Should SourceCodePositioner (and other classes) consider all kinds of line breaks?

[Marcono1234]

Currently the PMD class SourceCodePositioner only considers \n as line break (here). However, usually all JDK classes consider \n, \r and \r\n as line breaks, for example BufferedReader.readLine() or Files.readAllLines.

Would it make sense to adjust SourceCodePositioner (and maybe other PMD classes) to behave the same way? Or do you think that this is not worth it because it makes the code more error-prone and \n is the most common line break?

[oowekyala]

PMD 7 does consider all of \n, \r, and \r\n - the PR is still pending: #3893 (comment)

I'm not sure it makes sense to change this in PMD 6. SourceCodePositioner does work with \r\n because splitting lines on \n also splits correctly \r\n (see #3893 (comment)), and so text coordinates work the same. It doesn't support single \r, but that's AFAIK only used on machines that are relatively rare nowadays: https://en.wikipedia.org/wiki/Newline#Representation