Make Csound Document lexer inherit from RegexLexer instead of XmlLexer

1 files changed, 40 insertions, 34 deletions

diff --git a/pygments/lexers/csound.py b/pygments/lexers/csound.py
index f49cb409..cc1ee644 100644
--- a/pygments/lexers/csound.py
+++ b/pygments/lexers/csound.py
@@ -9,13 +9,13 @@
     :license: BSD, see LICENSE for details.
 """
 
-import re
+import copy, re
 
 from pygments.lexer import RegexLexer, bygroups, default, include, using, words
 from pygments.token import Comment, Keyword, Name, Number, Operator, Punctuation, \
     String, Text
 from pygments.lexers._csound_builtins import OPCODES
-from pygments.lexers.html import HtmlLexer, XmlLexer
+from pygments.lexers.html import HtmlLexer
 from pygments.lexers.python import PythonLexer
 from pygments.lexers.scripting import LuaLexer
 
@@ -311,7 +311,7 @@ class CsoundOrchestraLexer(CsoundLexer):
     }
 
 
-class CsoundDocumentLexer(XmlLexer):
+class CsoundDocumentLexer(RegexLexer):
     """
     For `Csound <http://csound.github.io>`_ documents.
 
@@ -321,34 +321,40 @@ class CsoundDocumentLexer(XmlLexer):
     name = 'Csound Document'
     aliases = ['csound-document', 'csound-csd']
     filenames = ['*.csd']
-    mimetypes = []
-
-    tokens = XmlLexer.tokens
-    for i, item in enumerate(tokens['root']):
-        if len(item) > 2 and item[2] == 'tag':
-            (tokens['root']).insert(i, (r'(<)(\s*)(CsInstruments)(\s*)',
-                                    bygroups(Name.Tag, Text, Name.Tag, Text),
-                                    ('orchestra content', 'tag')))
-            (tokens['root']).insert(i, (r'(<)(\s*)(CsScore)(\s*)',
-                                    bygroups(Name.Tag, Text, Name.Tag, Text),
-                                    ('score content', 'tag')))
-            (tokens['root']).insert(i, (r'(<)(\s*)(html)(\s*)',
-                                    bygroups(Name.Tag, Text, Name.Tag, Text),
-                                    ('HTML', 'tag')))
-            break
-
-    tokens['orchestra content'] = [
-        (r'(<)(\s*)(/)(\s*)(CsInstruments)(\s*)(>)',
-         bygroups(Name.Tag, Text, Name.Tag, Text, Name.Tag, Text, Name.Tag), '#pop'),
-        (r'.+?(?=<\s*/\s*CsInstruments\s*>)', using(CsoundOrchestraLexer))
-    ]
-    tokens['score content'] = [
-        (r'(<)(\s*)(/)(\s*)(CsScore)(\s*)(>)',
-         bygroups(Name.Tag, Text, Name.Tag, Text, Name.Tag, Text, Name.Tag), '#pop'),
-        (r'.+?(?=<\s*/\s*CsScore\s*>)', using(CsoundScoreLexer))
-    ]
-    tokens['HTML'] = [
-        (r'(<)(\s*)(/)(\s*)(html)(\s*)(>)',
-         bygroups(Name.Tag, Text, Name.Tag, Text, Name.Tag, Text, Name.Tag), '#pop'),
-        (r'.+?(?=<\s*/\s*html\s*>)', using(HtmlLexer))
-    ]
+
+    # These tokens are based on those in XmlLexer in pygments/lexers/html.py
+    tokens = {
+        'root': [
+            newline,
+            (r'/[*](.|\n)*?[*]/', Comment.Multiline),
+            (r'[^<&;/]+', Text),
+            (r'<\s*CsInstruments', Name.Tag, ('orchestra', 'tag')),
+            (r'<\s*CsScore', Name.Tag, ('score', 'tag')),
+            (r'<\s*[hH][tT][mM][lL]', Name.Tag, ('HTML', 'tag')),
+            (r'<\s*[\w:.-]+', Name.Tag, 'tag'),
+            (r'<\s*/\s*[\w:.-]+\s*>', Name.Tag)
+        ],
+        'orchestra': [
+            (r'<\s*/\s*CsInstruments\s*>', Name.Tag, '#pop'),
+            (r'(.|\n)+?(?=<\s*/\s*CsInstruments\s*>)', using(CsoundOrchestraLexer))
+        ],
+        'score': [
+            (r'<\s*/\s*CsScore\s*>', Name.Tag, '#pop'),
+            (r'(.|\n)+?(?=<\s*/\s*CsScore\s*>)', using(CsoundScoreLexer))
+        ],
+        'HTML': [
+            (r'<\s*/\s*[hH][tT][mM][lL]\s*>', Name.Tag, '#pop'),
+            (r'(.|\n)+?(?=<\s*/\s*[hH][tT][mM][lL]\s*>)', using(HtmlLexer))
+        ],
+        'tag': [
+            (r'\s+', Text),
+            (r'[\w.:-]+\s*=', Name.Attribute, 'attr'),
+            (r'/?\s*>', Name.Tag, '#pop')
+        ],
+        'attr': [
+            (r'\s+', Text),
+            (r'".*?"', String, '#pop'),
+            (r"'.*?'", String, '#pop'),
+            (r'[^\s>]+', String, '#pop')
+        ]
+    }