Some files not analyzed by Sonar "Invalid character encountered in file"

Question

I have a couple of files that are not analyzed with the following message:

Invalid character encountered in file [file name with full path] at line 9 for encoding UTF-8. Please fix file content or configure the encoding to be used using property 'sonar.sourceEncoding'.

In Visual Studio when I select File / Advance Save Options, the files were set to Western Europe (Windows) - Codepage 1252.

I changed it to Unicode (UTF-8 with Signature) - Codepage 65001.

But SonarQube still complains about the invalid characters. The "invalid characters" are comments in German with Umlaut-characters (ä,ö,ü)

What can I do to fix this (without removing the comments)?

Solution

The SonarQube Scanner for MSBuild and the SonarQube C# plugin currently expect all files of the project to have the UTF-8 encoding - and this is hardcoded.

There is a ticket to improve this in a future version: https://jira.sonarsource.com/browse/SONARMSBRU-174

Comments

[tobre]

How come it still did not work after changing the encoding to UTF-8?

[Dinesh Bolkensteyn]

I suspect that the file's encoding has not been properly changed. You can have a look at line 9 of that file in a hexadecimal editor to see for yourself whether it looks correct or not.