dev@glassfish.java.net

Re: verifying findbugs errors

From: Tim Quinn <tim.quinn_at_oracle.com>
Date: Mon, 4 Apr 2011 17:40:45 -0500

Hi, Jane.

On Apr 4, 2011, at 5:34 PM, Jane Young wrote:

> That's correct. I just recently updated the user.alias file so you
> are receiving e-mail notifications for the code that
> "nasaradu8"updated. I don't know nasaradu8 but the code that this
> person worked on is in appclient.

The messages I received because of that mapping are in security/core,
security/ejb.security, security/appclient.security.

Please map that user to someone on the security team.

Thanks.

- Tim

>
> The findbugs job that is sending out the massive e-mail notification
> is here (internal link):
> http://gf-hudson.us.oracle.com/hudson/job/gf-trunk-findbugs-low-priority
>
> I will work on the script to consolidate the e-mail notifications so
> it sends one e-mail to the recipients.
>
> On 4/4/11 3:23 PM, Tom Mueller wrote:
>> There is a users.alias file that translates users that are no
>> longer on the project into users that are.
>> See:
>> http://gf-hudson.us.oracle.com/hudson/job/gf-trunk-findbugs/ws/findbugstotext/users.alias/*view*/
>>
>> Tom
>>
>> On 4/4/2011 3:17 PM, Tim Quinn wrote:
>>>
>>> On Apr 4, 2011, at 5:08 PM, Justin Lee wrote:
>>>
>>>> Like many of you, I've received a glut of findbugs emails listing
>>>> various issues. I've started to whittle away on my list but I
>>>> can't seem to recreate that error report locally to verify the
>>>> issue. In fact, when I check the (internal) hudson job listed on
>>>> the wiki, it doesn't even list the issues I've been emailed
>>>> about. Running it locally on, say, admin/config-api, I get 0
>>>> errors before even changing the code. So my question, how can I
>>>> create that report locally so that I can do a pre/post check to
>>>> verify i'm satisfying the analysis parameters. Running "mvn
>>>> findbugs:findbugs" in that directory or the root doesn't seem
>>>> sufficient.
>>>
>>> I have found the same.
>>>
>>> I have aliased mvn to include -Dfindbugs.threshold=Low so the
>>> level should not explain the differences I see.
>>>
>>> Also, today I have received several messages in which all of the
>>> flagged lines have someone else's ID. The message says "...in the
>>> lines of code that you last modified..." but that doesn't seem to
>>> be accurate.
>>>
>>> Can anyone shed some light?
>>>
>>> - Tim
>