the changes are not picked up by the node. Do I have to
do something to force a reload of the file? This runs in the server, right?
What I did then was to create a range_custom() which is a copy of the
former, but without the type conversion. I tried that and it works like
a charm.
Thanks!
Pablo
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
On Thu, Mar 29, 2012 at 8:48 AM, Pablo Fernandez <pablo.fernandez@cscs.ch>wrote:> Dear stdlib''ers... > > I have just discovered the wonders of the parser functions, and got > impressed with the tens of functions that come with stdlib. First things > first... good work!!! Thanks!! > > And now the issue. It seems like the writer of the range() function did > not think about ranges with more than one digit that need leading zeros in > the first items, like "01..99", when you usually want to have 01, 02, and > so on. Ruby allows you to do ("01".."99") and that will do the right thing, > but the range() function provided with stdlib does some type conversion > (detects if it''s a number, and changes the type to integer) which converts > "01" to 1 breaking this possibility.Hey Pablo, Just a question - do you have a NEED for it to be [01...99] with the leading zero? If I understand correctly, the end result is the same, as the range function will cast 01 to 1 and not BREAK because you pass it 01 - correct? Maybe I''m missing why you would need to use 01 explicitly?> I tried to submit a bug report, but I just can list the open ones, can''t > make one myself. Is this intentional? How do I properly address this > request? >No problem. We''ve turned off Github Issues because we use our central Redmine server for bug tracking --> http://projects.puppetlabs.com/projects/modules/issues Please feel free to file any module tickets here!> > So, I tried to change that myself, but no matter what I do to the range.rb > file, the changes are not picked up by the node. Do I have to do something > to force a reload of the file? This runs in the server, right? > > What I did then was to create a range_custom() which is a copy of the > former, but without the type conversion. I tried that and it works like a > charm. > > Thanks! > Pablo > > -- > You received this message because you are subscribed to the Google Groups > "Puppet Users" group. > To post to this group, send email to puppet-users@googlegroups.com. > To unsubscribe from this group, send email to puppet-users+unsubscribe@** > googlegroups.com <puppet-users%2Bunsubscribe@googlegroups.com>. > For more options, visit this group at http://groups.google.com/** > group/puppet-users?hl=en<http://groups.google.com/group/puppet-users?hl=en> > . > >-- Gary Larizza Professional Services Engineer Puppet Labs -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
On Thu, Mar 29, 2012 at 8:57 AM, Gary Larizza <gary@puppetlabs.com> wrote:> > I tried to submit a bug report, but I just can list the open ones, can''t >> make one myself. Is this intentional? How do I properly address this >> request? >> > > No problem. We''ve turned off Github Issues because we use our central > Redmine server for bug tracking --> > http://projects.puppetlabs.com/projects/modules/issues Please feel free > to file any module tickets here! >Gary, Pablo, I actually just created a dedicated project [1] for the Standard Library. Ken Barber and I talked about this a couple of weeks ago; we''re of the opinion stdlib has grown enough to warrant a specific place to file issues. The main motivation is to enable everyone to easily target specific semantic versions in the issue tracker. For example, "stdlib misbehaves in 2.1.x but not in 2.2.x" This will allow us to target 2.1.x in without stdlib dominating all of the versions in the modules project. Pablo, if you file an issue, please file it in either project. I''m going to spend much of today ticket gardening and moving the stdlib tickets to the new project. The new project is set to "self subscribe" so please feel free to join the project in your account settings. [1] http://projects.puppetlabs.com/projects/stdlib Cheers, -Jeff -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Hi,> And now the issue. It seems like the writer of the range() > function did not think about ranges with more than one digit that > need leading zeros in the first items, like "01..99", when you > usually want to have 01, 02, and so on. Ruby allows you to do > ("01".."99") and that will do the right thing, but the range() > function provided with stdlib does some type conversion (detects > if it''s a number, and changes the type to integer) which converts > "01" to 1 breaking this possibility. > > > Hey Pablo, > > Just a question - do you have a NEED for it to be [01...99] with the > leading zero? If I understand correctly, the end result is the same, > as the range function will cast 01 to 1 and not BREAK because you pass > it 01 - correct? Maybe I''m missing why you would need to use 01 > explicitly?Yes, some of our machines are like "node01-node50", that''s why I tried to do that. The thing is that Ruby allows you to do that if you say ("00".."50") instead of (0..50). The only difference is that the array will be of strings, and the second of integers. Maybe an example illustrates better: irb(main):008:0> ("08".."12").to_a => ["08", "09", "10", "11", "12"] irb(main):010:0> (8..12).to_a => [8, 9, 10, 11, 12] But then, the range() function in stdlib, makes a type conversion, that breaks this possibility. I don''t understand why do you do that conversion... the function is just an interface to the "range" capability of Ruby, right? I think it should allow the programmer to choose the type he wants.> No problem. We''ve turned off Github Issues because we use our central > Redmine server for bug tracking --> > http://projects.puppetlabs.com/projects/modules/issues Please feel > free to file any module tickets here!I saw that, but I can only see the tickets, cannot open one myself. Do I have to create myself an account? How?> So, I tried to change that myself, but no matter what I do to the > range.rb file, the changes are not picked up by the node. Do I have to > do something to force a reload of the file? This runs in the server, > right?Do you know if, after editing a .rb file, I have to force some kind of reload to pick up the changes? Thanks! Pablo -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Gotcha, thanks! On 03/29/2012 06:23 PM, Jeff McCune wrote:> On Thu, Mar 29, 2012 at 8:57 AM, Gary Larizza <gary@puppetlabs.com > <mailto:gary@puppetlabs.com>> wrote: > > > I tried to submit a bug report, but I just can list the open > ones, can''t make one myself. Is this intentional? How do I > properly address this request? > > > No problem. We''ve turned off Github Issues because we use our > central Redmine server for bug tracking --> > http://projects.puppetlabs.com/projects/modules/issues Please > feel free to file any module tickets here! > > > Gary, Pablo, > > I actually just created a dedicated project [1] for the Standard > Library. Ken Barber and I talked about this a couple of weeks ago; > we''re of the opinion stdlib has grown enough to warrant a specific > place to file issues. The main motivation is to enable everyone to > easily target specific semantic versions in the issue tracker. For > example, "stdlib misbehaves in 2.1.x but not in 2.2.x" This will > allow us to target 2.1.x in without stdlib dominating all of the > versions in the modules project. > > Pablo, if you file an issue, please file it in either project. I''m > going to spend much of today ticket gardening and moving the stdlib > tickets to the new project. > > The new project is set to "self subscribe" so please feel free to join > the project in your account settings. > > [1] http://projects.puppetlabs.com/projects/stdlib > > Cheers, > -Jeff > -- > You received this message because you are subscribed to the Google > Groups "Puppet Users" group. > To post to this group, send email to puppet-users@googlegroups.com. > To unsubscribe from this group, send email to > puppet-users+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/puppet-users?hl=en.-- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
I needed this just yesterday, actually, for NFS mounts.
$PO_0_9 = prefix(range("0","9"),''/PO_0'')
$PO_10_31 =
prefix(range("10","31"),''/PO_'')
$PO_32_76 =
prefix(range("32","76"),''/PO_'')
nfs::po_mounts_50{ $PO_0_9:
options => ''soft,bg,tcp'',
}
nfs::po_mounts_50{ $PO_10_31:
options => ''soft,bg,tcp'',
}
nfs::po_mounts_55{ $PO_32_76:
options => ''soft,bg,tcp'',
}
I know how ugly this is, but I had to split the range in two just to get
the leading 0.
On Thu, Mar 29, 2012 at 11:57 AM, Gary Larizza <gary@puppetlabs.com>
wrote:
>
>
> On Thu, Mar 29, 2012 at 8:48 AM, Pablo Fernandez
<pablo.fernandez@cscs.ch>wrote:
>
>> Dear stdlib''ers...
>>
>> I have just discovered the wonders of the parser functions, and got
>> impressed with the tens of functions that come with stdlib. First
things
>> first... good work!!! Thanks!!
>>
>> And now the issue. It seems like the writer of the range() function did
>> not think about ranges with more than one digit that need leading zeros
in
>> the first items, like "01..99", when you usually want to have
01, 02, and
>> so on. Ruby allows you to do ("01".."99") and that
will do the right thing,
>> but the range() function provided with stdlib does some type conversion
>> (detects if it''s a number, and changes the type to integer)
which converts
>> "01" to 1 breaking this possibility.
>
>
> Hey Pablo,
>
> Just a question - do you have a NEED for it to be [01...99] with the
> leading zero? If I understand correctly, the end result is the same, as
> the range function will cast 01 to 1 and not BREAK because you pass it 01 -
> correct? Maybe I''m missing why you would need to use 01
explicitly?
>
>
>
>> I tried to submit a bug report, but I just can list the open ones,
can''t
>> make one myself. Is this intentional? How do I properly address this
>> request?
>>
>
> No problem. We''ve turned off Github Issues because we use our
central
> Redmine server for bug tracking -->
> http://projects.puppetlabs.com/projects/modules/issues Please feel free
> to file any module tickets here!
>
>
>>
>> So, I tried to change that myself, but no matter what I do to the
>> range.rb file, the changes are not picked up by the node. Do I have to
do
>> something to force a reload of the file? This runs in the server,
right?
>>
>> What I did then was to create a range_custom() which is a copy of the
>> former, but without the type conversion. I tried that and it works like
a
>> charm.
>>
>> Thanks!
>> Pablo
>>
>> --
>> You received this message because you are subscribed to the Google
Groups
>> "Puppet Users" group.
>> To post to this group, send email to puppet-users@googlegroups.com.
>> To unsubscribe from this group, send email to
puppet-users+unsubscribe@**
>> googlegroups.com <puppet-users%2Bunsubscribe@googlegroups.com>.
>> For more options, visit this group at http://groups.google.com/**
>>
group/puppet-users?hl=en<http://groups.google.com/group/puppet-users?hl=en>
>> .
>>
>>
>
>
> --
>
> Gary Larizza
> Professional Services Engineer
> Puppet Labs
>
> --
> You received this message because you are subscribed to the Google Groups
> "Puppet Users" group.
> To post to this group, send email to puppet-users@googlegroups.com.
> To unsubscribe from this group, send email to
> puppet-users+unsubscribe@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/puppet-users?hl=en.
>
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Hi,
irb(main):008:0> ("08".."12").to_a
> => ["08", "09", "10", "11",
"12"]
>
> irb(main):010:0> (8..12).to_a
> => [8, 9, 10, 11, 12]
>
> But then, the range() function in stdlib, makes a type conversion, that
> breaks this possibility. I don''t understand why do you do that
> conversion... the function is just an interface to the "range"
capability
> of Ruby, right? I think it should allow the programmer to choose the type
> he wants.
>
"With great power comes great responsibility" -- said that, please
allow
yourself to understand the following example:
("abc01".."abc99 ").to_a
Observe this little (which may happen) mistake after "abc99". That is
the
easily over-looked single space.
If you do that in your manifest, let say ... by accident... then your
Puppet Master will more likely become a happy candidate for OOM Killer to
consider :)
Do you know if, after editing a .rb file, I have to force some kind of
> reload to pick up the changes?
>
No, there is no need to reload, but this is not the solution. Instead, in
order to guarantee safe execution and also provide desired effect a
different approach can be taken: You can scan for how many string-encoded
digits you can fine within start of the range, count then and then based on
this counter collect (or map if you like Perl so much) generated range with
relevant substitution. If done right, then it will allow you to
zero-padding of any size at any place in the string e.g. test00, 00test
this00test, etc ...
I will update my version f range shortly to do precisely that.
KW
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/SsdPCfKazJwJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
On Thu, Mar 29, 2012 at 8:48 AM, Pablo Fernandez <pablo.fernandez@cscs.ch>wrote:> Dear stdlib''ers... > > I have just discovered the wonders of the parser functions, and got > impressed with the tens of functions that come with stdlib. First things > first... good work!!! Thanks!! > > And now the issue. It seems like the writer of the range() function did > not think about ranges with more than one digit that need leading zeros in > the first items, like "01..99", when you usually want to have 01, 02, and > so on. Ruby allows you to do ("01".."99") and that will do the right thing, > but the range() function provided with stdlib does some type conversion > (detects if it''s a number, and changes the type to integer) which converts > "01" to 1 breaking this possibility. I tried to submit a bug report, but I > just can list the open ones, can''t make one myself. Is this intentional? > How do I properly address this request? >Pablo, I just tried to reproduce this in the issue you reported at http://projects.puppetlabs.com/issues/13494 It looks like range("host01", "host99") works fine, returning [ "host01", "host02", ..., "host98", "host99" ] Is this what you''re looking for, or is it an issue only if the leading portion of the string is zero padded? -Jeff -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Hi,> It looks like range("host01", "host99") works fine, returning [ > "host01", "host02", ..., "host98", "host99" ] > > Is this what you''re looking for, or is it an issue only if the leading > portion of the string is zero padded?This is exactly what I was looking for, which indeed is much nicer than doing prefix(range(blabla)). In any case, I still don''t see the reason for the artificial type change inside that function, but anyway... BR/Pablo -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
On Thu, Mar 29, 2012 at 11:50 PM, Pablo Fernandez <pablo.fernandez@cscs.ch>wrote:> Hi, > > It looks like range("host01", "host99") works fine, returning [ "host01", >> "host02", ..., "host98", "host99" ] >> >> Is this what you''re looking for, or is it an issue only if the leading >> portion of the string is zero padded? >> > > This is exactly what I was looking for, which indeed is much nicer than > doing prefix(range(blabla)). > > In any case, I still don''t see the reason for the artificial type change > inside that function, but anyway... >The function isn''t actually doing the type change. Puppet itself massages strings that look like integers into actual integers. -Jeff -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Hi, [...] I changed how the range works (also making it resists Ruby''s string permutation bomb). Ideas implemented so far: https://gist.github.com/2277148 https://gist.github.com/2277144 https://gist.github.com/2270106 Also, the "a..b" and "a...b" notation should also be supported. I will push new version later. KW -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/SCJwCfMu9kEJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
I think you are wrong, check the source code:
# Check whether we have integer value if so then make it so ...
if start.match(/^\d+$/)
start = start.to_i
stop = stop.to_i
else
start = start.to_s
stop = stop.to_s
end
Besides, the change made by Krzysztof are awesome, specially
"node[01-45]", simply perfect.
Thanks!
Pablo
On 03/30/2012 06:49 PM, Jeff McCune wrote:> On Thu, Mar 29, 2012 at 11:50 PM, Pablo Fernandez
> <pablo.fernandez@cscs.ch <mailto:pablo.fernandez@cscs.ch>>
wrote:
>
> Hi,
>
> It looks like range("host01", "host99") works
fine, returning
> [ "host01", "host02", ..., "host98",
"host99" ]
>
> Is this what you''re looking for, or is it an issue only if
the
> leading portion of the string is zero padded?
>
>
> This is exactly what I was looking for, which indeed is much nicer
> than doing prefix(range(blabla)).
>
> In any case, I still don''t see the reason for the artificial
type
> change inside that function, but anyway...
>
>
> The function isn''t actually doing the type change. Puppet itself
> massages strings that look like integers into actual integers.
> -Jeff
>
> --
> You received this message because you are subscribed to the Google
> Groups "Puppet Users" group.
> To post to this group, send email to puppet-users@googlegroups.com.
> To unsubscribe from this group, send email to
> puppet-users+unsubscribe@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/puppet-users?hl=en.
--
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.