first attempt of codemirror fortran parser

fortran.js

CodeMirror.defineMode("fortran", function(conf, parserConf) {
    var ERRORCLASS = 'error';

    function wordRegexp(words) {
        return new RegExp("^((" + words.join(")|(") + "))\\b", "i");
    }

    var Operators = new RegExp("(\.and\.|\.or\.|\.eq\.|\.lt\.|\.le\.|\.gt\.|\.ge\.|\.ne\.|\.not\.|\.eqv\.|\.neqv\.)", "i");
    var Operators2 = new RegExp("((==)|(/=)|(=)|(<=)|(>=)|(/)|(//))"); //" |(\\*\\*)|\-|\+|\/\/|\/|(?!^)\*)");

    var singleDelimiters = new RegExp('^[\\(\\)\\[\\]\\,:=]')
    var stringPrefixes = new RegExp("^(([rub]|(ur)|(br))?(['\"]))", "i");
    var identifiers = new RegExp("^[_A-Za-z][_A-Za-z0-9]*");

    var commonkeywords = ["abstract", "accept", "allocatable", "allocate",
                          "array", "assign", "asynchronous", "backspace",
                          "bind", "block( data)?", "byte", "call", "case",
                          "class", "close", "common", "contains",
                          "continue", "cycle", "data", "deallocate",
                          "decode", "deferred", "dimension", "do",
                          "elemental", "else", "encode", "end( file)?",
                          "endif", "entry", "enumerator", "equivalence",
                          "exit", "external", "extrinsic", "final",
                          "forall", "format", "function", "generic",
                          "goto", "if", "implicit", "import", "include",
                          "inquire", "intent", "interface", "intrinsic",
                          "module", "namelist", "non_intrinsic",
                          "non_overridable", "none", "nopass",
                          "nullify", "open", "optional", "options",
                          "parameter", "pass", "pause", "pointer",
                          "print", "private", "program", "protected",
                          "public", "pure", "read", "recursive", "result",
                          "return", "rewind", "save", "select", "sequence",
                          "stop", "subroutine", "target", "then", "type",
                          "use", "value", "volatile", "where", "while",
                          "write"];
    var blockskeywords = ["program", "subroutine", "module", "type", "call", "block"];
    var dataTypes = ["c_bool", "c_char", "c_double", "c_double_complex",
                     "c_float", "c_float_complex", "c_funptr", "c_int",
                     "c_int16_t", "c_int32_t", "c_int64_t", "c_int8_t",
                     "c_int_fast16_t", "c_int_fast32_t", "c_int_fast64_t",
                     "c_int_fast8_t", "c_int_least16_t", "c_int_least32_t",
                     "c_int_least64_t", "c_int_least8_t", "c_intmax_t",
                     "c_intptr_t", "c_long", "c_long_double",
                     "c_long_double_complex", "c_long_long", "c_ptr",
                     "c_short", "c_signed_char", "c_size_t", "character",
                     "complex", "double complex", "double precision",
                     "integer", "logical", "real"];
    var commonBuiltins = ["abort", "abs", "access", "achar", "acos",
                          "adjustl", "adjustr", "aimag", "aint", "alarm",
                          "all", "allocated", "alog", "amax", "amin",
                          "amod", "and", "anint", "any", "asin",
                          "associated", "atan", "besj", "besjn", "besy",
                          "besyn", "bit_size", "btest", "cabs", "ccos",
                          "ceiling", "cexp", "char", "chdir", "chmod",
                          "clog", "cmplx", "command_argument_count",
                          "complex", "conjg", "cos", "cosh", "count",
                          "cpu_time", "cshift", "csin", "csqrt", "ctime",
                          "c_funloc", "c_loc", "c_associated", "c_null_ptr",
                          "c_null_funptr", "c_f_pointer", "c_null_char",
                          "c_alert", "c_backspace", "c_form_feed",
                          "c_new_line", "c_carriage_return",
                          "c_horizontal_tab", "c_vertical_tab", "dabs",
                          "dacos", "dasin", "datan", "date_and_time",
                          "dbesj", "dbesj", "dbesjn", "dbesy", "dbesy",
                          "dbesyn", "dble", "dcos", "dcosh", "ddim", "derf",
                          "derfc", "dexp", "digits", "dim", "dint", "dlog",
                          "dlog", "dmax", "dmin", "dmod", "dnint",
                          "dot_product", "dprod", "dsign", "dsinh",
                          "dsin", "dsqrt", "dtanh", "dtan", "dtime",
                          "eoshift", "epsilon", "erf", "erfc", "etime",
                          "exit", "exp", "exponent", "extends_type_of",
                          "fdate", "fget", "fgetc", "float", "floor",
                          "flush", "fnum", "fputc", "fput", "fraction",
                          "fseek", "fstat", "ftell", "gerror", "getarg",
                          "get_command", "get_command_argument",
                          "get_environment_variable", "getcwd",
                          "getenv", "getgid", "getlog", "getpid",
                          "getuid", "gmtime", "hostnm", "huge", "iabs",
                          "iachar", "iand", "iargc", "ibclr", "ibits",
                          "ibset", "ichar", "idate", "idim", "idint",
                          "idnint", "ieor", "ierrno", "ifix", "imag",
                          "imagpart", "index", "int", "ior", "irand",
                          "isatty", "ishft", "ishftc", "isign",
                          "iso_c_binding", "is_iostat_end", "is_iostat_eor",
                          "itime", "kill", "kind", "lbound", "len", "len_trim",
                          "lge", "lgt", "link", "lle", "llt", "lnblnk", "loc",
                          "log", "logical", "long", "lshift", "lstat", "ltime",
                          "matmul", "max", "maxexponent", "maxloc", "maxval",
                          "mclock", "merge", "move_alloc", "min", "minexponent",
                          "minloc", "minval", "mod", "modulo", "mvbits",
                          "nearest", "new_line", "nint", "not", "or", "pack",
                          "perror", "precision", "present", "product", "radix",
                          "rand", "random_number", "random_seed", "range",
                          "real", "realpart", "rename", "repeat", "reshape",
                          "rrspacing", "rshift", "same_type_as", "scale",
                          "scan", "second", "selected_int_kind",
                          "selected_real_kind", "set_exponent", "shape",
                          "short", "sign", "signal", "sinh", "sin", "sleep",
                          "sngl", "spacing", "spread", "sqrt", "srand", "stat",
                          "sum", "symlnk", "system", "system_clock", "tan",
                          "tanh", "time", "tiny", "transfer", "transpose",
                          "trim", "ttynam", "ubound", "umask", "unlink",
                          "unpack", "verify", "xor", "zabs", "zcos", "zexp",
                          "zlog", "zsin", "zsqrt"];
    var keywords = wordRegexp(commonkeywords.concat(dataTypes));
    var builtins = wordRegexp(commonBuiltins);
    var blocks = wordRegexp(blockskeywords);
    var indentInfo = null;

    // tokenizers
    function tokenBase(stream, state) {
        // Handle scope changes
        if (stream.sol()) {
            var scopeOffset = state.scopes[0].offset;
            if (stream.eatSpace()) {
                var lineOffset = stream.indentation();
                if (lineOffset > scopeOffset) {
                    indentInfo = 'indent';
                } else if (lineOffset < scopeOffset) {
                    indentInfo = 'dedent';
                }
                return null;
            } else {
                if (scopeOffset > 0) {
                    dedent(stream, state);
                }
            }
        }
        if (stream.eatSpace()) {
            return null;
        }

        var ch = stream.peek();

        // Handle Comments
        if (ch === '!') {
            stream.skipToEnd();
            return 'comment';
        }

        // Handle Numbers Literals
        if (stream.match(/\b[\+\-]?[0-9]+\.?[0-9a-zA-Z_]*\b/, false)) {
            return 'number';
        }

        // Handle Strings
        if (stream.match(stringPrefixes)) {
            state.tokenize = tokenStringFactory(stream.current());
            return state.tokenize(stream, state);
        }

        if (stream.match(Operators) || stream.match(Operators2)){
            return 'operator';
        }
        if (stream.match(singleDelimiters)) {
            return null;
        }

        if (stream.match(keywords)) {
            return 'keyword';
        }

        if (stream.match(builtins)) {
            return 'builtin';
        }

        if (stream.match(identifiers)) {
            if (state.lastToken.match(blocks)){
                return 'def';
            }
            return 'variable';
        }

        // Handle non-detected items
        stream.next();
        return ERRORCLASS;
    }

    function tokenStringFactory(delimiter) {
        while ('rub'.indexOf(delimiter.charAt(0).toLowerCase()) >= 0) {
            delimiter = delimiter.substr(1);
        }
        var singleline = delimiter.length == 1;
        var OUTCLASS = 'string';

        function tokenString(stream, state) {
            while (!stream.eol()) {
                stream.eatWhile(/[^'"\\]/);
                if (stream.eat('\\')) {
                    stream.next();
                    if (singleline && stream.eol()) {
                        return OUTCLASS;
                    }
                } else if (stream.match(delimiter)) {
                    state.tokenize = tokenBase;
                    return OUTCLASS;
                } else {
                    stream.eat(/['"]/);
                }
            }
            if (singleline) {
                if (parserConf.singleLineStringErrors) {
                    return ERRORCLASS;
                } else {
                    state.tokenize = tokenBase;
                }
            }
            return OUTCLASS;
        }
        tokenString.isString = true;
        return tokenString;
    }

    function indent(stream, state, type) {
        type = type || 'py';
        var indentUnit = 0;
        if (type === 'py') {
            if (state.scopes[0].type !== 'py') {
                state.scopes[0].offset = stream.indentation();
                return;
            }
            for (var i = 0; i < state.scopes.length; ++i) {
                if (state.scopes[i].type === 'py') {
                    indentUnit = state.scopes[i].offset + conf.indentUnit;
                    break;
                }
            }
        } else {
            indentUnit = stream.column() + stream.current().length;
        }
        state.scopes.unshift({
            offset: indentUnit,
            type: type
        });
    }

    function dedent(stream, state, type) {
        type = type || 'py';
        if (state.scopes.length == 1) return;
        if (state.scopes[0].type === 'py') {
            var _indent = stream.indentation();
            var _indent_index = -1;
            for (var i = 0; i < state.scopes.length; ++i) {
                if (_indent === state.scopes[i].offset) {
                    _indent_index = i;
                    break;
                }
            }
            if (_indent_index === -1) {
                return true;
            }
            while (state.scopes[0].offset !== _indent) {
                state.scopes.shift();
            }
            return false;
        } else {
            if (type === 'py') {
                state.scopes[0].offset = stream.indentation();
                return false;
            } else {
                if (state.scopes[0].type != type) {
                    return true;
                }
                state.scopes.shift();
                return false;
            }
        }
    }


    var external = {
        startState: function(basecolumn) {
            return {
              tokenize: tokenBase,
              scopes: [{offset:basecolumn || 0, type:'py'}],
              lastStyle: null,
              lastToken: null,
              lambda: false,
              dedent: 0
          };
        },

        token: function(stream, state) {
            var style = state.tokenize(stream, state);

            state.lastStyle = style;

            var current = stream.current();
            if (current && style) {
                state.lastToken = current;
            }

            if (stream.eol() && state.lambda) {
                state.lambda = false;
            }
            return style;
        },

        indent: function(state) {
            if (state.tokenize != tokenBase) {
                return state.tokenize.isString ? CodeMirror.Pass : 0;
            }

            return state.scopes[0].offset;
        },

        lineComment: "!"
    };
    return external;
});

CodeMirror.defineMIME("text/x-fortran", "fortran");