Adding Groovy Linter to Github Super-Linter about npm-groovy-lint

[admiralAwkbar]

Tried to get it working a different way and it just doesn't show output for the second file...

bash-5.0# npm-groovy-lint --path "./test" --files "**/*.groovy" --verbose
/test/good.groovy
  2     warning  Class should be marked with one of @GrailsCompileStatic, @CompileStatic or @CompileDynamic  CompileStatic           


npm-groovy-lint results in 2 linted files:
┌─────────┬───────────┬─────────────┐
│ (index) │ Severity  │ Total found │
├─────────┼───────────┼─────────────┤
│    0    │  'Error'  │      0      │
│    1    │ 'Warning' │      1      │
│    2    │  'Info'   │      0      │

As from above, I have a good.groovy and a bad.groovy, the good is a syntax-correct file, and the bad is set to:

/* Hello World in Groovy */
pr ('some dank")  

Does it seem to just skip it in output?

[nvuillam]

I think that if you want to analyze the results of several groovy files, you should use --output json :)
The text output is more for a human to read

[admiralAwkbar]

@nvuillam Can you help me understand why it's not showing data for the second file with the bad data? Is my command incorrect or am I looking at this wrong? Any help is much appreciated

[nvuillam]

Plz can you fork the repo , add your test case in errors.test.js and send me the link, so I can reproduce the issue ?

[admiralAwkbar]

Hey @nvuillam sorry got sidetracked and didn't have a ton of time...
You can recreate this test by:

• Install npm-groovy-lint
• create a folder called test
  • mkdir test
• create file groovy-bad.groovy

class Example {
   static void main(String[] args) {
      File file = new File("E:/Example.txt")
      println "The file ${file.absolutePath} has ${file.length()} bytes"
   }
}
class Example{
some noise here
}

• Run linter against the file

bash-5.0# npm-groovy-lint --failon error --path "./test" --files "**/bad.groovy"

npm-groovy-lint results in 1 linted files:
┌─────────┬───────────┬─────────────┐
│ (index) │ Severity  │ Total found │
├─────────┼───────────┼─────────────┤
│    0    │  'Error'  │      0      │
│    1    │ 'Warning' │      0      │
│    2    │  'Info'   │      0      │
└─────────┴───────────┴─────────────┘
bash-5.0# echo $?
0

As you can see, it shows 1 linted file, but none of the output, and error code is 0

Can you help me understand what i may be doing wrong?

Working case

If I point it to a file with decent data, it works as expected

Example of good.groovy:

class Example {
   static void main(String[] args) {
      File file = new File("E:/Example.txt")
      println "The file ${file.absolutePath} has ${file.length()} bytes"
   }
}

Output

bash-5.0# npm-groovy-lint --failon error --path "./test" --files "**/good.groovy"
/tests/good.groovy
  1     warning  Class `Example` is the only class in `good.groovy`. In such a case the file and the class should have the same name.  ClassNameSameAsFilename
  1     warning  Class should be marked with one of @GrailsCompileStatic, @CompileStatic or @CompileDynamic  CompileStatic
  2     info     Class does not start with a blank line after the opening brace  ClassStartsWithBlankLine
  2     info     The method main in class Example is at the incorrect indent level: Expected column 5 but was 4  Indentation
  3     info     The statement on line 3 in class Example is at the incorrect indent level: Expected column 9 but was 7  Indentation
  3     warning  The use of java.io.File violates the Enterprise Java Bean specification  JavaIoPackageAccess
  3     info     The String 'E:/Example.txt' can be wrapped in single quotes instead of double quotes  UnnecessaryGString
  4     info     The statement on line 4 in class Example is at the incorrect indent level: Expected column 9 but was 7  Indentation
  5     info     Line ends with whitespace characters  TrailingWhitespace
  6     info     Class does not end with a blank line before the closing brace  ClassEndsWithBlankLine


npm-groovy-lint results in 1 linted files:
┌─────────┬───────────┬─────────────┐
│ (index) │ Severity  │ Total found │
├─────────┼───────────┼─────────────┤
│    0    │  'Error'  │      0      │
│    1    │ 'Warning' │      3      │
│    2    │  'Info'   │      7      │
└─────────┴───────────┴─────────────┘
bash-5.0#

[nvuillam]

It seems that CodeNarc does not write in its XML output file when there is a parse error

I'll try to find a solution

[nvuillam]

Working on the solution... I'll keep you informed :)

[admiralAwkbar]

@nvuillam awesome. Even if it just marked the file as failed to parse and exited with error, that would achieve the main goal

[nvuillam]

@admiralAwkbar please can you check with latest version ? (v5.7.0)

I had to embed Ant in the package, but now every file is parsed before being sent to CodeNarc, so all parsing errors are returned

Please reopen if you still see any problem :)