-
Improvement
-
Resolution: Fixed
-
Minor
-
None
Hi there
First of all - sorry if this is not the right way of asking such questions.
I'm trying to migrate from freestyle PMD plugin based duplication report to pipelines DRY plugin based duplication report.
In our build flow, we launch ant build that produces cpd-pmd XML report inside of a docker container. This docker image/container is the only place where source code lives.
After XML is ready it is copied to a side and docker container is destroyed.
On next pipeline step I'm trying to publish DRY report which fails with
java.io.FileNotFoundException: /var/www/project/sourceFile.php (No such file or directory)
I've traced this to below line in dry plugin source code:
https://github.com/jenkinsci/dry-plugin/blob/dry-2.46/src/main/java/hudson/plugins/dry/parser/DuplicationParserRegistry.java#L91
I understand the reason why exception is thrown - plugin excepts/assumes that all the files mentioned in XML are available and it calculates some kind of hash for each (probably for diffing with previous runs reasons - just guessing).
Since XML itself contains enough information to make the report itself - I would love to see an option that could be used to disable this part of code.
Is it reasonable/possible to implement such improvement within dry plugin?
[JENKINS-41323] Do not throw exception during hash computation if source code is not available
Łukasz Wojciechowski
added a comment - Thanks drulli for picking this up.
I'm aware that those additional functionalities won't work as expected but this is IMO justified as long as core functionality (publishing the base report information) is there.
Łukasz Wojciechowski
added a comment - Thanks drulli for picking this up.
I'm aware that those additional functionalities won't work as expected but this is IMO justified as long as core functionality (publishing the base report information) is there. Code changed in jenkins
User: Ulli Hafner
Path:
src/main/java/hudson/plugins/analysis/util/ContextHashCode.java
http://jenkins-ci.org/commit/analysis-core-plugin/8538c6ee3ce7774ee0113d9de651e7bc70bcfd5d
Log:
[FIXED JENKINS-41323] Use filename as hash code on IO exceptions.
Code changed in jenkins
User: Ulli Hafner
Path:
src/main/java/hudson/plugins/analysis/core/AbstractAnnotationParser.java
src/main/java/hudson/plugins/analysis/core/AnnotationsClassifier.java
http://jenkins-ci.org/commit/analysis-core-plugin/cf832abe4b6fd605345ef042c611aae1602f8976
Log:
JENKINS-41323 Use filename as hash code on IO exceptions.
Code changed in jenkins
User: Ulli Hafner
Path:
dry.iml
pom.xml
src/main/java/hudson/plugins/dry/parser/DuplicationParserRegistry.java
http://jenkins-ci.org/commit/dry-plugin/fb0c13b27b367a5273e0dd40fc44cada19a32be5
Log:
JENKINS-41323 Use filename as hash code on IO exceptions.
Code changed in jenkins
User: Ulli Hafner
Path:
checkstyle.iml
http://jenkins-ci.org/commit/checkstyle-plugin/db1c028e0ed2a4ae26551a6cadf2dd84dcf0b90f
Log:
[FIXED JENKINS-41323] Use filename as hash code on IO exceptions.
I think it should not harm if I replace the exception with a logging statement (or ignore it anyway). The error will show up anyway when you click into the source code view that should visualize the warning (not as exception but as a error message in the view).
Note that without source files the detection of new and fixed warnings will not work well.