From 52fd55ee6ac8cfd01bdecaa63a47554810c8b8fa Mon Sep 17 00:00:00 2001
From: Golmote <golmote@hotmail.com>
Date: Fri, 12 Jun 2015 20:20:41 +0200
Subject: [PATCH] Relax markup grammar to allow non-ASCII chars in tag names
 and attributes. Fix #585

---
 components/prism-markup.js     | 10 +++++-----
 components/prism-markup.min.js |  2 +-
 examples/prism-markup.html     |  7 ++++++-
 prism.js                       | 10 +++++-----
 4 files changed, 17 insertions(+), 12 deletions(-)

diff --git a/components/prism-markup.js b/components/prism-markup.js
index f4ab4f3651..febdc98171 100644
--- a/components/prism-markup.js
+++ b/components/prism-markup.js
@@ -4,13 +4,13 @@ Prism.languages.markup = {
 	'doctype': /<!DOCTYPE.+?>/,
 	'cdata': /<!\[CDATA\[[\w\W]*?]]>/i,
 	'tag': {
-		pattern: /<\/?[\w:-]+\s*(?:\s+[\w:-]+(?:=(?:("|')(\\?[\w\W])*?\1|[^\s'">=]+))?\s*)*\/?>/i,
+		pattern: /<\/?[^\s>\/]+\s*(?:\s+[^\s>\/=]+(?:=(?:("|')(\\?[\w\W])*?\1|[^\s'">=]+))?\s*)*\/?>/i,
 		inside: {
 			'tag': {
-				pattern: /^<\/?[\w:-]+/i,
+				pattern: /^<\/?[^\s>\/]+/i,
 				inside: {
 					'punctuation': /^<\/?/,
-					'namespace': /^[\w-]+?:/
+					'namespace': /^[^\s>\/:]+:/
 				}
 			},
 			'attr-value': {
@@ -21,9 +21,9 @@ Prism.languages.markup = {
 			},
 			'punctuation': /\/?>/,
 			'attr-name': {
-				pattern: /[\w:-]+/,
+				pattern: /[^\s>\/]+/,
 				inside: {
-					'namespace': /^[\w-]+?:/
+					'namespace': /^[^\s>\/:]+:/
 				}
 			}
 
diff --git a/components/prism-markup.min.js b/components/prism-markup.min.js
index fbf30552b9..dece1fb7eb 100644
--- a/components/prism-markup.min.js
+++ b/components/prism-markup.min.js
@@ -1 +1 @@
-Prism.languages.markup={comment:/<!--[\w\W]*?-->/,prolog:/<\?.+?\?>/,doctype:/<!DOCTYPE.+?>/,cdata:/<!\[CDATA\[[\w\W]*?]]>/i,tag:{pattern:/<\/?[\w:-]+\s*(?:\s+[\w:-]+(?:=(?:("|')(\\?[\w\W])*?\1|[^\s'">=]+))?\s*)*\/?>/i,inside:{tag:{pattern:/^<\/?[\w:-]+/i,inside:{punctuation:/^<\/?/,namespace:/^[\w-]+?:/}},"attr-value":{pattern:/=(?:('|")[\w\W]*?(\1)|[^\s>]+)/i,inside:{punctuation:/=|>|"/}},punctuation:/\/?>/,"attr-name":{pattern:/[\w:-]+/,inside:{namespace:/^[\w-]+?:/}}}},entity:/&#?[\da-z]{1,8};/i},Prism.hooks.add("wrap",function(t){"entity"===t.type&&(t.attributes.title=t.content.replace(/&amp;/,"&"))});
\ No newline at end of file
+Prism.languages.markup={comment:/<!--[\w\W]*?-->/,prolog:/<\?.+?\?>/,doctype:/<!DOCTYPE.+?>/,cdata:/<!\[CDATA\[[\w\W]*?]]>/i,tag:{pattern:/<\/?[^\s>\/]+\s*(?:\s+[^\s>\/=]+(?:=(?:("|')(\\?[\w\W])*?\1|[^\s'">=]+))?\s*)*\/?>/i,inside:{tag:{pattern:/^<\/?[^\s>\/]+/i,inside:{punctuation:/^<\/?/,namespace:/^[^\s>\/:]+:/}},"attr-value":{pattern:/=(?:('|")[\w\W]*?(\1)|[^\s>]+)/i,inside:{punctuation:/=|>|"/}},punctuation:/\/?>/,"attr-name":{pattern:/[^\s>\/]+/,inside:{namespace:/^[^\s>\/:]+:/}}}},entity:/&#?[\da-z]{1,8};/i},Prism.hooks.add("wrap",function(t){"entity"===t.type&&(t.attributes.title=t.content.replace(/&amp;/,"&"))});
\ No newline at end of file
diff --git a/examples/prism-markup.html b/examples/prism-markup.html
index d3a333c76d..7709c025fd 100644
--- a/examples/prism-markup.html
+++ b/examples/prism-markup.html
@@ -72,4 +72,9 @@ <h2>Invalid HTML</h2>
 <h2>Multi-line attribute values</h2>
 <pre><code>&lt;p title="foo
 bar
-baz"></code></pre>
\ No newline at end of file
+baz"></code></pre>
+
+<h2>XML tags with non-ASCII characters</h2>
+<pre><code>&lt;L&auml;ufer&gt;foo&lt;/L&auml;ufer&gt;
+&lt;tag l&auml;ufer="l&auml;ufer"&gt;bar&lt;/tag&gt;
+&lt;l&auml;ufer:tag&gt;baz&lt;/l&auml;ufer:tag&gt;</code></pre>
\ No newline at end of file
diff --git a/prism.js b/prism.js
index e74f7cd746..ace59c0437 100644
--- a/prism.js
+++ b/prism.js
@@ -441,13 +441,13 @@ Prism.languages.markup = {
 	'doctype': /<!DOCTYPE.+?>/,
 	'cdata': /<!\[CDATA\[[\w\W]*?]]>/i,
 	'tag': {
-		pattern: /<\/?[\w:-]+\s*(?:\s+[\w:-]+(?:=(?:("|')(\\?[\w\W])*?\1|[^\s'">=]+))?\s*)*\/?>/i,
+		pattern: /<\/?[^\s>\/]+\s*(?:\s+[^\s>\/=]+(?:=(?:("|')(\\?[\w\W])*?\1|[^\s'">=]+))?\s*)*\/?>/i,
 		inside: {
 			'tag': {
-				pattern: /^<\/?[\w:-]+/i,
+				pattern: /^<\/?[^\s>\/]+/i,
 				inside: {
 					'punctuation': /^<\/?/,
-					'namespace': /^[\w-]+?:/
+					'namespace': /^[^\s>\/:]+:/
 				}
 			},
 			'attr-value': {
@@ -458,9 +458,9 @@ Prism.languages.markup = {
 			},
 			'punctuation': /\/?>/,
 			'attr-name': {
-				pattern: /[\w:-]+/,
+				pattern: /[^\s>\/]+/,
 				inside: {
-					'namespace': /^[\w-]+?:/
+					'namespace': /^[^\s>\/:]+:/
 				}
 			}