jmeter-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Antonio Gomes Rodrigues <ra0...@gmail.com>
Subject Re: Error Prone (static analysis tool for Java that catches common programming mistakes) results
Date Mon, 14 Nov 2016 15:01:44 GMT
Hi,

I will fix all the warning I can after 3.1 release

Antonio

2016-10-31 22:09 GMT+01:00 Philippe Mouawad <philippe.mouawad@gmail.com>:

> Hi Antonio,
> My answers below.
> Do you have the full report ?
> Thanks
>
> On Thu, Oct 27, 2016 at 2:46 PM, Antonio Gomes Rodrigues <ra0077@gmail.com
> >
> wrote:
>
> > Hi all,
> >
> > I have run Error Prone in JMeter source code
> >
> >
> > **** It have found  2 [ReferenceEquality] Comparison using reference
> > equality instead of value equality
> > http://errorprone.info/bugpattern/ReferenceEquality
> >
> >
> >     [javac]
> > C:\Util\0_perso\TestError\trunk\src\core\org\apache\jmeter\samplers\
> > SampleSaveConfiguration.java:600:
> > warning: [ReferenceEquality] Comparison using reference equality instead
> of
> > value equality
> >     [javac]             stringValues = s.delimiter == delimiter ||
> > (delimiter != null && delimiter.equals(s.delimiter));
> >     [javac]                                        ^
> >     [javac]     (see http://errorprone.info/bugpattern/ReferenceEquality
> )
> >     [javac]   Did you mean 'stringValues = Objects.equals(s.delimiter,
> > delimiter) || (delimiter != null && delimiter.equals(s.delimiter));' or
> > 'stringValues = s.delimiter.equals(delimiter) || (delimiter != null &&
> > delimiter.equals(s.delimiter));'?
> >     [javac]
> > C:\Util\0_perso\TestError\trunk\src\core\org\apache\jmeter\samplers\
> > SampleSaveConfiguration.java:604:
> > warning: [ReferenceEquality] Comparison using reference equality instead
> of
> > value equality
> >     [javac]             complexValues = s.formatter == formatter ||
> > (formatter != null && formatter.equals(s.formatter));
> >     [javac]                                         ^
> >     [javac]     (see http://errorprone.info/bugpattern/ReferenceEquality
> )
> >     [javac]   Did you mean 'complexValues = Objects.equals(s.formatter,
> > formatter) || (formatter != null && formatter.equals(s.formatter));' or
> > 'complexValues = s.formatter.equals(formatter) || (formatter != null &&
> > formatter.equals(s.formatter));'?
> >
> >
> > I don't think there's a bug, but we could use Objects.equals to reduce
> syntax
>
> >
> > ****And some [ClassNewInstance] Class.newInstance() bypasses exception
> > checking; prefer getConstructor().newInstance()
> > http://errorprone.info/bugpattern/ClassNewInstance
> >
> >     [javac]
> > C:\Util\0_perso\TestError\trunk\src\core\org\apache\
> > jmeter\testelement\property\MapProperty.java:119:
> > warning: [ClassNewInstance] Class.newInstance() bypasses exception
> > checking; prefer getConstructor().newInstance()
> >     [javac]             Map<String, JMeterProperty> newCol =
> > value.getClass().newInstance();
> >
> > [javac]
> > ^
> >     [javac]     (see http://errorprone.info/bugpattern/ClassNewInstance)
> >     [javac]   Did you mean 'Map<String, JMeterProperty> newCol =
> > value.getClass().getConstructor().newInstance();'?
> >
>
> We should fix that at least for Java 9.
> We have something like 60 occurences
>
> >
> >
> > ****1 Synchronizing on non-final fields is not safe: if the field is ever
> > updated, different threads may end up locking on different objects
> > http://errorprone.info/bugpattern/SynchronizeOnNonFinalField
> >
> >     [javac]
> > C:\Util\0_perso\TestError\trunk\src\core\org\apache\
> > jmeter\reporters\Summariser.java:190:
> > warning: [SynchronizeOnNonFinalField] Synchronizing on non-final fields
> is
> > not safe: if the field is ever updated, different threads may end up
> > locking on different objects.
> >     [javac]         synchronized (myTotals) {
> >     [javac]                      ^
> >     [javac]     (see
> > http://errorprone.info/bugpattern/SynchronizeOnNonFinalField)
> >
>
> False positive for me in this case but others review is welcome.
>
> >
> > ****1 error [ChainingConstructorIgnoresParameter] The called constructor
> > accepts a parameter with the same name and type as one of its caller's
> > parameters, but its caller doesn't pass that parameter to it.  It's
> likely
> > that it was intended to.
> > http://errorprone.info/bugpattern/ChainingConstructorIgnoresParameter
> >
> >     [javac]
> > C:\Util\0_perso\TestError\trunk\src\core\org\apache\
> > jmeter\gui\util\FilePanel.java:44:
> > error: [ChainingConstructorIgnoresParameter] The called constructor
> > accepts
> > a parameter with the same name and type as one of its caller's
> parameters,
> > but its caller doesn't pass that parameter to it.  It's likely that it
> was
> > intended to.
> >     [javac]         this(title, (String) null, false);
> >     [javac]             ^
> >     [javac]     (see
> > http://errorprone.info/bugpattern/ChainingConstructorIgnoresParameter)
> >     [javac]   Did you mean 'this(title, filetype, false);'?
> >
>
> Good catch.
> It affected Html Assertion and IncludeController
>
> >
> >
> > Anybody have already use this tool and check if results ara false
> positive?
> >
>
> Seems to be a good bug detector.
>
> >
> > Error Prone : https://github.com/google/error-prone
> >
> >
> > Antonio
> >
>
>
>
> --
> Cordialement.
> Philippe Mouawad.
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message