You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
461 lines
12 KiB
461 lines
12 KiB
define([ |
|
"dojo/_base/lang", |
|
"dojo/_base/array", |
|
"dojo/dom", |
|
"dojo/dom-class" |
|
], function(lang, array, dom, domClass){ |
|
|
|
|
|
var dh = lang.getObject("dojox.highlight", true), |
|
C_NUMBER_RE = '\\b(0x[A-Za-z0-9]+|\\d+(\\.\\d+)?)' |
|
; |
|
/*===== |
|
dh = { |
|
// summary: |
|
// Syntax highlighting with language auto-detection package |
|
// description: |
|
// Syntax highlighting with language auto-detection package. |
|
// Released under CLA by the Dojo Toolkit, original BSD release |
|
// available from: http://softwaremaniacs.org/soft/highlight/ |
|
}; |
|
=====*/ |
|
|
|
dh.languages = dh.languages || {}; |
|
// constants |
|
|
|
dh.constants = { |
|
IDENT_RE: '[a-zA-Z][a-zA-Z0-9_]*', |
|
UNDERSCORE_IDENT_RE: '[a-zA-Z_][a-zA-Z0-9_]*', |
|
NUMBER_RE: '\\b\\d+(\\.\\d+)?', |
|
C_NUMBER_RE: C_NUMBER_RE, |
|
// Common modes |
|
APOS_STRING_MODE: { |
|
className: 'string', |
|
begin: '\'', end: '\'', |
|
illegal: '\\n', |
|
contains: ['escape'], |
|
relevance: 0 |
|
}, |
|
QUOTE_STRING_MODE: { |
|
className: 'string', |
|
begin: '"', |
|
end: '"', |
|
illegal: '\\n', |
|
contains: ['escape'], |
|
relevance: 0 |
|
}, |
|
BACKSLASH_ESCAPE: { |
|
className: 'escape', |
|
begin: '\\\\.', end: '^', |
|
relevance: 0 |
|
}, |
|
C_LINE_COMMENT_MODE: { |
|
className: 'comment', |
|
begin: '//', end: '$', |
|
relevance: 0 |
|
}, |
|
C_BLOCK_COMMENT_MODE: { |
|
className: 'comment', |
|
begin: '/\\*', end: '\\*/' |
|
}, |
|
HASH_COMMENT_MODE: { |
|
className: 'comment', |
|
begin: '#', end: '$' |
|
}, |
|
C_NUMBER_MODE: { |
|
className: 'number', |
|
begin: C_NUMBER_RE, end: '^', |
|
relevance: 0 |
|
} |
|
}; |
|
|
|
// utilities |
|
|
|
function esc(value){ |
|
return value.replace(/&/gm, '&').replace(/</gm, '<').replace(/>/gm, '>'); |
|
} |
|
|
|
function verifyText(block){ |
|
return array.every(block.childNodes, function(node){ |
|
return node.nodeType == 3 || String(node.nodeName).toLowerCase() == 'br'; |
|
}); |
|
} |
|
|
|
function blockText(block){ |
|
var result = []; |
|
array.forEach(block.childNodes, function(node){ |
|
if(node.nodeType == 3){ |
|
result.push(node.nodeValue); |
|
}else if(String(node.nodeName).toLowerCase() == 'br'){ |
|
result.push("\n"); |
|
}else{ |
|
throw 'Complex markup'; |
|
} |
|
}); |
|
return result.join(""); |
|
} |
|
|
|
function buildKeywordGroups(mode){ |
|
if(!mode.keywordGroups){ |
|
for(var key in mode.keywords){ |
|
var kw = mode.keywords[key]; |
|
if(kw instanceof Object){ // dojo.isObject? |
|
mode.keywordGroups = mode.keywords; |
|
}else{ |
|
mode.keywordGroups = {keyword: mode.keywords}; |
|
} |
|
break; |
|
} |
|
} |
|
} |
|
|
|
function buildKeywords(language){ |
|
if(language.defaultMode && language.modes){ |
|
buildKeywordGroups(language.defaultMode); |
|
array.forEach(language.modes, buildKeywordGroups); |
|
} |
|
} |
|
|
|
// main object |
|
|
|
var Highlighter = function(langName, textBlock){ |
|
// initialize the state |
|
this.langName = langName; |
|
this.lang = dh.languages[langName]; |
|
this.modes = [this.lang.defaultMode]; |
|
this.relevance = 0; |
|
this.keywordCount = 0; |
|
this.result = []; |
|
|
|
// build resources lazily |
|
if(!this.lang.defaultMode.illegalRe){ |
|
this.buildRes(); |
|
buildKeywords(this.lang); |
|
} |
|
|
|
// run the algorithm |
|
try{ |
|
this.highlight(textBlock); |
|
this.result = this.result.join(""); |
|
}catch(e){ |
|
if(e == 'Illegal'){ |
|
this.relevance = 0; |
|
this.keywordCount = 0; |
|
this.partialResult = this.result.join(""); |
|
this.result = esc(textBlock); |
|
}else{ |
|
throw e; |
|
} |
|
} |
|
}; |
|
|
|
lang.extend(Highlighter, { |
|
buildRes: function(){ |
|
array.forEach(this.lang.modes, function(mode){ |
|
if(mode.begin){ |
|
mode.beginRe = this.langRe('^' + mode.begin); |
|
} |
|
if(mode.end){ |
|
mode.endRe = this.langRe('^' + mode.end); |
|
} |
|
if(mode.illegal){ |
|
mode.illegalRe = this.langRe('^(?:' + mode.illegal + ')'); |
|
} |
|
}, this); |
|
this.lang.defaultMode.illegalRe = this.langRe('^(?:' + this.lang.defaultMode.illegal + ')'); |
|
}, |
|
|
|
subMode: function(lexeme){ |
|
var classes = this.modes[this.modes.length - 1].contains; |
|
if(classes){ |
|
var modes = this.lang.modes; |
|
for(var i = 0; i < classes.length; ++i){ |
|
var className = classes[i]; |
|
for(var j = 0; j < modes.length; ++j){ |
|
var mode = modes[j]; |
|
if(mode.className == className && mode.beginRe.test(lexeme)){ return mode; } |
|
} |
|
} |
|
} |
|
return null; |
|
}, |
|
|
|
endOfMode: function(lexeme){ |
|
for(var i = this.modes.length - 1; i >= 0; --i){ |
|
var mode = this.modes[i]; |
|
if(mode.end && mode.endRe.test(lexeme)){ return this.modes.length - i; } |
|
if(!mode.endsWithParent){ break; } |
|
} |
|
return 0; |
|
}, |
|
|
|
isIllegal: function(lexeme){ |
|
var illegalRe = this.modes[this.modes.length - 1].illegalRe; |
|
return illegalRe && illegalRe.test(lexeme); |
|
}, |
|
|
|
|
|
langRe: function(value, global){ |
|
var mode = 'm' + (this.lang.case_insensitive ? 'i' : '') + (global ? 'g' : ''); |
|
return new RegExp(value, mode); |
|
}, |
|
|
|
buildTerminators: function(){ |
|
var mode = this.modes[this.modes.length - 1], |
|
terminators = {}; |
|
if(mode.contains){ |
|
array.forEach(this.lang.modes, function(lmode){ |
|
if(array.indexOf(mode.contains, lmode.className) >= 0){ |
|
terminators[lmode.begin] = 1; |
|
} |
|
}); |
|
} |
|
for(var i = this.modes.length - 1; i >= 0; --i){ |
|
var m = this.modes[i]; |
|
if(m.end){ terminators[m.end] = 1; } |
|
if(!m.endsWithParent){ break; } |
|
} |
|
if(mode.illegal){ terminators[mode.illegal] = 1; } |
|
var t = []; |
|
for(i in terminators){ t.push(i); } |
|
mode.terminatorsRe = this.langRe("(" + t.join("|") + ")"); |
|
}, |
|
|
|
eatModeChunk: function(value, index){ |
|
var mode = this.modes[this.modes.length - 1]; |
|
|
|
// create terminators lazily |
|
if(!mode.terminatorsRe){ |
|
this.buildTerminators(); |
|
} |
|
|
|
value = value.substr(index); |
|
var match = mode.terminatorsRe.exec(value); |
|
if(!match){ |
|
return { |
|
buffer: value, |
|
lexeme: "", |
|
end: true |
|
}; |
|
} |
|
return { |
|
buffer: match.index ? value.substr(0, match.index) : "", |
|
lexeme: match[0], |
|
end: false |
|
}; |
|
}, |
|
|
|
keywordMatch: function(mode, match){ |
|
var matchStr = match[0]; |
|
if(this.lang.case_insensitive){ matchStr = matchStr.toLowerCase(); } |
|
for(var className in mode.keywordGroups){ |
|
if(matchStr in mode.keywordGroups[className]){ return className; } |
|
} |
|
return ""; |
|
}, |
|
|
|
buildLexemes: function(mode){ |
|
var lexemes = {}; |
|
array.forEach(mode.lexems, function(lexeme){ |
|
lexemes[lexeme] = 1; |
|
}); |
|
var t = []; |
|
for(var i in lexemes){ t.push(i); } |
|
mode.lexemsRe = this.langRe("(" + t.join("|") + ")", true); |
|
}, |
|
|
|
processKeywords: function(buffer){ |
|
var mode = this.modes[this.modes.length - 1]; |
|
if(!mode.keywords || !mode.lexems){ |
|
return esc(buffer); |
|
} |
|
|
|
// create lexemes lazily |
|
if(!mode.lexemsRe){ |
|
this.buildLexemes(mode); |
|
} |
|
|
|
mode.lexemsRe.lastIndex = 0; |
|
var result = [], lastIndex = 0, |
|
match = mode.lexemsRe.exec(buffer); |
|
while(match){ |
|
result.push(esc(buffer.substr(lastIndex, match.index - lastIndex))); |
|
var keywordM = this.keywordMatch(mode, match); |
|
if(keywordM){ |
|
++this.keywordCount; |
|
result.push('<span class="'+ keywordM +'">' + esc(match[0]) + '</span>'); |
|
}else{ |
|
result.push(esc(match[0])); |
|
} |
|
lastIndex = mode.lexemsRe.lastIndex; |
|
match = mode.lexemsRe.exec(buffer); |
|
} |
|
result.push(esc(buffer.substr(lastIndex, buffer.length - lastIndex))); |
|
return result.join(""); |
|
}, |
|
|
|
processModeInfo: function(buffer, lexeme, end) { |
|
var mode = this.modes[this.modes.length - 1]; |
|
if(end){ |
|
this.result.push(this.processKeywords(mode.buffer + buffer)); |
|
return; |
|
} |
|
if(this.isIllegal(lexeme)){ throw 'Illegal'; } |
|
var newMode = this.subMode(lexeme); |
|
if(newMode){ |
|
mode.buffer += buffer; |
|
this.result.push(this.processKeywords(mode.buffer)); |
|
if(newMode.excludeBegin){ |
|
this.result.push(lexeme + '<span class="' + newMode.className + '">'); |
|
newMode.buffer = ''; |
|
}else{ |
|
this.result.push('<span class="' + newMode.className + '">'); |
|
newMode.buffer = lexeme; |
|
} |
|
this.modes.push(newMode); |
|
this.relevance += typeof newMode.relevance == "number" ? newMode.relevance : 1; |
|
return; |
|
} |
|
var endLevel = this.endOfMode(lexeme); |
|
if(endLevel){ |
|
mode.buffer += buffer; |
|
if(mode.excludeEnd){ |
|
this.result.push(this.processKeywords(mode.buffer) + '</span>' + lexeme); |
|
}else{ |
|
this.result.push(this.processKeywords(mode.buffer + lexeme) + '</span>'); |
|
} |
|
while(endLevel > 1){ |
|
this.result.push('</span>'); |
|
--endLevel; |
|
this.modes.pop(); |
|
} |
|
this.modes.pop(); |
|
this.modes[this.modes.length - 1].buffer = ''; |
|
return; |
|
} |
|
}, |
|
|
|
highlight: function(value){ |
|
var index = 0; |
|
this.lang.defaultMode.buffer = ''; |
|
do{ |
|
var modeInfo = this.eatModeChunk(value, index); |
|
this.processModeInfo(modeInfo.buffer, modeInfo.lexeme, modeInfo.end); |
|
index += modeInfo.buffer.length + modeInfo.lexeme.length; |
|
}while(!modeInfo.end); |
|
if(this.modes.length > 1){ |
|
throw 'Illegal'; |
|
} |
|
} |
|
}); |
|
|
|
// more utilities |
|
|
|
function replaceText(node, className, text){ |
|
if(String(node.tagName).toLowerCase() == "code" && String(node.parentNode.tagName).toLowerCase() == "pre"){ |
|
// See these 4 lines? This is IE's notion of "node.innerHTML = text". Love this browser :-/ |
|
var container = document.createElement('div'), |
|
environment = node.parentNode.parentNode; |
|
container.innerHTML = '<pre><code class="' + className + '">' + text + '</code></pre>'; |
|
environment.replaceChild(container.firstChild, node.parentNode); |
|
}else{ |
|
node.className = className; |
|
node.innerHTML = text; |
|
} |
|
} |
|
function highlightStringLanguage(language, str){ |
|
var highlight = new Highlighter(language, str); |
|
return {result:highlight.result, langName:language, partialResult:highlight.partialResult}; |
|
} |
|
|
|
function highlightLanguage(block, language){ |
|
var result = highlightStringLanguage(language, blockText(block)); |
|
replaceText(block, block.className, result.result); |
|
} |
|
|
|
function highlightStringAuto(str){ |
|
var result = "", langName = "", bestRelevance = 2, |
|
textBlock = str; |
|
for(var key in dh.languages){ |
|
if(!dh.languages[key].defaultMode){ continue; } // skip internal members |
|
var highlight = new Highlighter(key, textBlock), |
|
relevance = highlight.keywordCount + highlight.relevance, relevanceMax = 0; |
|
if(!result || relevance > relevanceMax){ |
|
relevanceMax = relevance; |
|
result = highlight.result; |
|
langName = highlight.langName; |
|
} |
|
} |
|
return {result:result, langName:langName}; |
|
} |
|
|
|
function highlightAuto(block){ |
|
var result = highlightStringAuto(blockText(block)); |
|
if(result.result){ |
|
replaceText(block, result.langName, result.result); |
|
} |
|
} |
|
|
|
// the public API |
|
|
|
dojox.highlight.processString = function(/* String */ str, /* String? */lang){ |
|
// summary: |
|
// highlight a string of text |
|
// returns: Object |
|
// Object containing: |
|
// |
|
// - result - string of html with spans to apply formatting |
|
// - partialResult - if the formatting failed: string of html |
|
// up to the point of the failure, otherwise: undefined |
|
// - langName - the language used to do the formatting |
|
return lang ? highlightStringLanguage(lang, str) : highlightStringAuto(str); |
|
}; |
|
|
|
dojox.highlight.init = function(/* String|DomNode */ node){ |
|
// summary: |
|
// Highlight a passed node |
|
// description: |
|
// Syntax highlight a passed DomNode or String ID of a DomNode |
|
// example: |
|
// | dojox.highlight.init("someId"); |
|
// |
|
node = dom.byId(node); |
|
if(domClass.contains(node, "no-highlight")){ return; } |
|
if(!verifyText(node)){ return; } |
|
|
|
var classes = node.className.split(/\s+/), |
|
flag = array.some(classes, function(className){ |
|
if(className.charAt(0) != "_" && dh.languages[className]){ |
|
highlightLanguage(node, className); |
|
return true; // stop iterations |
|
} |
|
return false; // continue iterations |
|
}); |
|
if(!flag){ |
|
highlightAuto(node); |
|
} |
|
}; |
|
|
|
dh.Code = function(props, node){ |
|
// summary: |
|
// A class object to allow for dojoType usage with the highlight engine. This is |
|
// NOT a Widget in the conventional sense, and does not have any member functions for |
|
// the instance. This is provided as a convenience. You likely should be calling |
|
// `dojox.highlight.init` directly. |
|
// props: Object? |
|
// Unused. Pass 'null' or {}. Positional usage to allow `dojo.parser` to instantiate |
|
// this class as other Widgets would be. |
|
// node: String|DomNode |
|
// A String ID or DomNode reference to use as the root node of this instance. |
|
// example: |
|
// | <pre><code dojoType="dojox.highlight.Code">for(var i in obj){ ... }</code></pre> |
|
// |
|
// example: |
|
// | var inst = new dojox.highlight.Code({}, "someId"); |
|
// |
|
dh.init(node); |
|
}; |
|
|
|
return dh; |
|
|
|
});
|
|
|