Issue #10506 has been reported by Henrik Lindberg.
----------------------------------------
Bug #10506: Naming rules are inconsistent
https://projects.puppetlabs.com/issues/10506
Author: Henrik Lindberg
Status: Unreviewed
Priority: Normal
Assignee:
Category: language
Target version:
Affected Puppet version:
Keywords: usability syntax
Branch:
The rules for naming of classes is inconsistently applied to different language
concepts.
* It is allowed to name a class using *NUMBER* (e.g. '0', '0.0', '0477',
'0x1f', '1.3e-2').
* A dollar variable is parsed with the regular expression
`%r{\$(::)?([-\w]+::)*[-\w]+}` and can thus never refer to a variable in a
class named with a '.' (period) in the name
* A *NAME*, *CLASSREF*, or *CLASSNAME* is parsed with`
%r{((::)?[a-z0-9][-\w]*)(::[a-z0-9][-\w]*)*}` and can also not contain a '.'
(period)
* Regular expression matching injects values into variables `$0` - `$n`
(although not directly clashing since a class is typically not referenced with
a variable, it is still confusing to see say `$0::1`, when $0 is a regexp match)
* Class defaults can be set using a class reference (each segment starts with
upper case letter in a fully qualified name), this is strange when name is
numeric (how do you write an upper case '1'?). Compare:
using words:
class one($a) { class two($b) {} }
One { a=>10 }
One::Two {b=>20}
using numbers:
class 1($a) { class 2($b) {} }
1 { a=>10 }
1::2 {b=>20}
The second example (naturally) results in:
Could not parse for environment production: All resource specifications
require names at /private/tmp/test.pp:2 on node xxx
* A separate issue is that assignment to `$0` - `$n` may not take effect (the
variable is still undefined) in some cases.
* The documentation does not seem to mention the fact that classes may have
numerical names (the stricter rules are shown in several places).
I suggest that:
* Numerical names should be deprecated
* Assignment to `$0` - `$n` should be illegal (reserved for regular expression
matches)
In contrast, if trying to fix this by adding '.' (period) as valid in
variables, a lot of string interpolation will break (in similar ways to how the
altered string interpolation rules regarding '-' (hyphen) breaks existing code).
The interpolation issues could be resolved by deprecating the interpolation
short form (i.e. interpolating without braces), but that may be considered a
more serious imposition (even if recommended as best practice) - i.e. that
`"Hello $name"` must be written `"Hello ${name}"`.
It seems like 2.8 would be a good candidate for more stringent rules since
dynamic scoping of variables will be deprecated and people do have to inspect
their code and deal with naming/reference issues.
--
You have received this notification because you have either subscribed to it,
or are involved in it.
To change your notification preferences, please click here:
http://projects.puppetlabs.com/my/account
--
You received this message because you are subscribed to the Google Groups
"Puppet Bugs" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/puppet-bugs?hl=en.
