Is sevntu sandbox actual for now?

54 views
Skip to first unread message

Pavel Baranchikov

unread,
Dec 29, 2014, 2:18:03 AM12/29/14
to sevntu-c...@googlegroups.com
Hello!

I've implemented EnumValueNameCheck in February 2014 - about a year ago. I was informed, that all the new checks are to be placed in sandbox project before committing them to the main Checkstyle project.
During last 2-3- months I see vast new checks implemented in Ceckstyle main project, but my EnumValueNameCheck is not moved to main project.

Looks, like EnumValueNameCheck is forgotten. Should I make pull request to main project to propagate the check there?

Best regards,
Pavel.

Roman Ivanov

unread,
Dec 29, 2014, 11:17:21 AM12/29/14
to sevntu-c...@googlegroups.com
HI Pavel,

All new Checks that appear in Checkstyle were created due to GSoC activity, and we had to implement and seriously test them directly in checkstyle project.

There are NO other new Checks that come to main library during that year. I have a lot of requests for new Checks but rule is strict for all.

Your Check is not forgotten, I even have issue to it, that I still did not found time to formulate and put in public, I will do in next few weeks for sure.
Please do not propagate it to main library for now.

​thanks,
Roman Ivanov

Roman Ivanov

unread,
Dec 30, 2014, 8:41:52 PM12/30/14
to sevntu-c...@googlegroups.com
Hi Pavel,

I share with you my note, I do not have time right now to think a more about it, but if you have time, please think about it. 

Checkstyle - update for Enum name , read
, we need to change ClassEnum detection algorithm: enum that implemnetes interface, override of Enum class methods should not be reason to become ClassEnum.

That note I did few months ago, please review an article and let me know if my notes a incorrect.

thanks,
Roman Ivanov

Pavel Baranchikov

unread,
Jul 15, 2015, 1:01:08 PM7/15/15
to sevntu-c...@googlegroups.com

, we need to change ClassEnum detection algorithm: enum that implemnetes interface, override of Enum class methods should not be reason to become ClassEnum.

This is the same theoretical discussion,as naming of final static fields: there are several points of view on what is constant.

BTW, we did not manage to use the ClassEnum of ConstantEnum distinguish for internal usage in our company. So, maybe the distinguish code should be removed at all, forcing user just to specify regexp, which should default to

^([A-Z][A-Z0-9]*(_[A-Z0-9]+)*)|([a-z][a-zA-Z0-9]*)$

(either CamelNotation or UPPERCASE_NOTATION)

Best ragards,
Pavel.

Roman Ivanov

unread,
Aug 2, 2015, 10:33:05 PM8/2/15
to sevntu-c...@googlegroups.com
Hi Pavel, 

that is good example why sandbox project is required :) .

please remove code that will not work properly.
Please make by default such pattern to not conflict with Spring and hibernate - there a lot of Enums there.


thanks ,
Roman Ivanov

Pavel Baranchikov

unread,
Aug 4, 2015, 12:05:15 PM8/4/15
to Sevntu Checkstyle

Hi, Roman

that is good example why sandbox project is required :) .
Yes. Agree.
 
please remove code that will not work properly.
Please make by default such pattern to not conflict with Spring and hibernate - there a lot of Enums there.

Roman Ivanov

unread,
Sep 29, 2015, 9:32:07 PM9/29/15
to sevntu-c...@googlegroups.com
Reply all
Reply to author
Forward
0 new messages