improve Ox.tokenize, add Ox.identify
This commit is contained in:
rolux
parent
08aedd34fd
commit
5a2ecca0f1
1 changed files with 275 additions and 325 deletions
|
|
@ -134,10 +134,9 @@ Ox.doc = (function() {
|
|||
tokens = [];
|
||||
Ox.tokenize(source).forEach(function(token) {
|
||||
var match;
|
||||
token.source = source.substr(token.offset, token.length);
|
||||
if (token.type == 'comment' && (
|
||||
match = re.multiline.exec(token.source)
|
||||
|| re.singleline.exec(token.source)
|
||||
match = re.multiline.exec(token.value)
|
||||
|| re.singleline.exec(token.value)
|
||||
)) {
|
||||
blocks.push(match[1]);
|
||||
tokens.push([]);
|
||||
|
|
@ -162,7 +161,9 @@ Ox.doc = (function() {
|
|||
if (/^[A-Z]/.test(item.name)) {
|
||||
// main item
|
||||
// include leading whitespace
|
||||
item.source = parseTokens(tokens[i]);
|
||||
item.source = parseTokens(tokens[i]).map(function(token) {
|
||||
return token.value;
|
||||
}).join('');
|
||||
item.line = source.slice(0, item.source[0].offset)
|
||||
.split('\n').length;
|
||||
items.push(item);
|
||||
|
|
@ -304,215 +305,21 @@ Ox.doc = (function() {
|
|||
}());
|
||||
|
||||
/*@
|
||||
Ox.minify <f> Minifies JavaScript
|
||||
(source) -> <s> Minified JavaScript
|
||||
(file, callback) -> <u> undefined
|
||||
source <s> JavaScript source
|
||||
file <s> JavaScript file
|
||||
callback <f> Callback function
|
||||
> Ox.minify('for (a in b)\n{\t\tc = void 0;\n}')
|
||||
'for(a in b)\n{c=void 0;}'
|
||||
> Ox.minify('return a; return 0; return "";')
|
||||
'return a;return 0;return"";'
|
||||
> Ox.minify('return\na;\nreturn\n0;\nreturn\n"";')
|
||||
'return\na;return\n0;return\n"";'
|
||||
Ox.identify <f> Returns the type of a JavaScript identifier
|
||||
(str) -> <s> Type
|
||||
Type can be <code>constant</code>, <code>identifier</code>,
|
||||
<code>keyword</code>, <code>method</code>, <code>object</code> or
|
||||
<code>property</code>
|
||||
@*/
|
||||
Ox.minify = function() {
|
||||
// see https://github.com/douglascrockford/JSMin/blob/master/README
|
||||
// and http://inimino.org/~inimino/blog/javascript_semicolons
|
||||
if (arguments.length == 1) {
|
||||
return minify(arguments[0]);
|
||||
} else {
|
||||
Ox.get(arguments[0], function(source) {
|
||||
arguments[1](minify(source));
|
||||
});
|
||||
}
|
||||
function isCommentOrLinebreakOrWhitespace(token) {
|
||||
return token.type == 'comment' || isLinebreakOrWhitespace(token);
|
||||
}
|
||||
function isIdentifierOrNumber(token) {
|
||||
return Ox.contains([
|
||||
'constant', 'identifier', 'keyword',
|
||||
'number', 'method', 'object', 'property'
|
||||
], token.type);
|
||||
}
|
||||
function isIdentifierOrNumberOrString(token) {
|
||||
return isIdentifierOrNumber(token) || token.type == 'string';
|
||||
}
|
||||
function isLinebreakOrWhitespace(token) {
|
||||
return Ox.contains(['linebreak', 'whitespace'], token.type);
|
||||
}
|
||||
function minify(source) {
|
||||
var tokens = Ox.tokenize(source),
|
||||
length = tokens.length,
|
||||
ret = '';
|
||||
function getValue(token) {
|
||||
return source.substr(token.offset, token.length);
|
||||
}
|
||||
tokens.forEach(function(token, i) {
|
||||
var next, nextToken, previousToken;
|
||||
if (isLinebreakOrWhitespace(token)) {
|
||||
previousToken = i == 0 ? null : tokens[i - 1];
|
||||
next = i + 1;
|
||||
while (
|
||||
next < length
|
||||
&& isCommentOrLinebreakOrWhitespace(tokens[next])
|
||||
) {
|
||||
next++;
|
||||
}
|
||||
nextToken = next == length ? null : tokens[next];
|
||||
}
|
||||
if (token.type == 'linebreak') {
|
||||
// replace a linebreak between two tokens that are identifiers
|
||||
// or numbers or strings or unary operators or grouping
|
||||
// operators with a single newline, otherwise remove it
|
||||
if (
|
||||
previousToken && nextToken && (
|
||||
isIdentifierOrNumberOrString(previousToken)
|
||||
|| Ox.contains([
|
||||
'++', '--', ')', ']', '}'
|
||||
], getValue(previousToken))
|
||||
) && (
|
||||
isIdentifierOrNumberOrString(nextToken)
|
||||
|| Ox.contains([
|
||||
'+', '-', '++', '--', '~', '!', '(', '[', '{'
|
||||
], getValue(nextToken))
|
||||
)
|
||||
) {
|
||||
ret += '\n';
|
||||
}
|
||||
} else if (token.type == 'whitespace') {
|
||||
// replace whitespace between two tokens that are identifiers or
|
||||
// numbers, or between a token that ends with "+" or "-" and one
|
||||
// that begins with "+" or "-", with a single space, otherwise
|
||||
// remove it
|
||||
if (
|
||||
previousToken && nextToken && ((
|
||||
isIdentifierOrNumber(previousToken)
|
||||
&& isIdentifierOrNumber(nextToken)
|
||||
) || (
|
||||
Ox.contains([
|
||||
'+', '-', '++', '--'
|
||||
], getValue(previousToken))
|
||||
&& Ox.contains([
|
||||
'+', '-', '++', '--'
|
||||
], getValue(nextToken))
|
||||
))
|
||||
) {
|
||||
ret += ' ';
|
||||
}
|
||||
} else if (token.type != 'comment') {
|
||||
// remove comments and leave all other tokens untouched
|
||||
ret += getValue(token);
|
||||
}
|
||||
});
|
||||
return ret;
|
||||
}
|
||||
};
|
||||
|
||||
/*@
|
||||
Ox.test <f> Takes JavaScript, runs inline tests, returns results
|
||||
@*/
|
||||
Ox.test = function(file, callback) {
|
||||
Ox.doc(file, function(items) {
|
||||
var tests = [];
|
||||
items.forEach(function(item) {
|
||||
item.examples && item.examples.some(function(example) {
|
||||
return example.result;
|
||||
}) && item.examples.forEach(function(example) {
|
||||
Ox.Log('TEST', example.statement);
|
||||
var actual = eval(example.statement);
|
||||
if (example.result) {
|
||||
tests.push({
|
||||
actual: JSON.stringify(actual),
|
||||
expected: example.result,
|
||||
name: item.name,
|
||||
section: item.section,
|
||||
statement: example.statement,
|
||||
passed: Ox.isEqual(eval(
|
||||
'Ox.test.result = ' + example.result
|
||||
), actual)
|
||||
});
|
||||
}
|
||||
});
|
||||
});
|
||||
callback(tests);
|
||||
});
|
||||
};
|
||||
|
||||
/*@
|
||||
Ox.tokenize <f> Tokenizes JavaScript
|
||||
(source) -> <[o]> Array of tokens
|
||||
length <n> Length of the token
|
||||
offset <n> Offset of the token
|
||||
type <s> Type of the token
|
||||
Type can be <code>"comment"</code>, <code>"constant"</code>,
|
||||
<code>"identifier"</code>, <code>"keyword"</code>,
|
||||
<code>"linebreak"</code>, <code>"method"</code>,
|
||||
<code>"number"</code>, <code>"object"</code>,
|
||||
<code>"operator"</code>, <code>"property"</code>,
|
||||
<code>"regexp"</code>, <code>"string"</code>
|
||||
or <code>"whitespace"</code>
|
||||
source <s> JavaScript source code
|
||||
@*/
|
||||
// FIXME: constant/method/object/property is of interest
|
||||
// for syntax highlighting, but may not belong here
|
||||
// FIXME: backport python version
|
||||
// FIXME: numbers (hex, exp, etc.)
|
||||
Ox.tokenize = (function() {
|
||||
|
||||
// see https://github.com/mozilla/narcissus/blob/master/lib/jslex.js
|
||||
// and https://developer.mozilla.org/en/JavaScript/Reference
|
||||
|
||||
var identifier = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ$_',
|
||||
linebreak = '\n\r',
|
||||
number = '0123456789',
|
||||
operator = [
|
||||
// arithmetic
|
||||
'+', '-', '*', '/', '%', '++', '--',
|
||||
// assignment
|
||||
'=', '+=', '-=', '*=', '/=', '%=',
|
||||
'&=', '|=', '^=', '<<=', '>>=', '>>>=',
|
||||
// bitwise
|
||||
'&', '|', '^', '~', '<<', '>>', '>>>',
|
||||
// comparison
|
||||
'==', '!=', '===', '!==', '>', '>=', '<', '<=',
|
||||
// conditional
|
||||
'?', ':',
|
||||
// grouping
|
||||
'(', ')', '[', ']', '{', '}',
|
||||
// logical
|
||||
'&&', '||', '!',
|
||||
// other
|
||||
'.', ',', ';'
|
||||
],
|
||||
regexp = 'abcdefghijklmnopqrstuvwxyz',
|
||||
string = '\'"',
|
||||
whitespace = ' \t',
|
||||
word = {
|
||||
Ox.identify = (function() {
|
||||
// see https://developer.mozilla.org/en/JavaScript/Reference
|
||||
var identifiers = {
|
||||
constant: [
|
||||
// Math
|
||||
'E', 'LN2', 'LN10', 'LOG2E', 'LOG10E', 'PI', 'SQRT1_2', 'SQRT2',
|
||||
// Number
|
||||
'MAX_VALUE', 'MIN_VALUE', 'NEGATIVE_INFINITY', 'POSITIVE_INFINITY'
|
||||
],
|
||||
keyword: [
|
||||
'break',
|
||||
'case', 'catch', 'class', 'const', 'continue',
|
||||
'debugger', 'default', 'delete', 'do',
|
||||
'else', 'enum', 'export', 'extends',
|
||||
'false', 'finally', 'for', 'function',
|
||||
'if', 'implements', 'import', 'in', 'instanceof', 'interface',
|
||||
'let', 'module',
|
||||
'new', 'null',
|
||||
'package', 'private', 'protected', 'public',
|
||||
'return',
|
||||
'super', 'switch', 'static',
|
||||
'this', 'throw', 'true', 'try', 'typeof',
|
||||
'var', 'void',
|
||||
'yield',
|
||||
'while', 'with',
|
||||
],
|
||||
method: [
|
||||
// Array
|
||||
'concat',
|
||||
|
|
@ -580,7 +387,8 @@ Ox.tokenize = (function() {
|
|||
'match',
|
||||
'replace',
|
||||
'search', 'slice', 'split', 'substr', 'substring',
|
||||
'toLocaleLowerCase', 'toLocaleUpperCase', 'toLowerCase', 'toUpperCase', 'trim',
|
||||
'toLocaleLowerCase', 'toLocaleUpperCase',
|
||||
'toLowerCase', 'toUpperCase', 'trim',
|
||||
// Window
|
||||
'addEventListener', 'alert', 'atob',
|
||||
'blur', 'btoa',
|
||||
|
|
@ -637,137 +445,279 @@ Ox.tokenize = (function() {
|
|||
'toolbar', 'top'
|
||||
]
|
||||
};
|
||||
|
||||
return function(source) {
|
||||
|
||||
source = source.replace(/\r\n/g, '\n').replace(/\r/g, '\n');
|
||||
|
||||
var cursor = 0,
|
||||
tokenize = {
|
||||
comment: function() {
|
||||
while (char = source[++cursor]) {
|
||||
if (next == '/' && char == '\n') {
|
||||
break;
|
||||
} else if (next == '*' && char + source[cursor + 1] == '*/') {
|
||||
cursor += 2;
|
||||
break;
|
||||
}
|
||||
}
|
||||
},
|
||||
identifier: function() {
|
||||
var str;
|
||||
while ((identifier + number).indexOf(source[++cursor]) > -1) {}
|
||||
str = source.slice(start, cursor);
|
||||
Ox.forEach(word, function(value, key) {
|
||||
if (value.indexOf(str) > -1) {
|
||||
type = key;
|
||||
return function(identifier) {
|
||||
var ret;
|
||||
if (Ox.KEYWORDS.indexOf(identifier) > -1) {
|
||||
ret = 'keyword'
|
||||
} else {
|
||||
ret = 'identifier'
|
||||
Ox.forEach(identifiers, function(words, type) {
|
||||
if (words.indexOf(identifier) > -1) {
|
||||
ret = type;
|
||||
Ox.Break();
|
||||
}
|
||||
});
|
||||
},
|
||||
linebreak: function() {
|
||||
while (linebreak.indexOf(source[++cursor]) > -1) {}
|
||||
},
|
||||
number: function() {
|
||||
while ((number + '.').indexOf(source[++cursor]) > -1) {}
|
||||
},
|
||||
operator: function() {
|
||||
while (operator.indexOf(char += source[++cursor]) > -1) {}
|
||||
},
|
||||
regexp: function() {
|
||||
while ((char = source[++cursor]) != '/') {
|
||||
char == '\\' && ++cursor;
|
||||
if (cursor == source.length) {
|
||||
break;
|
||||
}
|
||||
}
|
||||
while (regexp.indexOf(source[++cursor]) > -1) {}
|
||||
},
|
||||
string: function() {
|
||||
var delimiter = char;
|
||||
while ((char = source[++cursor]) != delimiter) {
|
||||
char == '\\' && ++cursor;
|
||||
if (cursor == source.length) {
|
||||
break;
|
||||
}
|
||||
}
|
||||
++cursor;
|
||||
},
|
||||
whitespace: function() {
|
||||
while (whitespace.indexOf(source[++cursor]) > -1) {}
|
||||
}
|
||||
},
|
||||
tokens = [],
|
||||
type;
|
||||
return ret;
|
||||
};
|
||||
})();
|
||||
|
||||
while (cursor < source.length) {
|
||||
var char = source[cursor],
|
||||
next = source[cursor + 1],
|
||||
start = cursor;
|
||||
if (char == '/' && (next == '/' || next == '*')) {
|
||||
type = 'comment';
|
||||
} else if (identifier.indexOf(char) > -1) {
|
||||
type = 'identifier';
|
||||
} else if (linebreak.indexOf(char) > -1) {
|
||||
type = 'linebreak';
|
||||
} else if (number.indexOf(char) > -1) {
|
||||
type = 'number';
|
||||
} else if (string.indexOf(char) > -1) {
|
||||
type = 'string';
|
||||
} else if (whitespace.indexOf(char) > -1) {
|
||||
type = 'whitespace';
|
||||
} else if (char == '/') {
|
||||
type = isRegExp() ? 'regexp' : 'operator';
|
||||
} else if (operator.indexOf(char) > -1) {
|
||||
type = 'operator';
|
||||
}
|
||||
tokenize[type]();
|
||||
tokens.push({
|
||||
length: cursor - start,
|
||||
offset: start,
|
||||
type: type
|
||||
/*@
|
||||
Ox.minify <f> Minifies JavaScript
|
||||
(source) -> <s> Minified JavaScript
|
||||
(file, callback) -> <u> undefined
|
||||
source <s> JavaScript source
|
||||
file <s> JavaScript file
|
||||
callback <f> Callback function
|
||||
> Ox.minify('for (a in b)\n{\t\tc = void 0;\n}')
|
||||
'for(a in b)\n{c=void 0;}'
|
||||
> Ox.minify('return a; return 0; return "";')
|
||||
'return a;return 0;return"";'
|
||||
> Ox.minify('return\na;\nreturn\n0;\nreturn\n"";')
|
||||
'return\na;return\n0;return\n"";'
|
||||
@*/
|
||||
Ox.minify = function() {
|
||||
// see https://github.com/douglascrockford/JSMin/blob/master/README
|
||||
// and http://inimino.org/~inimino/blog/javascript_semicolons
|
||||
if (arguments.length == 1) {
|
||||
return minify(arguments[0]);
|
||||
} else {
|
||||
Ox.get(arguments[0], function(source) {
|
||||
arguments[1](minify(source));
|
||||
});
|
||||
}
|
||||
|
||||
function isRegExp() {
|
||||
// checks if a forward slash is the beginning of a regexp,
|
||||
// as opposed to the beginning of an operator
|
||||
// see http://www.mozilla.org/js/language/js20-2000-07/rationale/syntax.html#regular-expressions
|
||||
var index = tokens.length,
|
||||
isRegExp = false,
|
||||
offset = 0,
|
||||
prevToken,
|
||||
prevString;
|
||||
// scan back to the previous significant token,
|
||||
// or the beginning of the source
|
||||
function minify(source) {
|
||||
var tokens = Ox.tokenize(source),
|
||||
length = tokens.length,
|
||||
ret = '';
|
||||
tokens.forEach(function(token, i) {
|
||||
var next, nextToken, prevToken;
|
||||
if (['linebreak', 'whitespace'].indexOf(token.type) > -1) {
|
||||
prevToken = i == 0 ? null : tokens[i - 1];
|
||||
next = i + 1;
|
||||
while (
|
||||
tokens[--index] !== void 0 && [
|
||||
'comment', 'linebreak', 'whitespace'
|
||||
].indexOf(tokens[index].type) > -1
|
||||
next < length && ['comment', 'linebreak', 'whitespace']
|
||||
.indexOf(tokens[next].type) > -1
|
||||
) {
|
||||
offset += tokens[index].length;
|
||||
next++;
|
||||
}
|
||||
if (typeof tokens[index] == 'undefined') {
|
||||
// source begins with forward slash
|
||||
nextToken = next == length ? null : tokens[next];
|
||||
}
|
||||
if (token.type == 'linebreak') {
|
||||
// replace a linebreak between two tokens that are identifiers
|
||||
// or numbers or strings or unary operators or grouping
|
||||
// operators with a single newline, otherwise remove it
|
||||
if (
|
||||
prevToken && nextToken && (
|
||||
['identifier', 'number', 'string'].indexOf(prevToken.type) > -1
|
||||
|| ['++', '--', ')', ']', '}'].indexOf(prevToken.value) > -1
|
||||
) && (
|
||||
['identifier', 'number', 'string'].indexOf(nextToken.type) > -1
|
||||
|| ['+', '-', '++', '--', '~', '!', '(', '[', '{'].indexOf(nextToken.value) > -1
|
||||
)
|
||||
) {
|
||||
ret += '\n';
|
||||
}
|
||||
} else if (token.type == 'whitespace') {
|
||||
// replace whitespace between two tokens that are identifiers or
|
||||
// numbers, or between a token that ends with "+" or "-" and one
|
||||
// that begins with "+" or "-", with a single space, otherwise
|
||||
// remove it
|
||||
if (
|
||||
prevToken && nextToken && ((
|
||||
['identifier', 'number'].indexOf(prevToken.type) > -1
|
||||
&& ['identifier', 'number'].indexOf(nextToken.type) > -1
|
||||
) || (
|
||||
['+', '-', '++', '--'].indexOf(prevToken.value) > -1
|
||||
&& ['+', '-', '++', '--'].indexOf(nextToken.value) > -1
|
||||
))
|
||||
) {
|
||||
ret += ' ';
|
||||
}
|
||||
} else if (token.type != 'comment') {
|
||||
// remove comments and leave all other tokens untouched
|
||||
ret += token.value;
|
||||
}
|
||||
});
|
||||
return ret;
|
||||
}
|
||||
};
|
||||
|
||||
/*@
|
||||
Ox.test <f> Takes JavaScript, runs inline tests, returns results
|
||||
@*/
|
||||
Ox.test = function(file, callback) {
|
||||
Ox.doc(file, function(items) {
|
||||
var tests = [];
|
||||
items.forEach(function(item) {
|
||||
item.examples && item.examples.some(function(example) {
|
||||
return example.result;
|
||||
}) && item.examples.forEach(function(example) {
|
||||
Ox.Log('TEST', example.statement);
|
||||
var actual = eval(example.statement);
|
||||
if (example.result) {
|
||||
tests.push({
|
||||
actual: JSON.stringify(actual),
|
||||
expected: example.result,
|
||||
name: item.name,
|
||||
section: item.section,
|
||||
statement: example.statement,
|
||||
passed: Ox.isEqual(eval(
|
||||
'Ox.test.result = ' + example.result
|
||||
), actual)
|
||||
});
|
||||
}
|
||||
});
|
||||
});
|
||||
callback(tests);
|
||||
});
|
||||
};
|
||||
|
||||
/*@
|
||||
Ox.tokenize <f> Tokenizes JavaScript
|
||||
(source) -> <[o]> Array of tokens
|
||||
column <n> Column of the token
|
||||
line <n> Line of the token
|
||||
type <s> Type of the token
|
||||
Type can be <code>"comment"</code>, <code>"identifier"</code>,
|
||||
<code>"linebreak"</code>, <code>"number"</code>,
|
||||
<code>"operator"</code>, <code>"regexp"</code>,
|
||||
<code>"string"</code> or <code>"whitespace"</code>
|
||||
value <s> Value of the token
|
||||
source <s> JavaScript source code
|
||||
@*/
|
||||
// FIXME: numbers (hex, exp, etc.)
|
||||
Ox.tokenize = (function() {
|
||||
|
||||
// see https://github.com/mozilla/narcissus/blob/master/lib/lexer.js
|
||||
|
||||
var comment = ['//', '/*'],
|
||||
identifier = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ$_',
|
||||
linebreak = '\n\r',
|
||||
number = '0123456789',
|
||||
operator = [
|
||||
// arithmetic
|
||||
'+', '-', '*', '/', '%', '++', '--',
|
||||
// assignment
|
||||
'=', '+=', '-=', '*=', '/=', '%=',
|
||||
'&=', '|=', '^=', '<<=', '>>=', '>>>=',
|
||||
// bitwise
|
||||
'&', '|', '^', '~', '<<', '>>', '>>>',
|
||||
// comparison
|
||||
'==', '!=', '===', '!==', '>', '>=', '<', '<=',
|
||||
// conditional
|
||||
'?', ':',
|
||||
// grouping
|
||||
'(', ')', '[', ']', '{', '}',
|
||||
// logical
|
||||
'&&', '||', '!',
|
||||
// other
|
||||
'.', ',', ';'
|
||||
],
|
||||
regexp = 'abcdefghijklmnopqrstuvwxyz',
|
||||
string = '\'"',
|
||||
whitespace = ' \t';
|
||||
|
||||
function isRegExp(tokens) {
|
||||
// Returns true if the current token is the beginning of a RegExp, as
|
||||
// opposed to the beginning of an operator
|
||||
var i = tokens.length - 1, isRegExp, token
|
||||
// Scan back to the previous significant token, or to the beginning of
|
||||
// the source
|
||||
while (i >= 0 && [
|
||||
'comment', 'linebreak', 'whitespace'
|
||||
].indexOf(tokens[i].type) > -1) {
|
||||
i--;
|
||||
}
|
||||
if (i == -1) {
|
||||
// Source begins with a forward slash
|
||||
isRegExp = true;
|
||||
} else {
|
||||
prevToken = tokens[index];
|
||||
prevString = source.substr(
|
||||
cursor - prevToken.length - offset, prevToken.length
|
||||
);
|
||||
token = tokens[i];
|
||||
isRegExp = (
|
||||
prevToken.type == 'keyword'
|
||||
&& ['false', 'null', 'true'].indexOf(prevString) == -1
|
||||
token.type == 'identifier'
|
||||
&& Ox.identify(token.value) == 'keyword'
|
||||
&& ['false', 'null', 'true'].indexOf(token.value) == -1
|
||||
) || (
|
||||
prevToken.type == 'operator'
|
||||
&& ['++', '--', ')', ']', '}'].indexOf(prevString) == -1
|
||||
);
|
||||
token.type == 'operator'
|
||||
&& ['++', '--', ')', ']', '}'].indexOf(token.value) == -1
|
||||
)
|
||||
}
|
||||
return isRegExp;
|
||||
}
|
||||
|
||||
return function(source) {
|
||||
var char,
|
||||
column = 0,
|
||||
cursor = 0,
|
||||
delimiter,
|
||||
length = source.length,
|
||||
line = 0,
|
||||
lines,
|
||||
next,
|
||||
tokens = [],
|
||||
start,
|
||||
type,
|
||||
value;
|
||||
source = source.replace(/\r\n/g, '\n').replace(/\r/g, '\n');
|
||||
while (cursor < length) {
|
||||
start = cursor;
|
||||
char = source[cursor];
|
||||
if (comment.indexOf(delimiter = char + source[cursor + 1]) > -1) {
|
||||
type = 'comment';
|
||||
++cursor;
|
||||
while (char = source[++cursor]) {
|
||||
if (delimiter == '//' && char == '\n') {
|
||||
break;
|
||||
} else if (delimiter == '/*' && char + source[cursor + 1] == '*/') {
|
||||
cursor += 2;
|
||||
break;
|
||||
}
|
||||
}
|
||||
} else if (identifier.indexOf(char) > -1) {
|
||||
type = 'identifier';
|
||||
while ((identifier + number).indexOf(source[++cursor]) > -1) {}
|
||||
} else if (linebreak.indexOf(char) > -1) {
|
||||
type = 'linebreak';
|
||||
while (linebreak.indexOf(source[++cursor]) > -1) {}
|
||||
} else if (number.indexOf(char) > -1) {
|
||||
type = 'number';
|
||||
while ((number + '.').indexOf(source[++cursor]) > -1) {}
|
||||
} else if (char == '/' && isRegExp(tokens)) {
|
||||
type = 'regexp';
|
||||
while ((char = source[++cursor]) != '/' && cursor < length) {
|
||||
char == '\\' && ++cursor;
|
||||
}
|
||||
while (regexp.indexOf(source[++cursor]) > -1) {}
|
||||
} else if (operator.indexOf(char) > -1) {
|
||||
type = 'operator';
|
||||
while (operator.indexOf(char += source[++cursor]) > -1 && cursor < length) {}
|
||||
} else if (string.indexOf(delimiter = char) > -1) {
|
||||
type = 'string';
|
||||
while ((char = source[++cursor]) != delimiter && cursor < length) {
|
||||
char == '\\' && ++cursor;
|
||||
}
|
||||
++cursor;
|
||||
} else if (whitespace.indexOf(char) > -1) {
|
||||
type = 'whitespace';
|
||||
while (whitespace.indexOf(source[++cursor]) > -1) {}
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
value = source.slice(start, cursor);
|
||||
tokens.push({column: column, line: line, type: type, value: value});
|
||||
if (type == 'comment') {
|
||||
lines = value.split('\n');
|
||||
column = lines[lines.length - 1].length;
|
||||
line += lines.length - 1;
|
||||
} else if (type == 'linebreak') {
|
||||
column = 0;
|
||||
line += value.length;
|
||||
} else {
|
||||
column += value.length;
|
||||
}
|
||||
}
|
||||
return tokens;
|
||||
|
||||
};
|
||||
|
||||
}());
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue