SonarQube output format startColumn needs to be zero based

[karlajw]

Thank you for creating the issue!

• [x ] Yes, I've searched similar issues on GitHub and didn't find any.
• [x ] Yes, I've included all information below (version, config, etc).

Please include the following information:

Version of woke

0.15.0, 0.15.1, 0.16.1

Config file

default config

Go environment

$ go version && go env
# paste output here

Verbose output of running

Woke generated an issue report that has the same value of 'startColumn' and 'endColumn' (both with a value of 1) when there is only one column (ie. file name, java package name)

{
"issues":[
{
"engineId":"woke",
"ruleId":"dummy",
"primaryLocation":{
"message":"Filename finding: dummy may be insensitive, use placeholder, sample instead",
"filePath":"src/main/java/com/testco/pocproject/dummy",
"textRange":{
"startLine":1,
"startColumn":1,
"endColumn":1
}
},
"type":"CODE_SMELL",
"severity":"MAJOR"
},

......

]
}

SonarQube throws the following runtime error when trying to import the above issue format:

ERROR: Error during SonarScanner execution
java.lang.IllegalArgumentException: Start pointer [line=1, lineOffset=1] should be before end pointer [line=1, lineOffset=1]

SonarQube documentation shows the startColumn field as zero based https://docs.sonarqube.org/latest/analysis/generic-issue/

The Hadolint project has had a similar issue eg. hadolint/hadolint#658

[cognitivegears]

I've duplicated this issue with a local Sonarqube instance. This happens for file/path results where the start index and end index are both set to 1. Sonarqube doesn't like having these set to the same value and fails to import. Created a potential fix as #162 and added a test to expose / validate the issue resolution.

[karlajw]

Thanks for the quick fix!

[caitlinelfring]

Released in v0.17.1

[caitlinelfring]

Thanks for the quick fix @cognitivegears!!!