[wildfly-dev] Tyr - pull request format checker

Martin Stefanko mstefank at redhat.com
Wed Apr 10 06:35:06 EDT 2019


For the demo I've started with this
<https://gist.github.com/xstefank/3a79f2f199ee1e449a44c607120a9d30>. For
additional checks I have only something to verify that issue in title
matches the link in the description. Any ideas are welcome!

I was also thinking about making distinction between issues / bugs and
features if that would be useful.

I've created a google document to track this -
https://docs.google.com/document/d/1ZuD5fyOlGTVDh96jiuM6jTdofm4fG6gZqdfBNRdZp8I/edit?usp=sharing
.

Martin


On Wed, Apr 10, 2019 at 11:10 AM Darran Lofthouse <
darran.lofthouse at jboss.com> wrote:

> Do we have somewhere yet that we can start to look into an initial set of
> rules to discuss what the initial set should be?
>
> On Wed, Apr 10, 2019 at 6:32 AM Martin Stefanko <mstefank at redhat.com>
> wrote:
>
>> > The time the cross becomes a problem is if we adopt a set of
>> verification rules that we know will regularly trigger a failure to be
>> ignored, if the failure cases to be ignored are truly exceptional cases
>> then the cross will be useful.
>>
>> It would be beneficial to adjust rules as we'll see how useful they are
>> as time progresses. Configuration of rules is external to the service so it
>> will be possible.
>>
>> Martin
>>
>>
>> On Tue, Apr 9, 2019 at 5:49 PM Darran Lofthouse <
>> darran.lofthouse at jboss.com> wrote:
>>
>>> > OTOH if the mergers decide they want to do something I don't want a
>>>> tool preventing us doing it, which is what my questions were driving at.
>>>> We're way smarter than the tool.
>>>>
>>>> The status should not prevent a possibility of merge but if the cross
>>>> next to the commit would be a problem we can add an override to disable tyr
>>>> check (meaning make it pass manually) with a comment or similar as
>>>> suggested above.
>>>>
>>>
>>> The time the cross becomes a problem is if we adopt a set of
>>> verification rules that we know will regularly trigger a failure to be
>>> ignored, if the failure cases to be ignored are truly exceptional cases
>>> then the cross will be useful.
>>>
>>>
>>>>
>>>> Martin
>>>>
>>>>
>>>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/wildfly-dev/attachments/20190410/d1d18b19/attachment-0001.html 


More information about the wildfly-dev mailing list