Defining your own syntax highlighting in Crucible or Fisheye

Still need help?

The Atlassian Community is here for you.

Ask the community

Creating your own

All existing syntax highlighting definitions can be found at FISHEYE_INST/syntax directory. For example, java.def:

java.def
syntaxdef java {

/\s+/m : ;

/(a(bstract|ssert)|b(oolean|reak|yte)|c(a(se|tch)|har|lass|on(st|tinue))\
|d(efault|o(|uble))|e(lse|xtends|num)|f(alse|inal(|ly)|loat|or)|goto\
|i(f|mp(lements|ort)|n(stanceof|t(|erface)))|long|n(ative|ew|ull)\
|p(ackage|r(ivate|otected)|ublic)|return|s(hort|t(atic|rictfp)|uper|witch|ynchronized)\
|t(h(is|row(|s))|r(ansient|ue|y))|vo(id|latile)|while)\b/ : {
      region {
        type=keyword;
        index=word;
      }
}

/[\p{L}_\$][\p{L}\p{N}_\$]*/ : {
      region {
         type=identifier;
         index=word;
      }
  }

#annotation
/@[\p{L}_\$][\p{L}\p{N}_\$]*/ : {
      region {
         type=annotation;
         index=word;
      }
  }


# string literals
/"/ : {
    context {
            /\\"/: ;
            /\\./: ;
            /$/m : exit;
            "\"" : exit;
        }
        region ${ALL} {
               type=string;
        }
    }

# char literal
/'/ : {
    context {
            /\\./: ;
            /$/m : exit;
            "'" : exit;
        }
        region ${ALL} {
               type=char_literal;
        }
    }


/\/\/.*$/m : {
      todo(${0});
      region {
         type=comment;
	     index=prose;
	     findlinks=true;
      }
   }


/\/\*\*(.*?)\*\//s : {
      javadoc(${1});
      region {
         type=comment;
         index=prose;
         findlinks=true;
      }
   }

   context javadoc {

      /(\B@[a-zA-Z0-9-_.]+)\s*([^@\*]*)/ : {
         region ${1} {
            type=commentmeta;
	     }
      }
   }

/(0x[0-9a-f][0-9a-f_]*l?)|(0b[0-1][0-1_]*l?)|([0-9][0-9_]*(\.[0-9][0-9_]*)?|\.[0-9][0-9_]+)(e(\+|-)?[0-9][0-9_]*)?(f|l|d)?/i : {
       region {
          type=numeric;
       }
   }

/\/\*(.*?)\*\//s : {
      todo(${1});
      region {
         type=comment;
         findlinks=true;
      }
      region ${1} {
	     index=prose;
      }
   }

   context todo {
      /(?i)todo\s.*/ : {
        region {
          type=todo;
        }
      }

   }

} 


As can be seen in above, these syntax highlighting definition files are just Regular Expressions, so if you know Regular Expressions and the semantics of the programming language you want to write a new definition for, you should be able to translate one of these existing definitions to what you need in your new file. Usually there is no need to create your own definition from scratch for a programming language that is not already there, you can use an existing template that best matches the definition you need as a starting point.

Make sure to following these steps:

  1. Copy one of the existing definitions into your FISHEYE_INST/syntax directory. (Note, by making your modifications in the FISHEYE_INST directory, your changes will persist across upgrades. If you make the changes in FISHEYE_HOME, they will need to be reapplied each upgrade).
  2. Modify the filename and the definition in the FISHEYE_INST/syntax directory. E.g. you can change java.def to objectiveC.def and modify the definition to suit your programming language.
  3. Copy the FISHEYE_HOME/syntax/filename.map to your FISHEYE_INST/syntax directory and modify it to map file extensions to definitions.
  4. Restart Fisheye to pick up the change.

Grammar

This is a sample grammar of a .def file:

Grammar
syntaxdef : "syntaxdef" identifier LCURLY ( matchrule | context )* RCURLY;
 
context : "context" identifier LCURLY contextBody RCURLY;
 
contextBody : ( matchrule | context )*;
 
matchrule : ( stringmatcher | regexpmatcher ) COLON matchbody;
 
stringmatcher: STRING_LITERAL;
       
regexpmatcher : REGEXP;
 
matchbody :
        ( ("exit")?  SEMI)
    |
        (
            LCURLY
            ("predicate" ASSIGN dottedIdentifier)?
            (
               ( ("exit")?  SEMI)
               |
               (            
                    ((           
                        (identifier (LPAREN submatchref RPAREN)? SEMI)
                        |
                        ("context" (LPAREN submatchref RPAREN)? LCURLY contextBody RCURLY)
                    ))*
                    (regiondecl)*
               )
            )
            ("filter" ASSIGN dottedIdentifier)*
            RCURLY
        )
    ;
 
regiondecl
    : "region" (submatchref)? LCURLY
        (
            identifier (DOT identifier)* ASSIGN
            (identifier|STRING_LITERAL) SEMI
        )*
        RCURLY;
 
submatchref :  DOLLAR LCURLY (identifier|NUMINT) RCURLY;
 
dottedIdentifier : IDENT ("." IDENT)*;

LPAREN          :   '('     ;
RPAREN          :   ')'     ;
LCURLY          :   '{'     ;
RCURLY          :   '}'     ;
COLON           :   ':'     ;
SEMI            :   ';'     ;
DOT             :   '.'     ;
ASSIGN          :   '='     ;

Examples

Below are some examples found online and / or shared by customers:


TypeScript
# Basic TypeScript syntax highlighting rules for FishEye/Crucible.
#
# Put this file to the <FISHEYE_INST>/syntax directory and update the
# <FISHEYE_INST>/syntax/filename.map file to contain the following lines:
#
# "**/*.ts" typescript.def "TypeScript"
#
# FishEye/Crucible might need to be restarted to have the changes take effect.
#
# References:
# * https://confluence.atlassian.com/display/FISHKB/Defining+your+own+syntax+highlighting+in+Crucible+or+Fisheye
# * https://confluence.atlassian.com/display/FISHKB/Configure+syntax+highlighting+for+non-standard+file+extensions
#
# Notes:
# * this schema considers identifiers that start with an uppercase letter to represent classes or constants, so
#   it treats such identifiers as 'symbol' for better colors. If you don't want such treatment, replace the 
#   'symbol' with 'identifier'.
#
# TODO: multiline strings do not really work, e.g.:
# var mySuperString = 'this is a \
# multiline string';

syntaxdef typescript {

# whitespaces
/\s+/m : ;

# keywords
/\b(a(ny|s)|b(oolean|reak)|c(a(se|tch)|lass|on(st(|ructor)|tinue))|d(e(bugger|clare|fault|lete)|o)|e(lse|num|x(port|tends))|f(alse|inally|or|rom|unction)|get|i(f|mp(lements|ort)|n(|stanceof|terface))|let|module|n(ew|u(ll|mber))|of|p(ackage|r(ivate|otected)|ublic)|r(equire|eturn)|s(et|t(atic|ring)|uper|witch|ymbol)|t(h(is|row)|r(ue|y)|ype(|of))|v(ar|oid)|w(hile|ith)|yield)\b/ : {
    region {
        type=keyword;
        index=word;
    }
}

# decorators
/@[a-zA-Z_][a-zA-Z0-9_]*/ : {
    region {
        type=decorator;
        index=word;
    }
}

# backquote string literal
/`/ : {
    context {
        /\\`/: ;
        "`" : exit;
        /(\$\{.*?\})/ : {
            region ${1} {
                type=identifier;
                index=word;
            }
        }
    }
    region ${ALL} {
        type=string;
    }
}

# single-quote string literal
/'/ : {
    context {
        /\\'/: ;
        /$/m : exit;
        "'" : exit;
    }
    region ${ALL} {
        type=string;
    }
}

# double-quote string literal
/"/ : {
    context {
        /\\"/: ;
        /$/m : exit;
        "\"" : exit;
    }
    region ${ALL} {
        type=string;
    }
}

# numbers
/(0x[0-9a-f]+)|(([0-9]+\.?)|([0-9]*\.[0-9]+))/i : {
    region {
        type=numeric;
    }
}

# Identifiers
/[a-z_$][a-zA-Z_0-9]*/ : {
    region {
        type=identifier;
        index=word;
    }
}

# probable classes or constant expressions
/[A-Z][a-zA-Z_0-9]*/ : {
    region {
        type=symbol;
        index=word;
    }
}

# multiline comment
/\/\*(.*?)\*\//s : {
    todo(${1});
    region {
        type=comment;
        findlinks=true;
    }
    region ${1} {
        index=prose;
    }
}

# one-line comment
/\/\/.*$/m : {
    todo(${0});
    region {
        type=comment;
        index=prose;
        findlinks=true;
    }
}

context todo {
    /(?i)(todo):?\s.*/ : {
        region {
            type=todo;
        }
    }
}

# regex
/
 \/
 ([^\n\r*\\\/]|\\[^\r\n])# RegularExpressionFirstChar
/x : {
    context { # RegularExpresionChar*
        /\\[^\r\n]/ : ;
        /\// : exit;
        /[\r\n]/ : exit; # for safety
    }
    region ${ALL} {
        type=regex;
    }
}

}
Chapel
syntaxdef chpl {

# whitespace
/\s+/m : ;

# keywords
/(align|as|atomic|\
begin|bool|borrowed|break|by|\
catch|class|clone|cobegin|coforall|complex|config|const|continue|\
defer|delete|destroy|dmapped|domain|do|\
else|enum|except|export|extern|\
false|forall|forwarding|for|\
if|imag|index|init|inline|inout|integral|int|in|iter|\
label|lambda|let|lifetime|locale|local|\
module|\
new|nil|noinit|\
only|on|opaque|otherwise|out|override|owned|\
param|pragma|primitive|private|proc|prototype|public|\
range|real|record|reduce|ref|require|return|\
scan|select|serial|shared|single|sparse|string|subdomain|sync|\
then|this|throws|throw|true|try|type|\
uint|union|unmanaged|use|\
var|\
when|where|while|with|\
yield|\
zip)\b/ : {
       region {
         type=keyword;
         index=word;
       }
    }

# preprocessor line
/^#.*$/m : {
       region {
         type=keyword; # obviously not a keyword. this will change when we shake out a cannonical list of types
       }
    }

# char literal
/L?'(\\.|\\[0-7]{3}|\\x[0-9a-fA-F]{2}|.)'/ : {
       region {
          type=char_literal;
       }    
}

# string literal
/L?"/ : {
    context {
        /\\"/: ;
        /\\./: ;
        /$/m : exit;
        "\"" : exit;
    }
    region ${ALL} {
           type=string;
    }
}

#numeric literal
/(0x[0-9a-f]+(lu|ul|[ul])?)|((([0-9]+\.?)|([0-9]*\.[0-9]+))(e(\+|-)?[0-9]+)?(fl|lf|[lf])?)/i : {
       region {
          type=numeric;
       }
   }# identifier
/[a-zA-Z_][a-zA-Z_0-9]*/ : {
      region {
         type=identifier;
         index=word;
      }
   }

# comment
/\/\*.*?\*\//s : {
      region {
         type=comment;
         index=prose;
         findlinks=true;
      }
    }

# inline comment
/\/\/.*$/m : {
      region {
         type=comment;
         index=prose;
         findlinks=true;
      }
   }
}
Kotlin
syntaxdef kotlin {


/\s+/m : ;
#keywords
/(a(bstract|nnotation|s)|break|c(atch|lass|o(mpanion|n(st|tinue))|rossinline)\
|d(ata|elegate|o|ynamic)|e(lse|num|xternal)|f(i(eld|le|nally)|or|un|alse)|get\
|i(f|mport|n(fix|it|line|terface)|s|t)|n(oinline|ull)|o(bject|pe(n|rator)|ut)\
|p(a(ckage|ram)|roperty)|re(ceiver|ified|turn)|s(e(aled|tparam)|u(per|spend))\
|t(ailrec|h(is|row)|r(y|ue)|ype(alias|of))|va(l|rarg)|wh(en|ile))\b/ : {
      region {
        type=keyword;
        index=word;
      }
}

/[\p{L}_\$][\p{L}\p{N}_\$]*/ : {
      region {
         type=identifier;
         index=word;
      }
  }

#annotation
/@[\p{L}_\$][\p{L}\p{N}_\$]*/ : {
      region {
         type=annotation;
         index=word;
      }
  }


# string literals
/"/ : {
    context {
            /\\"/: ;
            /\\./: ;
            /$/m : exit;
            "\"" : exit;
        }
        region ${ALL} {
               type=string;
        }
    }

# char literal
/'/ : {
    context {
            /\\./: ;
            /$/m : exit;
            "'" : exit;
        }
        region ${ALL} {
               type=char_literal;
        }
    }


/\/\/.*$/m : {
      todo(${0});
      region {
         type=comment;
         index=prose;
         findlinks=true;
      }
   }


/\/\*\*(.*?)\*\//s : {
      javadoc(${1});
      region {
         type=comment;
         index=prose;
         findlinks=true;
      }
   }

   context javadoc {

      /(\B@[a-zA-Z0-9-_.]+)\s*([^@\*]*)/ : {
         region ${1} {
            type=commentmeta;
         }
      }
   }

/(0x[0-9a-f][0-9a-f_]*l?)|(0b[0-1][0-1_]*l?)|([0-9][0-9_]*(\.[0-9][0-9_]*)?|\.[0-9][0-9_]+)(e(\+|-)?[0-9][0-9_]*)?(f|l|d)?/i : {
       region {
          type=numeric;
       }
   }

/\/\*(.*?)\*\//s : {
      todo(${1});
      region {
         type=comment;
         findlinks=true;
      }
      region ${1} {
         index=prose;
      }
   }

   context todo {
      /(?i)todo\s.*/ : {
        region {
          type=todo;
        }
      }

   }

}


Configure syntax highlighting for non-standard file extensions


Last modified on Aug 15, 2022

Was this helpful?

Yes
No
Provide feedback about this article
Powered by Confluence and Scroll Viewport.