lexeme ++ whitespace was in the wrong order. This matters because any newline characters in either lexeme orwhitespace will reset the column number in reposition. Therefore the two strings must be concatenated in the same order in which they where lexed, otherwise newlines in the whitespace portion won't reset the column correctly.
I don't know if this matters for record-dot-preprocessor, but I was working on an indentation aware fork of this repo where this caused issues that took time to track down.
lexeme ++ whitespace
was in the wrong order. This matters because any newline characters in eitherlexeme
orwhitespace
will reset the column number inreposition
. Therefore the two strings must be concatenated in the same order in which they where lexed, otherwise newlines in thewhitespace
portion won't reset the column correctly.I don't know if this matters for
record-dot-preprocessor
, but I was working on an indentation aware fork of this repo where this caused issues that took time to track down.