Hi Anne,

Answers inline.

1. after integrate PMD into source code, the compiled jar files' sizes will
> be increased? if so, did you notice how much gets increased?
>

There is no increase or impact at all on the compiled jars. The PMD plugin
is a source code analyzer, it doesn't look at the compiled code. It just
runs as part of the maven build cycle.


>
> 2. +1 for sonar upstream, I saw hbase is also included.
>

Ok I can take care of this.


>
> 3. are you aware of any native apache tools to detect race concurrency and
> race condition in a static way, such as predict
> <https://www.runtimeverification.com/predict/>. If so, it will be
> benefitted to include such a tool;
>

No I'm not unfortunately. The Sonar run will pick up on this stuff.

Colm.


> Thanks,
> Anne
>
> On Wed, Dec 9, 2015 at 2:18 AM, Colm O hEigeartaigh <[email protected]>
> wrote:
>
>> Hi Hao,
>>
>> Answers below.
>>
>>  > - What are the potentail bad behaviors we can catch throgh this plugin?
>>
>> To quote the PMD page: "It finds common programming flaws like unused
>> variables, empty catch blocks, unnecessary object creation, and so forth.
>> ". Essentially it imposes a constraint on how "messy" the source code is
>> allowed get. For example, currently the Sentry source is littered with
>> unused imports, etc.
>>
>>  > - Is it possible to use PMD to detect possible memory leakage?
>>
>> No, that is not the purpose of the tool. If the project is interested, I
>> could ask INFRA to add Sentry to analysis.apache.org, which runs
>> SonarQube
>> against the source and reports issues. That does detect possible memory
>> leakage (amongst many many other issues, a lot of them trivial).
>>
>>  > - Will the PMD plugin introduce some false alarms?And how we can handle
>> that?
>>
>> Not really. Sometimes it will report an error that you would like to keep.
>> So for example, it errors on unused constructor or method parameters.
>> Sometimes you might like to keep the parameter for backwards compatibility
>> reasons. You can add a comment to the line of code to skip the check "//
>> NOPMD" or else add a Java annotation to skip the check for a method,
>> class,
>> etc.:
>>
>> http://pmd.sourceforge.net/pmd-5.1.1/suppressing.html
>>
>> Thanks,
>>
>> Colm.
>>
>> On Tue, Dec 8, 2015 at 7:43 PM, Hao Hao <[email protected]> wrote:
>>
>> > Hi Colm,
>> >
>> > Thanks a lot for proposing a new way for managing the code quality.
>> > Questions:
>> >
>> >
>> >    - What are the potentail bad behaviors we can catch throgh this
>> plugin?
>> >    - Is it possible to use PMD to detect possible memory leakage?
>> >    - Will the PMD plugin introduce some false alarms?And how we can
>> handle
>> >    that?
>> >
>> >
>> > Thanks!
>> >
>> > Best,
>> > Hao
>> >
>> > On Tue, Dec 8, 2015 at 11:13 AM, Colm O hEigeartaigh <
>> [email protected]>
>> > wrote:
>> >
>> > > Hi all,
>> > >
>> > > I would like to get some opinions about applying the PMD maven plugin
>> to
>> > > Sentry, aka the Project Mess Detector:
>> > >
>> > > https://maven.apache.org/plugins/maven-pmd-plugin/
>> > > https://pmd.github.io/
>> > >
>> > > Essentially, PMD is a source code analyzer that checks for some common
>> > > issues that occur. The maven plugin enforces PMD by throwing a build
>> > error
>> > > if a regression is detected. It can be skipped via a maven profile to
>> > avoid
>> > > having to run it if you are making local changes.
>> > >
>> > > Quite a few of the other Apache projects I'm involved in use PMD as
>> part
>> > of
>> > > the build cycle (CXF, WSS4J, Santuario, Kerby, etc.).
>> > >
>> > > I've done a small bit of work with it already in Sentry, and would
>> like
>> > to
>> > > get some feedback on whether to proceed with submitting a patch or
>> not.
>> > > Most of the issues are fairly trivial such as unused imports,
>> variables,
>> > > empty catch blocks etc.
>> > >
>> > > Thoughts?
>> > >
>> > > Colm.
>> > >
>> > >
>> > > --
>> > > Colm O hEigeartaigh
>> > >
>> > > Talend Community Coder
>> > > http://coders.talend.com
>> > >
>> >
>>
>>
>>
>> --
>> Colm O hEigeartaigh
>>
>> Talend Community Coder
>> http://coders.talend.com
>>
>
>
>
> --
> Thanks,
> Anne
>



-- 
Colm O hEigeartaigh

Talend Community Coder
http://coders.talend.com

Reply via email to