[EC63] UnnecessarilyAssignValuesToVariables rule on caught exception

[natixis-caen]

Stop to mark unused exception variables as code smell.

[natixis-caen]

Hi @dedece35 , I checked the list and this PR should be ok regarding the criteria.

[utarwyn]

LGTM 👌 Can you update the CHANGELOG?

[dedece35]

ok for me, but, please remove "throws Exception" because no Exception really thrown.

[dedece35]

@natixis-caen : last modification , then I will approve this PR :p

[dedece35]

To discuss in core-team : deprecation of this rule only for Java

[sonarqubecloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

80.0% Coverage
0.0% Duplication

[utarwyn]

@dedece35
To discuss in core-team : deprecation of this rule only for Java

Why this rule should be deprecated? Already in SonarQube? In fact, there are already 22 Java rules in SonarQube about not using variables/code. https://rules.sonarsource.com/java/tag/unused/RSPEC-1481/ Perhaps EC63 is a kind of duplicate? to be discussed.

[dedece35]

Hi @utarwyn, @glalloue, @jhertout,

after checking completyly this rule (currently, only implemented in JAVA), you're right @utarwyn, there is already 2 native SonarQube rules for 2 subjects "unused variable" and "return directly instead of creation of temporary variable" implemented in our ecoCode rule EC63 :

• unused variable : https://rules.sonarsource.com/java/tag/unused/RSPEC-1481/
• useless assignment : https://rules.sonarsource.com/java/tag/unused/RSPEC-1854
• immediately return : https://rules.sonarsource.com/java/RSPEC-1488/

Second thing, all implemented tests show that our EC63 rule is a duplication of 3 previous native Sonarqube rules : in each line of test files that we check if we have an issue reported by our rule, there is also a native Sonarqube rule that reports also an issue.

maybe we can use "ecocode" tag to native SaonrQube rules with our tool : https://github.com/green-code-initiative/ecoCode-common/tree/main/tools/rules_config#tagging-scripts
please see issue green-code-initiative/creedengo-common#26

for me, we can deprecate this rule, now. OK with me ?
if we are ok to deprecate it, 3 issues and the current PR will be closed :p and a new PR will be created to depreciate this EC63 rule.

[dedece35]

draft : waiting for #258

[dedece35]

Hi @natixis-caen and @utarwyn,
as #258 has been merged to deprecate rule EC63, this PR is closed without merging.