On Tue, Aug 5, 2014 at 3:50 PM, rahul <[email protected]> wrote:

> So to summarize, this is our plan for Rubocop:
>
> - We propose to enable AndOr cop in small chunks, giving preference to
> those files/directories that are heavily in development.
> - For AndOr, the conclusion seems to be to avoid keywords completely, and
> ensure that the instances where they are used are changed do not hurt
> readability.
> - As a prototype, we have turned on AndOr on lib/pops directory PR 2892
>

Also a heads-up that for pull requests:

1) a week or so ago, we added a travis job that fails if any of the
.rubocop.yml enabled cops report anything (these are just the cops that
were uncontroversial at the beginning of this thread)

2) just now, I turned on houndci.com which will comment on pull requests
based on the same configuration

Note that hound *can* be configured with a separate config file of its own,
but we don't have one, so it falls back to the .rubocop.yml. If we wanted
to have a set of cops which triggered comments on the PRs, but didn't
figure travis fails, we could get that by having a separate houndci.yml.
Not sure what I think of that, but just putting the idea out there.

Kylo


>
> On Tuesday, July 29, 2014 11:00:46 PM UTC-7, Kylo Ginsberg wrote:
>
>> On Tue, Jul 29, 2014 at 4:42 PM, Andy Parker <[email protected]>
>> wrote:
>>
>>> Right now the PRs are doing a mechanical transformation to remove a
>>> keyword that we don't want to use. What is missing is that it isn't
>>> transforming the code into what later changes to that code should preserve.
>>> Or put another way, if we got a PR that contained new code that looked like
>>> that we would reject it, I think. It passes the test of not using
>>> disallowed operators, but it doesn't pass the test of being written in a
>>> form that a reader would expect.
>>>
>>
>> I agree that the purely mechanical transformation applied to the
>> genuinely flow control cases introduces constructs that would slow me down
>> as a code reader (and that I'd be very unlikely to write).
>>
>> So are there objections to converting such cases to use 'if', etc?
>> Personally I'd find that clearer and easier to read. And it would still
>> allow us to eliminate the and/or keywords which we've identified as the
>> source of some bugs/confusion.
>>
>> Kylo
>>
>  --
> You received this message because you are subscribed to the Google Groups
> "Puppet Developers" 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-dev/ab872dad-c258-4e09-81b3-8c13f17bc968%40googlegroups.com
> <https://groups.google.com/d/msgid/puppet-dev/ab872dad-c258-4e09-81b3-8c13f17bc968%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>
> For more options, visit https://groups.google.com/d/optout.
>



-- 
Kylo Ginsberg
[email protected]

*Join us at PuppetConf 2014 <http://www.puppetconf.com/>, September
20-24 in San Francisco*
*Register by September 8th to take advantage of the Final Countdown
<https://www.eventbrite.com/e/puppetconf-2014-tickets-7666774529?discount=FinalCountdown>
*
*—**save $149!*

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" 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-dev/CALsUZFEHveiiDug7xWBT%2BWo0PKR0V-FrGXF%3DE%3DY%3D9-KqLrh4Ag%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to