Merge pull request #808 from Golmote/prism-parser

Add support for Parser

components.js

@@ -341,6 +341,11 @@ var components = {
 			"title": "PARI/GP",
 			"owner": "Golmote"
 		},
+		"parser": {
+			"title": "Parser",
+			"require": "markup",
+			"owner": "Golmote"
+		},
 		"pascal": {
 			"title": "Pascal",
 			"owner": "Golmote"

components/prism-markup.js

@@ -4,7 +4,7 @@ Prism.languages.markup = {
 	'doctype': /<!DOCTYPE[\w\W]+?>/,
 	'cdata': /<!\[CDATA\[[\w\W]*?]]>/i,
 	'tag': {
-		pattern: /<\/?[^\s>\/=.]+(?:\s+[^\s>\/=]+(?:=(?:("|')(?:\\\1|\\?(?!\1)[\w\W])*\1|[^\s'">=]+))?)*\s*\/?>/i,
+		pattern: /<\/?(?!\d)[^\s>\/=.$<]+(?:\s+[^\s>\/=]+(?:=(?:("|')(?:\\\1|\\?(?!\1)[\w\W])*\1|[^\s'">=]+))?)*\s*\/?>/i,
 		inside: {
 			'tag': {
 				pattern: /^<\/?[^\s>\/]+/i,

components/prism-parser.js

@@ -0,0 +1,66 @@
+Prism.languages.parser = Prism.languages.extend('markup', {
+	'keyword': {
+		pattern: /(^|[^^])(?:\^(?:case|eval|for|if|switch|throw)\b|@(?:BASE|CLASS|GET(?:_DEFAULT)?|OPTIONS|SET_DEFAULT|USE)\b)/,
+		lookbehind: true
+	},
+	'variable': {
+		pattern: /(^|[^^])\B\$(?:\w+|(?=[.\{]))(?:(?:\.|::?)\w+)*(?:\.|::?)?/,
+		lookbehind: true,
+		inside: {
+			'punctuation': /\.|:+/
+		}
+	},
+	'function': {
+		pattern: /(^|[^^])\B[@^]\w+(?:(?:\.|::?)\w+)*(?:\.|::?)?/,
+		lookbehind: true,
+		inside: {
+			'keyword': {
+				pattern: /(^@)(?:GET_|SET_)/,
+				lookbehind: true
+			},
+			'punctuation': /\.|:+/
+		}
+	},
+	'escape': {
+		pattern: /\^(?:[$^;@()\[\]{}"':]|#[a-f\d]*)/i,
+		alias: 'builtin'
+	},
+	'punctuation': /[\[\](){};]/
+});
+Prism.languages.insertBefore('parser', 'keyword', {
+	'parser-comment': {
+		pattern: /(\s)#.*/,
+		lookbehind: true,
+		alias: 'comment'
+	},
+	'expression': {
+		// Allow for 3 levels of depth
+		pattern: /(^|[^^])\((?:[^()]|\((?:[^()]|\((?:[^()])*\))*\))*\)/,
+		lookbehind: true,
+		inside: {
+			'string': {
+				pattern: /(^|[^^])(["'])(?:(?!\2)[^^]|\^[\s\S])*\2/,
+				lookbehind: true
+			},
+			'keyword': Prism.languages.parser.keyword,
+			'variable': Prism.languages.parser.variable,
+			'function': Prism.languages.parser.function,
+			'boolean': /\b(?:true|false)\b/,
+			'number': /\b(?:0x[a-f\d]+|\d+\.?\d*(?:e[+-]?\d+)?)\b/i,
+			'escape': Prism.languages.parser.escape,
+			'operator': /[~+*\/\\%]|!(?:\|\|?|=)?|&&?|\|\|?|==|<[<=]?|>[>=]?|-[fd]?|\b(?:def|eq|ge|gt|in|is|le|lt|ne)\b/,
+			'punctuation': Prism.languages.parser.punctuation
+		}
+	}
+});
+Prism.languages.insertBefore('inside', 'punctuation', {
+	'expression': Prism.languages.parser.expression,
+	'keyword': Prism.languages.parser.keyword,
+	'variable': Prism.languages.parser.variable,
+	'function': Prism.languages.parser.function,
+	'escape': Prism.languages.parser.escape,
+	'parser-punctuation': {
+		pattern: Prism.languages.parser.punctuation,
+		alias: 'punctuation'
+	}
+}, Prism.languages.parser['tag'].inside['attr-value']);

examples/prism-parser.html

@@ -0,0 +1,97 @@
+<h1>Parser</h1>
+<p>To use this language, use the class "language-parser".</p>
+
+<h2>Comments</h2>
+<pre><code>$foo[bar] # Some comment</code></pre>
+
+<h2>Variables and functions</h2>
+<pre><code>@navigation[]
+$sections[^table::load[sections.cfg]]
+$sections.uri</code></pre>
+
+<h2>Literals</h2>
+<pre><code>$foo(3+$bar)
+^switch[$sMode]{
+	^case[def]{$result(true)}
+}
+^if(in "/news/"){}</code></pre>
+
+<h2>Escape sequences</h2>
+<pre><code>^^
+^"
+^;</code></pre>
+
+<h2>Embedded in markup</h2>
+<pre><code>&lt;nav>
+	&lt;ul>
+	^sections.menu{
+		&lt;li>
+			&lt;a href="$sections.uri">$sections.name&lt;/a>
+		&lt;/li>
+	}
+	&lt;/ul>
+&lt;/nav></code></pre>
+
+<h2>Full example</h2>
+<pre><code>@CLASS
+MyTable
+
+@create[uParam]
+^switch[$uParam.CLASS_NAME]{
+   ^case[string;void]{$t[^table::create{$uParam}]}
+   ^case[table;MyTable]{$t[^table::create[$uParam]]}
+   ^case[DEFAULT]{^throw[MyTable;Unsupported type $uParam.CLASS_NAME]}
+}
+
+# method will return value in different calling contexts
+@GET[sMode]
+^switch[$sMode]{
+   ^case[table]{$result[$t]}
+   ^case[bool]{$result($t!=0)}
+   ^case[def]{$result(true)}
+   ^case[expression;double]{$result($t)}
+   ^case[DEFAULT]{^throw[MyTable;Unsupported mode '$sMode']}
+}
+
+
+# method will handle access to the "columns"
+@GET_DEFAULT[sName]
+$result[$t.$sName]
+
+
+# wrappers for all existing methods are required
+@count[]
+^t.count[]
+
+@menu[jCode;sSeparator]
+^t.menu{$jCode}[$sSeparator]
+
+
+# new functionality
+@remove[iOffset;iLimit]
+$iLimit(^iLimit.int(0))
+$t[^t.select(^t.offset[]<$iOffset || ^t.offset[]>=$iOffset+$iLimit)]</code></pre>
+
+<h2>Known failures</h2>
+<p>There are certain edge cases where Prism will fail.
+	There are always such cases in every regex-based syntax highlighter.
+	However, Prism dares to be open and honest about them.
+	If a failure is listed here, it doesn’t mean it will never be fixed. This is more of a “known bugs” list, just with a certain type of bug.
+</p>
+
+<h3>Comment-like substrings</h3>
+<pre><code>("fo #o")</code></pre>
+
+<h3>Tag-like substrings</h3>
+<pre><code>("fo&lt;div>o")</code></pre>
+
+<h3>Code block starting with a comment</h3>
+<pre><code># Doesn't work
+# Does work</code></pre>
+<pre><code> # Does work when prefixed with a space</code></pre>
+
+<h3>Comments inside expressions break literals and operators</h3>
+<pre><code>^if(
+    $age>=4  # not too young
+    && $age<=80  # and not too old
+)</code></pre>

plugins/autoloader/prism-autoloader.js

@@ -4,7 +4,7 @@
 	}
 
 	// The dependencies map is built automatically with gulp
-	var lang_dependencies = /*languages_placeholder[*/{"javascript":"clike","actionscript":"javascript","aspnet":"markup","bison":"c","c":"clike","csharp":"clike","cpp":"c","coffeescript":"javascript","crystal":"ruby","css-extras":"css","d":"clike","dart":"clike","fsharp":"clike","glsl":"clike","go":"clike","groovy":"clike","haml":"ruby","handlebars":"markup","jade":"javascript","java":"clike","less":"css","markdown":"markup","nginx":"clike","objectivec":"c","php":"clike","php-extras":"php","processing":"clike","qore":"clike","jsx":["markup","javascript"],"ruby":"clike","sass":"css","scss":"css","scala":"java","smarty":"markup","swift":"clike","textile":"markup","twig":"markup","typescript":"javascript","wiki":"markup"}/*]*/;
+	var lang_dependencies = /*languages_placeholder[*/{"javascript":"clike","actionscript":"javascript","aspnet":"markup","bison":"c","c":"clike","csharp":"clike","cpp":"c","coffeescript":"javascript","crystal":"ruby","css-extras":"css","d":"clike","dart":"clike","fsharp":"clike","glsl":"clike","go":"clike","groovy":"clike","haml":"ruby","handlebars":"markup","jade":"javascript","java":"clike","less":"css","markdown":"markup","nginx":"clike","objectivec":"c","parser":"markup","php":"clike","php-extras":"php","processing":"clike","qore":"clike","jsx":["markup","javascript"],"ruby":"clike","sass":"css","scss":"css","scala":"java","smarty":"markup","swift":"clike","textile":"markup","twig":"markup","typescript":"javascript","wiki":"markup"}/*]*/;
 
 	var lang_data = {};
 

prism.js

@@ -449,7 +449,7 @@ Prism.languages.markup = {
 	'doctype': /<!DOCTYPE[\w\W]+?>/,
 	'cdata': /<!\[CDATA\[[\w\W]*?]]>/i,
 	'tag': {
-		pattern: /<\/?[^\s>\/=.]+(?:\s+[^\s>\/=]+(?:=(?:("|')(?:\\\1|\\?(?!\1)[\w\W])*\1|[^\s'">=]+))?)*\s*\/?>/i,
+		pattern: /<\/?(?!\d)[^\s>\/=.$<]+(?:\s+[^\s>\/=]+(?:=(?:("|')(?:\\\1|\\?(?!\1)[\w\W])*\1|[^\s'">=]+))?)*\s*\/?>/i,
 		inside: {
 			'tag': {
 				pattern: /^<\/?[^\s>\/]+/i,

tests/languages/parser/boolean_feature.test

@@ -0,0 +1,21 @@
+(true)
+(false)
+
+----------------------------------------------------
+
+[
+	["expression", [
+		["punctuation", "("],
+		["boolean", "true"],
+		["punctuation", ")"]
+	]],
+	["expression", [
+		["punctuation", "("],
+		["boolean", "false"],
+		["punctuation", ")"]
+	]]
+]
+
+----------------------------------------------------
+
+Checks for booleans inside expressions.

tests/languages/parser/escape_feature.test

@@ -0,0 +1,68 @@
+^$
+^^
+^;
+^@
+^(
+^)
+^[
+^]
+^{
+^}
+^"
+^'
+^:
+^#
+^#20
+^#af
+^#AF
+
+^^date::now
+^$foobar
+
+<div class="foo^^bar">
+
+----------------------------------------------------
+
+[
+	["escape", "^$"],
+	["escape", "^^"],
+	["escape", "^;"],
+	["escape", "^@"],
+	["escape", "^("],
+	["escape", "^)"],
+	["escape", "^["],
+	["escape", "^]"],
+	["escape", "^{"],
+	["escape", "^}"],
+	["escape", "^\""],
+	["escape", "^'"],
+	["escape", "^:"],
+	["escape", "^#"],
+	["escape", "^#20"],
+	["escape", "^#af"],
+	["escape", "^#AF"],
+
+	["escape", "^^"], "date::now\r\n",
+	["escape", "^$"], "foobar\r\n\r\n",
+
+	["tag", [
+		["tag", [
+			["punctuation", "<"],
+			"div"
+		]],
+		["attr-name", ["class"]],
+		["attr-value", [
+			["punctuation", "="],
+			["punctuation", "\""],
+			"foo",
+			["escape", "^^"],
+			"bar",
+			["punctuation", "\""]
+		]],
+		["punctuation", ">"]
+	]]
+]
+
+----------------------------------------------------
+
+Checks for escapes.

tests/languages/parser/expression_feature.test

@@ -0,0 +1,58 @@
+((3-(9-2))*4)
+^eval(4+2)
+
+<div class="foo-^eval(4+2)">
+
+----------------------------------------------------
+
+[
+	["expression", [
+		["punctuation", "("],
+		["punctuation", "("],
+		["number", "3"],
+		["operator", "-"],
+		["punctuation", "("],
+		["number", "9"],
+		["operator", "-"],
+		["number", "2"],
+		["punctuation", ")"],
+		["punctuation", ")"],
+		["operator", "*"],
+		["number", "4"],
+		["punctuation", ")"]
+	]],
+	["keyword", "^eval"],
+
+	["expression", [
+		["punctuation", "("],
+		["number", "4"],
+		["operator", "+"],
+		["number", "2"],
+		["punctuation", ")"]
+	]],
+
+	["tag", [
+		["tag", [
+			["punctuation", "<"],
+			"div"
+		]],
+		["attr-name", ["class"]],
+		["attr-value", [
+			["punctuation", "="],
+			["punctuation", "\""],
+			"foo-",
+			["keyword", "^eval"],
+			["expression", [
+				["punctuation", "("],
+				["number", "4"], ["operator", "+"], ["number", "2"],
+				["punctuation", ")"]
+			]],
+			["punctuation", "\""]
+		]],
+		["punctuation", ">"]
+	]]
+]
+
+----------------------------------------------------
+
+Checks for expressions, up to 3 levels of depth.