Output FilesToAnalyze.txt as utf-8

[NatMarchand]

Hi,
I'm currently encountering issues as I need to do a sed to replace paths in FilesToAnalyze.txt files as those files are utf-16le encoded.

Can you change them to utf-8 ? Just removing Encoding parameters in the file should do the trick as default value is utf-8.
Thanks !

sonar-scanner-msbuild/src/SonarScanner.MSBuild.Tasks/Targets/SonarQube.Integration.targets

Line 294 in 45df416

Encoding="Unicode" />

[pavel-mikula-sonarsource]

The explicit argument would be preferred. We need whatever will produce UTF-8 with BOM

[duncanp-sonar]

@NatMarchand FilesToAnalyze.txt is generated and consumed by the scanner i.e. it's an implementation detail that could change.

Why do you need to modify the contents of the file? Is it to work round a bug, or is there a new feature you'd like the scanner to support?

cc @tom-howlett-sonarsource

[NatMarchand]

I'm currently using sonarqube with gitlab and in our pipeline, starting the analysis, running tests and completing analysis are on different jobs. When the repo is cloned, there's a random string in the path (which is the id of the provisioned runner) which makes the path not the same accross all the jobs. Therefore, as all paths are absolute, I need to rewrite and change the base path.
Something like that:

    - find results/ -type f -name "coverage.opencover.xml" -exec sed -i "s /src ${CI_PROJECT_DIR} gI" {} \;
    - find .sonarqube/ -type f -exec sed -i "s /builds/.*/${CI_PROJECT_PATH} ${CI_PROJECT_DIR} gI" {} \;
    - find . -type f -name "*.RoslynCA.json" -exec sed -i "s /builds/.*/${CI_PROJECT_PATH} ${CI_PROJECT_DIR} gI" {} \;
    - dotnet-sonarscanner end -d:sonar.login="${SONAR_TOKEN}"