/**
* Copyright 2013 Craig Campbell
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
* Rainbow is a simple code syntax highlighter
*
* @preserve @version 1.2
* @url rainbowco.de
*/
window['Rainbow'] = (function() {
/**
* array of replacements to process at the end
*
* @type {Object}
*/
var replacements = {},
/**
* an array of start and end positions of blocks to be replaced
*
* @type {Object}
*/
replacement_positions = {},
/**
* an array of the language patterns specified for each language
*
* @type {Object}
*/
language_patterns = {},
/**
* an array of languages and whether they should bypass the default patterns
*
* @type {Object}
*/
bypass_defaults = {},
/**
* processing level
*
* replacements are stored at this level so if there is a sub block of code
* (for example php inside of html) it runs at a different level
*
* @type {number}
*/
CURRENT_LEVEL = 0,
/**
* constant used to refer to the default language
*
* @type {number}
*/
DEFAULT_LANGUAGE = 0,
/**
* used as counters so we can selectively call setTimeout
* after processing a certain number of matches/replacements
*
* @type {number}
*/
match_counter = 0,
/**
* @type {number}
*/
replacement_counter = 0,
/**
* @type {null|string}
*/
global_class,
/**
* @type {null|Function}
*/
onHighlight;
/**
* adds a class to a given code block
*
* @param {Element} el
* @param {string} class_name class name to add
* @returns void
*/
function _addClass(el, class_name) {
el.className += el.className ? ' ' + class_name : class_name;
}
/**
* checks if a block has a given class
*
* @param {Element} el
* @param {string} class_name class name to check for
* @returns {boolean}
*/
function _hasClass(el, class_name) {
return (' ' + el.className + ' ').indexOf(' ' + class_name + ' ') > -1;
}
/**
* gets the language for this block of code
*
* @param {Element} block
* @returns {string|null}
*/
function _getLanguageForBlock(block) {
// if this doesn't have a language but the parent does then use that
// this means if for example you have: <pre data-language="php">
// with a bunch of <code> blocks inside then you do not have
// to specify the language for each block
var language = block.getAttribute('data-language') || block.parentNode.getAttribute('data-language');
// this adds support for specifying language via a css class
// you can use the Google Code Prettify style: <pre class="lang-php">
// or the HTML5 style: <pre><code class="language-php">
if (!language) {
var pattern = /\blang(?:uage)?-(\w+)/,
match = block.className.match(pattern) || block.parentNode.className.match(pattern);
if (match) {
language = match[1];
}
}
return language;
}
/**
* makes sure html entities are always used for tags
*
* @param {string} code
* @returns {string}
*/
function _htmlEntities(code) {
return code.replace(/</g, '<').replace(/>/g, '>').replace(/&(?![\w\#]+;)/g, '&');
}
/**
* determines if a new match intersects with an existing one
*
* @param {number} start1 start position of existing match
* @param {number} end1 end position of existing match
* @param {number} start2 start position of new match
* @param {number} end2 end position of new match
* @returns {boolean}
*/
function _intersects(start1, end1, start2, end2) {
if (start2 >= start1 && start2 < end1) {
return true;
}
return end2 > start1 && end2 < end1;
}
/**
* determines if two different matches have complete overlap with each other
*
* @param {number} start1 start position of existing match
* @param {number} end1 end position of existing match
* @param {number} start2 start position of new match
* @param {number} end2 end position of new match
* @returns {boolean}
*/
function _hasCompleteOverlap(start1, end1, start2, end2) {
// if the starting and end positions are exactly the same
// then the first one should stay and this one should be ignored
if (start2 == start1 && end2 == end1) {
return false;
}
return start2 <= start1 && end2 >= end1;
}
/**
* determines if the match passed in falls inside of an existing match
* this prevents a regex pattern from matching inside of a bigger pattern
*
* @param {number} start - start position of new match
* @param {number} end - end position of new match
* @returns {boolean}
*/
function _matchIsInsideOtherMatch(start, end) {
for (var key in replacement_positions[CURRENT_LEVEL]) {
key = parseInt(key, 10);
// if this block completely overlaps with another block
// then we should remove the other block and return false
if (_hasCompleteOverlap(key, replacement_positions[CURRENT_LEVEL][key], start, end)) {
delete replacement_positions[CURRENT_LEVEL][key];
delete replacements[CURRENT_LEVEL][key];
}
if (_intersects(key, replacement_positions[CURRENT_LEVEL][key], start, end)) {
return true;
}
}
return false;
}
/**
* takes a string of code and wraps it in a span tag based on the name
*
* @param {string} name name of the pattern (ie keyword.regex)
* @param {string} code block of code to wrap
* @returns {string}
*/
function _wrapCodeInSpan(name, code) {
return '<span class="' + name.replace(/\./g, ' ') + (global_class ? ' ' + global_class : '') + '">' + code + '</span>';
}
/**
* finds out the position of group match for a regular expression
*
* @see http://stackoverflow.com/questions/1985594/how-to-find-index-of-groups-in-match
*
* @param {Object} match
* @param {number} group_number
* @returns {number}
*/
function _indexOfGroup(match, group_number) {
var index = 0,
i;
for (i = 1; i < group_number; ++i) {
if (match[i]) {
index += match[i].length;
}
}
return index;
}
/**
* matches a regex pattern against a block of code
* finds all matches that should be processed and stores the positions
* of where they should be replaced within the string
*
* this is where pretty much all the work is done but it should not
* be called directly
*
* @param {RegExp} pattern
* @param {string} code
* @returns void
*/
function _processPattern(regex, pattern, code, callback)
{
if (typeof regex === "undefined" || regex === null) {
//console.warn("undefined regular expression")
return callback();
}
var match = regex.exec(code);
if (!match) {
return callback();
}
++match_counter;
// treat match 0 the same way as name
if (!pattern['name'] && typeof pattern['matches'][0] == 'string') {
pattern['name'] = pattern['matches'][0];
delete pattern['matches'][0];
}
var replacement = match[0],
start_pos = match.index,
end_pos = match[0].length + start_pos,
/**
* callback to process the next match of this pattern
*/
processNext = function() {
var nextCall = function() {
_processPattern(regex, pattern, code, callback);
};
// every 100 items we process let's call set timeout
// to let the ui breathe a little
//return match_counter % 100 > 0 ? nextCall() : setTimeout(nextCall, 0);
//ACE
//the timeout brings up errors when multiple blocks are being highlighted quickly one after one
return nextCall();
};
// if this is not a child match and it falls inside of another
// match that already happened we should skip it and continue processing
if (_matchIsInsideOtherMatch(start_pos, end_pos)) {
return processNext();
}
/**
* callback for when a match was successfully processed
*
* @param {string} replacement
* @returns void
*/
var onMatchSuccess = function(replacement) {
// if this match has a name then wrap it in a span tag
if (pattern['name']) {
replacement = _wrapCodeInSpan(pattern['name'], replacement);
}
// console.log('LEVEL', CURRENT_LEVEL, 'replace', match[0], 'with', replacement, 'at position', start_pos, 'to', end_pos);
// store what needs to be replaced with what at this position
if (!replacements[CURRENT_LEVEL]) {
replacements[CURRENT_LEVEL] = {};
replacement_positions[CURRENT_LEVEL] = {};
}
replacements[CURRENT_LEVEL][start_pos] = {
'replace': match[0],
'with': replacement
};
// store the range of this match so we can use it for comparisons
// with other matches later
replacement_positions[CURRENT_LEVEL][start_pos] = end_pos;
// process the next match
processNext();
},
// if this pattern has sub matches for different groups in the regex
// then we should process them one at a time by rerunning them through
// this function to generate the new replacement
//
// we run through them backwards because the match position of earlier
// matches will not change depending on what gets replaced in later
// matches
group_keys = keys(pattern['matches']),
/**
* callback for processing a sub group
*
* @param {number} i
* @param {Array} group_keys
* @param {Function} callback
*/
processGroup = function(i, group_keys, callback) {
if (i >= group_keys.length) {
return callback(replacement);
}
var processNextGroup = function() {
processGroup(++i, group_keys, callback);
},
block = match[group_keys[i]];
// if there is no match here then move on
if (!block) {
return processNextGroup();
}
var group = pattern['matches'][group_keys[i]],
language = group['language'],
/**
* process group is what group we should use to actually process
* this match group
*
* for example if the subgroup pattern looks like this
* 2: {
* 'name': 'keyword',
* 'pattern': /true/g
* }
*
* then we use that as is, but if it looks like this
*
* 2: {
* 'name': 'keyword',
* 'matches': {
* 'name': 'special',
* 'pattern': /whatever/g
* }
* }
*
* we treat the 'matches' part as the pattern and keep
* the name around to wrap it with later
*/
process_group = group['name'] && group['matches'] ? group['matches'] : group,
/**
* takes the code block matched at this group, replaces it
* with the highlighted block, and optionally wraps it with
* a span with a name
*
* @param {string} block
* @param {string} replace_block
* @param {string|null} match_name
*/
_replaceAndContinue = function(block, replace_block, match_name) {
replacement = _replaceAtPosition(_indexOfGroup(match, group_keys[i]), block, match_name ? _wrapCodeInSpan(match_name, replace_block) : replace_block, replacement);
processNextGroup();
};
// if this is a sublanguage go and process the block using that language
if (language) {
return _highlightBlockForLanguage(block, language, function(code) {
_replaceAndContinue(block, code);
});
}
// if this is a string then this match is directly mapped to selector
// so all we have to do is wrap it in a span and continue
if (typeof group === 'string') {
return _replaceAndContinue(block, block, group);
}
// the process group can be a single pattern or an array of patterns
// _processCodeWithPatterns always expects an array so we convert it here
_processCodeWithPatterns(block, process_group.length ? process_group : [process_group], function(code) {
_replaceAndContinue(block, code, group['matches'] ? group['name'] : 0);
});
};
processGroup(0, group_keys, onMatchSuccess);
}
/**
* should a language bypass the default patterns?
*
* if you call Rainbow.extend() and pass true as the third argument
* it will bypass the defaults
*/
function _bypassDefaultPatterns(language)
{
return bypass_defaults[language];
}
/**
* returns a list of regex patterns for this language
*
* @param {string} language
* @returns {Array}
*/
function _getPatternsForLanguage(language) {
var patterns = language_patterns[language] || [],
default_patterns = language_patterns[DEFAULT_LANGUAGE] || [];
return _bypassDefaultPatterns(language) ? patterns : patterns.concat(default_patterns);
}
/**
* substring replace call to replace part of a string at a certain position
*
* @param {number} position the position where the replacement should happen
* @param {string} replace the text we want to replace
* @param {string} replace_with the text we want to replace it with
* @param {string} code the code we are doing the replacing in
* @returns {string}
*/
function _replaceAtPosition(position, replace, replace_with, code) {
var sub_string = code.substr(position);
return code.substr(0, position) + sub_string.replace(replace, replace_with);
}
/**
* sorts an object by index descending
*
* @param {Object} object
* @return {Array}
*/
function keys(object) {
var locations = [],
replacement,
pos;
for(var location in object) {
if (object.hasOwnProperty(location)) {
locations.push(location);
}
}
// numeric descending
return locations.sort(function(a, b) {
return b - a;
});
}
/**
* processes a block of code using specified patterns
*
* @param {string} code
* @param {Array} patterns
* @returns void
*/
function _processCodeWithPatterns(code, patterns, callback)
{
// we have to increase the level here so that the
// replacements will not conflict with each other when
// processing sub blocks of code
++CURRENT_LEVEL;
// patterns are processed one at a time through this function
function _workOnPatterns(patterns, i)
{
// still have patterns to process, keep going
if (i < patterns.length) {
return _processPattern(patterns[i]['pattern'], patterns[i], code, function() {
_workOnPatterns(patterns, ++i);
});
}
// we are done processing the patterns
// process the replacements and update the DOM
_processReplacements(code, function(code) {
// when we are done processing replacements
// we are done at this level so we can go back down
delete replacements[CURRENT_LEVEL];
delete replacement_positions[CURRENT_LEVEL];
--CURRENT_LEVEL;
callback(code);
});
}
_workOnPatterns(patterns, 0);
}
/**
* process replacements in the string of code to actually update the markup
*
* @param {string} code the code to process replacements in
* @param {Function} onComplete what to do when we are done processing
* @returns void
*/
function _processReplacements(code, onComplete) {
/**
* processes a single replacement
*
* @param {string} code
* @param {Array} positions
* @param {number} i
* @param {Function} onComplete
* @returns void
*/
function _processReplacement(code, positions, i, onComplete) {
if (i < positions.length) {
++replacement_counter;
var pos = positions[i],
replacement = replacements[CURRENT_LEVEL][pos];
code = _replaceAtPosition(pos, replacement['replace'], replacement['with'], code);
// process next function
var next = function() {
_processReplacement(code, positions, ++i, onComplete);
};
// use a timeout every 250 to not freeze up the UI
//return replacement_counter % 250 > 0 ? next() : setTimeout(next, 0);
//ACE
//the timeout brings up errors when multiple blocks are being highlighted quickly one after one
return next()
}
onComplete(code);
}
var string_positions = keys(replacements[CURRENT_LEVEL]);
_processReplacement(code, string_positions, 0, onComplete);
}
/**
* takes a string of code and highlights it according to the language specified
*
* @param {string} code
* @param {string} language
* @param {Function} onComplete
* @returns void
*/
function _highlightBlockForLanguage(code, language, onComplete) {
var patterns = _getPatternsForLanguage(language);
_processCodeWithPatterns(_htmlEntities(code), patterns, onComplete);
}
/**
* highlight an individual code block
*
* @param {Array} code_blocks
* @param {number} i
* @returns void
*/
function _highlightCodeBlock(code_blocks, i, onComplete) {
if (i < code_blocks.length) {
var block = code_blocks[i],
language = _getLanguageForBlock(block);
if (!_hasClass(block, 'rainbow') && language) {
language = language.toLowerCase();
_addClass(block, 'rainbow');
return _highlightBlockForLanguage(block.innerHTML, language, function(code) {
block.innerHTML = code;
// reset the replacement arrays
replacements = {};
replacement_positions = {};
// if you have a listener attached tell it that this block is now highlighted
if (onHighlight) {
onHighlight(block, language);
}
// process the next block
setTimeout(function() {
_highlightCodeBlock(code_blocks, ++i, onComplete);
}, 0);
});
}
return _highlightCodeBlock(code_blocks, ++i, onComplete);
}
if (onComplete) {
onComplete();
}
}
/**
* start highlighting all the code blocks
*
* @returns void
*/
function _highlight(node, onComplete) {
// the first argument can be an Event or a DOM Element
// I was originally checking instanceof Event but that makes it break
// when using mootools
//
// @see https://github.com/ccampbell/rainbow/issues/32
//
node = node && typeof node.getElementsByTagName == 'function' ? node : document;
var pre_blocks = node.getElementsByTagName('pre'),
code_blocks = node.getElementsByTagName('code'),
i,
final_pre_blocks = [],
final_code_blocks = [];
// first loop through all pre blocks to find which ones to highlight
// also strip whitespace
for (i = 0; i < pre_blocks.length; ++i) {
// strip whitespace around code tags when they are inside of a pre tag
// this makes the themes look better because you can't accidentally
// add extra linebreaks at the start and end
//
// when the pre tag contains a code tag then strip any extra whitespace
// for example
// <pre>
// <code>var foo = true;</code>
// </pre>
//
// will become
// <pre><code>var foo = true;</code></pre>
//
// if you want to preserve whitespace you can use a pre tag on its own
// without a code tag inside of it
if (pre_blocks[i].getElementsByTagName('code').length) {
pre_blocks[i].innerHTML = pre_blocks[i].innerHTML.replace(/^\s+/, '').replace(/\s+$/, '');
continue;
}
// if the pre block has no code blocks then we are going to want to
// process it directly
final_pre_blocks.push(pre_blocks[i]);
}
// @see http://stackoverflow.com/questions/2735067/how-to-convert-a-dom-node-list-to-an-array-in-javascript
// we are going to process all <code> blocks
for (i = 0; i < code_blocks.length; ++i) {
final_code_blocks.push(code_blocks[i]);
}
_highlightCodeBlock(final_code_blocks.concat(final_pre_blocks), 0, onComplete);
}
/**
* public methods
*/
return {
/**
* extends the language pattern matches
*
* @param {*} language name of language
* @param {*} patterns array of patterns to add on
* @param {boolean|null} bypass if true this will bypass the default language patterns
*/
extend: function(language, patterns, bypass) {
// if there is only one argument then we assume that we want to
// extend the default language rules
if (arguments.length == 1) {
patterns = language;
language = DEFAULT_LANGUAGE;
}
bypass_defaults[language] = bypass;
language_patterns[language] = patterns.concat(language_patterns[language] || []);
},
/**
* call back to let you do stuff in your app after a piece of code has been highlighted
*
* @param {Function} callback
*/
onHighlight: function(callback) {
onHighlight = callback;
},
/**
* method to set a global class that will be applied to all spans
*
* @param {string} class_name
*/
addClass: function(class_name) {
global_class = class_name;
},
/**
* starts the magic rainbow
*
* @returns void
*/
color: function() {
// if you want to straight up highlight a string you can pass the string of code,
// the language, and a callback function
if (typeof arguments[0] == 'string') {
return _highlightBlockForLanguage(arguments[0], arguments[1], arguments[2]);
}
// if you pass a callback function then we rerun the color function
// on all the code and call the callback function on complete
if (typeof arguments[0] == 'function') {
return _highlight(0, arguments[0]);
}
// otherwise we use whatever node you passed in with an optional
// callback function as the second parameter
_highlight(arguments[0], arguments[1]);
}
};
}) ();
/**
* adds event listener to start highlighting
*/
(function() {
if (document.addEventListener) {
return document.addEventListener('DOMContentLoaded', Rainbow.color, false);
}
}) ();
// When using Google closure compiler in advanced mode some methods
// get renamed. This keeps a public reference to these methods so they can
// still be referenced from outside this library.
Rainbow["onHighlight"] = Rainbow.onHighlight;
Rainbow["addClass"] = Rainbow.addClass;
|