Custom Language rule works fine when viewed in PsiViewer but show BAD_CHARACTER

Please always link to full sources. This seems to be a problem with lexer.

Hi Yann, 

           Sorry. Here are the details. 

fofik.bnf:

{
    parserClass = "com.fofik.parser.FofikParser"
    psiPackage = "com.fofik.psi"
    psiImplPackage = "com.fofik.psi.impl"

    elementTypeHolderClass = "com.fofik.psi.FofikTypes"
    elementTypeClass = "com.fofik.psi.FofikElementType"
    psiClassPrefix = "Fofik"
    psiVisitorName = "FofikElementVisitor"
    tokenTypeClass = "com.fofik.psi.FofikTokenType"

    psiImplUtilClass = "com.fofik.psi.impl.FofikPsiImplUtil"
    extends("literal|reference_expression")=value
      tokens = [
          L_CURLY='{'
          R_CURLY='}'
          L_PAREN = '('
          R_PAREN=')'
          SEMICOLON = ';'
          DOT = '.'
          PLUS = '+'
          COMMA = ','
          ASSIGN = 'regexp:[:=]'
          NUMBER="regexp:((0|[1-9][0-9]*)(\.[0-9]+)?([eE][+-]?[0-9]*)?)|Infinity|-Infinity|NaN"
          LINE_COMMENT='regexp://.*"/*" ([^*]|\*+[^*/])* (\*+"/")?'
          BLOCK_COMMENT='regexp:/\*([^*]|\*+[^*/])*(\*+/)?'
          DOUBLE_QUOTED_STRING='regexp:"([^\\"\r\n]|\\[^\r\n])*"?'
          SINGLE_QUOTED_STRING="regexp:'([^\\'\r\n]|\\[^\r\n])*'?"
          IDENTIFIER="regexp:[[:jletterdigit:]~!()*\-./@\^<>=]+"
          CRLF = "regexp:[\r\n]"
      ]
}
root ::= '{' property+ '}'
object ::= '{' (property)* '}'
property ::= property_name ('=' value ';')  {
methods = [ getName getValue]
mixin = "com.fofik.psi.impl.FofikNamedElementImpl"
implements = "com.fofik.psi.FofikNamedElement"
pin(".*") = 1
}
private property_name ::= literal |reference_expression
literal ::= SINGLE_QUOTED_STRING|DOUBLE_QUOTED_STRING|IDENTIFIER
reference_expression ::= IDENTIFIER

value ::= object| literal | reference_expression

 fofik.lex:

package com.fofik;

import com.intellij.lexer.FlexLexer;
import com.intellij.psi.tree.IElementType;
import static com.fofik.psi.FofikTypes.*;
import com.intellij.psi.TokenType;
import static com.intellij.psi.TokenType.WHITE_SPACE;
import static com.intellij.psi.TokenType.BAD_CHARACTER;


%%

%class FofikLexer
%implements FlexLexer
%unicode
%function advance
%type IElementType
%eof{  return;
%eof}

WHITE_SPACE=\s+
LINE_COMMENT="//"[^\n]*|"#"[^\n]*
BLOCK_COMMENT="/"\*([^*]|\*+[^*/])*(\*+"/")?
DOUBLE_QUOTED_STRING=\"([^\\\"]|\\[^\r\n])*\"?
SINGLE_QUOTED_STRING='([^\\']|\\[^\r\n])*'?
NUMBER=((0|[1-9][0-9]*)(\.[0-9]+)?([eE][+-]?[0-9]*)?)|Infinity|-Infinity|NaN
IDENTIFIER=[a-zA-Z_][a-zA-Z0-9_]*
%%
<YYINITIAL> {
    {WHITE_SPACE}             { return TokenType.WHITE_SPACE;}
  "{"                         { return L_CURLY; }
  "}"                         { return R_CURLY; }
       "("                         { return L_PAREN; }
        ")"                         { return R_PAREN; }
  ";"                         { return SEMICOLON; }
  "."                         { return DOT; }
  "+"                         { return PLUS; }
  ","                         { return COMMA;}
  {DOUBLE_QUOTED_STRING}      { return DOUBLE_QUOTED_STRING;}
  {SINGLE_QUOTED_STRING}      { return SINGLE_QUOTED_STRING;}
  {LINE_COMMENT}              { return LINE_COMMENT; }
  {BLOCK_COMMENT}             { return BLOCK_COMMENT; }
  {IDENTIFIER}                { return IDENTIFIER; }
      {NUMBER}                    { return NUMBER; }
}

[^] { return BAD_CHARACTER; }

Unfortunately, I cannot upload my project outside of the company network. Please do let me know if there is any further detail required.Thanks again. 

Sorry but it is impossible to diagnose this without having the full project sources.

Yann,  I'm working on moving the pieces specific to plugin into another project.