Add Treetop lexer

3 files changed, 535 insertions, 1 deletions

diff --git a/pygments/lexers/_mapping.py b/pygments/lexers/_mapping.py
index 0eca2682..e3c45a47 100644
--- a/pygments/lexers/_mapping.py
+++ b/pygments/lexers/_mapping.py
@@ -253,6 +253,7 @@ LEXERS = {
     'TeaTemplateLexer': ('pygments.lexers.templates', 'Tea', ('tea',), ('*.tea',), ('text/x-tea',)),
     'TexLexer': ('pygments.lexers.text', 'TeX', ('tex', 'latex'), ('*.tex', '*.aux', '*.toc'), ('text/x-tex', 'text/x-latex')),
     'TextLexer': ('pygments.lexers.special', 'Text only', ('text',), ('*.txt',), ('text/plain',)),
+    'TreetopLexer': ('pygments.lexers.parsers', 'Treetop', ('treetop',), ('*.treetop', '*.tt'), ()),
     'UrbiscriptLexer': ('pygments.lexers.other', 'UrbiScript', ('urbiscript',), ('*.u',), ('application/x-urbiscript',)),
     'VGLLexer': ('pygments.lexers.other', 'VGL', ('vgl',), ('*.rpf',), ()),
     'ValaLexer': ('pygments.lexers.compiled', 'Vala', ('vala', 'vapi'), ('*.vala', '*.vapi'), ('text/x-vala',)),
diff --git a/pygments/lexers/parsers.py b/pygments/lexers/parsers.py
index 2b5f954f..70e07dc5 100644
--- a/pygments/lexers/parsers.py
+++ b/pygments/lexers/parsers.py
@@ -28,7 +28,8 @@ __all__ = ['RagelLexer', 'RagelEmbeddedLexer', 'RagelCLexer', 'RagelDLexer',
            'AntlrPerlLexer', 'AntlrRubyLexer', 'AntlrCppLexer',
            #'AntlrCLexer',
            'AntlrCSharpLexer', 'AntlrObjectiveCLexer',
-           'AntlrJavaLexer', "AntlrActionScriptLexer"]
+           'AntlrJavaLexer', "AntlrActionScriptLexer",
+           'TreetopLexer']
 
 
 class RagelLexer(RegexLexer):
@@ -693,3 +694,80 @@ class AntlrActionScriptLexer(DelegatingLexer):
     def analyse_text(text):
         return AntlrLexer.analyse_text(text) and \
                re.search(r'^\s*language\s*=\s*ActionScript\s*;', text, re.M)
+
+class TreetopBaseLexer(RegexLexer):
+    """
+    A base lexer for `Treetop <http://treetop.rubyforge.org/>`_ grammars.
+    Not for direct use; use TreetopLexer instead.
+    """
+
+    tokens = {
+        'root': [
+            include('space'),
+            (r'require[ \t]+[^\n\r]+[\n\r]', Other),
+            (r'module\b', Keyword.Namespace, 'module'),
+            (r'grammar\b', Keyword, 'grammar'),
+        ],
+        'module': [
+            include('space'),
+            include('end'),
+            (r'module\b', Keyword, '#push'),
+            (r'grammar\b', Keyword, 'grammar'),
+            (r'[A-Z][A-Za-z_0-9]*(?:::[A-Z][A-Za-z_0-9]*)*', Name.Namespace),
+        ],
+        'grammar': [
+            include('space'),
+            include('end'),
+            (r'rule\b', Keyword, 'rule'),
+            (r'include\b', Keyword, 'include'),
+            (r'[A-Z][A-Za-z_0-9]*', Name),
+        ],
+        'include': [
+            include('space'),
+            (r'[A-Z][A-Za-z_0-9]*(?:::[A-Z][A-Za-z_0-9]*)*', Name.Class, '#pop'),
+        ],
+        'rule': [
+            include('space'),
+            include('end'),
+            (r'"(\\\\|\\"|[^"])*"', String.Double),
+            (r"'(\\\\|\\'|[^'])*'", String.Single),
+            (r'([A-Za-z_][A-Za-z_0-9]*)(:)', bygroups(Name.Label, Punctuation)),
+            (r'[A-Za-z_][A-Za-z_0-9]*', Name),
+            (r'[()]', Punctuation),
+            (r'[?+*/&!~]', Operator),
+            (r'\[(?:\\.|\[:\^?[a-z]+:\]|[^\\\]])+\]', String.Regex),
+            (r'([0-9]*)(\.\.)([0-9]*)', bygroups(Number.Integer, Operator, Number.Integer)),
+            (r'(<)([^>]+)(>)', bygroups(Punctuation, Name.Class, Punctuation)),
+            (r'{', Punctuation, 'inline_module'),
+            (r'\.', String.Regex),
+        ],
+        'inline_module': [
+            (r'{', Other, 'ruby'),
+            (r'}', Punctuation, '#pop'),
+            (r'[^{}]*', Other),
+        ],
+        'ruby': [
+            (r'{', Other, '#push'),
+            (r'}', Other, '#pop'),
+            (r'[^{}]*', Other),
+        ],
+        'space': [
+            (r'[ \t\n\r]+', Whitespace),
+            (r'#[^\n]*', Comment.Single),
+        ],
+        'end': [
+            (r'end\b', Keyword, '#pop'),
+        ],
+    }
+
+class TreetopLexer(DelegatingLexer):
+    """
+    A lexer for `Treetop <http://treetop.rubyforge.org/>`_ grammars.
+    """
+
+    name = 'Treetop'
+    aliases = ['treetop']
+    filenames = ['*.treetop', '*.tt']
+
+    def __init__(self, **options):
+        super(TreetopLexer, self).__init__(RubyLexer, TreetopBaseLexer, **options)
diff --git a/tests/examplefiles/metagrammar.treetop b/tests/examplefiles/metagrammar.treetop
new file mode 100644
index 00000000..acd6af63
--- /dev/null
+++ b/tests/examplefiles/metagrammar.treetop
@@ -0,0 +1,455 @@
+module Treetop
+  module Compiler
+    grammar Metagrammar
+      rule treetop_file
+        requires:(space? require_statement)* prefix:space? module_or_grammar suffix:space? {
+          def compile
+            requires.text_value + prefix.text_value + module_or_grammar.compile + suffix.text_value
+          end
+        }
+      end
+
+      rule require_statement
+        prefix:space? "require" [ \t]+ [^\n\r]+ [\n\r]
+      end
+
+      rule module_or_grammar
+	module_declaration / grammar
+      end
+
+      rule module_declaration
+        prefix:('module' space name:([A-Z] alphanumeric_char* ('::' [A-Z] alphanumeric_char*)*) space) module_contents:(module_declaration / grammar) suffix:(space 'end') {
+          def compile
+            prefix.text_value + module_contents.compile + suffix.text_value
+          end
+
+	  def parser_name
+	    prefix.name.text_value+'::'+module_contents.parser_name
+	  end
+        }
+      end
+
+      rule grammar
+        'grammar' space grammar_name space ('do' space)? declaration_sequence space? 'end' <Grammar>
+      end
+
+      rule grammar_name
+        ([A-Z] alphanumeric_char*)
+      end
+
+      rule declaration_sequence
+        head:declaration tail:(space declaration)* <DeclarationSequence> {
+          def declarations
+            [head] + tail
+          end
+
+          def tail
+            super.elements.map { |elt| elt.declaration }
+          end
+        }
+        /
+        '' {
+          def compile(builder)
+          end
+        }
+      end
+
+      rule declaration
+        parsing_rule / include_declaration
+      end
+
+      rule include_declaration
+        'include' space [A-Z] (alphanumeric_char / '::')* {
+          def compile(builder)
+            builder << text_value
+          end
+        }
+      end
+
+      rule parsing_rule
+        'rule' space nonterminal space ('do' space)? parsing_expression space 'end' <ParsingRule>
+      end
+
+      rule parsing_expression
+        choice / sequence / primary
+      end
+
+      rule choice
+        head:alternative tail:(space? '/' space? alternative)+ <Choice> {
+          def alternatives
+            [head] + tail
+          end
+
+          def tail
+            super.elements.map {|elt| elt.alternative}
+          end
+
+          def inline_modules
+            (alternatives.map {|alt| alt.inline_modules }).flatten
+          end
+        }
+      end
+
+      rule sequence
+        head:labeled_sequence_primary tail:(space labeled_sequence_primary)+ node_class_declarations <Sequence> {
+          def sequence_elements
+            [head] + tail
+          end
+
+          def tail
+            super.elements.map {|elt| elt.labeled_sequence_primary }
+          end
+
+          def inline_modules
+            (sequence_elements.map {|elt| elt.inline_modules}).flatten +
+            [sequence_element_accessor_module] +
+            node_class_declarations.inline_modules
+          end
+
+          def inline_module_name
+            node_class_declarations.inline_module_name
+          end
+        }
+      end
+
+      rule alternative
+        sequence / primary
+      end
+
+      rule primary
+        prefix atomic {
+          def compile(address, builder, parent_expression=nil)
+            prefix.compile(address, builder, self)
+          end
+
+          def prefixed_expression
+            atomic
+          end
+
+          def inline_modules
+            atomic.inline_modules
+          end
+
+          def inline_module_name
+            nil
+          end
+        }
+        /
+	prefix space? predicate_block {
+          def compile(address, builder, parent_expression=nil)
+            prefix.compile(address, builder, self)
+          end
+          def prefixed_expression
+            predicate_block
+          end
+          def inline_modules
+            []
+          end
+        }
+        /
+        atomic suffix node_class_declarations {
+          def compile(address, builder, parent_expression=nil)
+            suffix.compile(address, builder, self)
+          end
+
+          def optional_expression
+            atomic
+          end
+
+          def node_class_name
+            node_class_declarations.node_class_name
+          end
+
+          def inline_modules
+            atomic.inline_modules + node_class_declarations.inline_modules
+          end
+
+          def inline_module_name
+            node_class_declarations.inline_module_name
+          end
+        }
+        /
+        atomic node_class_declarations {
+          def compile(address, builder, parent_expression=nil)
+            atomic.compile(address, builder, self)
+          end
+
+          def node_class_name
+            node_class_declarations.node_class_name
+          end
+
+          def inline_modules
+            atomic.inline_modules + node_class_declarations.inline_modules
+          end
+
+          def inline_module_name
+            node_class_declarations.inline_module_name
+          end
+        }
+      end
+
+      rule labeled_sequence_primary
+        label sequence_primary {
+          def compile(lexical_address, builder)
+            sequence_primary.compile(lexical_address, builder)
+          end
+
+          def inline_modules
+            sequence_primary.inline_modules
+          end
+
+          def label_name
+            if label.name
+              label.name
+            elsif sequence_primary.instance_of?(Nonterminal)
+              sequence_primary.text_value
+            else
+              nil
+            end
+          end
+        }
+      end
+
+      rule label
+        (alpha_char alphanumeric_char*) ':' {
+          def name
+            elements[0].text_value
+          end
+        }
+        /
+        '' {
+          def name
+            nil
+          end
+        }
+      end
+
+      rule sequence_primary
+        prefix atomic {
+          def compile(lexical_address, builder)
+            prefix.compile(lexical_address, builder, self)
+          end
+
+          def prefixed_expression
+            elements[1]
+          end
+
+          def inline_modules
+            atomic.inline_modules
+          end
+
+          def inline_module_name
+            nil
+          end
+        }
+        /
+        prefix space? predicate_block {
+          def compile(address, builder, parent_expression=nil)
+            prefix.compile(address, builder, self)
+          end
+          def prefixed_expression
+            predicate_block
+          end
+          def inline_modules
+            []
+          end
+        }
+        /
+        atomic suffix {
+          def compile(lexical_address, builder)
+            suffix.compile(lexical_address, builder, self)
+          end
+
+          def node_class_name
+            nil
+          end
+
+          def inline_modules
+            atomic.inline_modules
+          end
+
+          def inline_module_name
+            nil
+          end
+        }
+        /
+        atomic
+      end
+
+      rule suffix
+        repetition_suffix / optional_suffix
+      end
+
+      rule optional_suffix
+        '?' <Optional>
+      end
+
+      rule node_class_declarations
+        node_class_expression trailing_inline_module {
+          def node_class_name
+            node_class_expression.node_class_name
+          end
+
+          def inline_modules
+            trailing_inline_module.inline_modules
+          end
+
+          def inline_module
+            trailing_inline_module.inline_module
+          end
+
+          def inline_module_name
+            inline_module.module_name if inline_module
+          end
+        }
+      end
+
+      rule repetition_suffix
+        '+' <OneOrMore> / '*' <ZeroOrMore> / occurrence_range
+      end
+
+      rule occurrence_range
+	space? min:([0-9])* '..' max:([0-9])* <OccurrenceRange>
+      end
+
+      rule prefix
+        '&' <AndPredicate> / '!' <NotPredicate> / '~' <TransientPrefix>
+      end
+
+      rule atomic
+        terminal
+        /
+        nonterminal
+        /
+        parenthesized_expression
+      end
+
+      rule parenthesized_expression
+        '(' space? parsing_expression space? ')' <ParenthesizedExpression> {
+          def inline_modules
+            parsing_expression.inline_modules
+          end
+        }
+      end
+
+      rule nonterminal
+        !keyword_inside_grammar (alpha_char alphanumeric_char*) <Nonterminal>
+      end
+
+      rule terminal
+        quoted_string / character_class / anything_symbol
+      end
+
+      rule quoted_string
+        (single_quoted_string / double_quoted_string) {
+          def string
+            super.text_value
+          end
+        }
+      end
+
+      rule double_quoted_string
+        '"' string:(!'"' ("\\\\" / '\"' / .))* '"' <Terminal>
+      end
+
+      rule single_quoted_string
+        "'" string:(!"'" ("\\\\" / "\\'" / .))* "'" <Terminal>
+      end
+
+      rule character_class
+        '[' characters:(!']' ('\\' . / bracket_expression / !'\\' .))+ ']' <CharacterClass> {
+          def characters
+            super.text_value
+          end
+        }
+      end
+
+      rule bracket_expression
+         '[:' '^'? (
+           'alnum' / 'alpha' / 'blank' / 'cntrl' / 'digit' / 'graph' / 'lower' /
+           'print' / 'punct' / 'space' / 'upper' / 'xdigit' / 'word'
+         ) ':]'
+      end
+
+      rule anything_symbol
+        '.' <AnythingSymbol>
+      end
+
+      rule node_class_expression
+        space '<' (!'>' .)+ '>' {
+          def node_class_name
+            elements[2].text_value
+          end
+        }
+        /
+        '' {
+          def node_class_name
+            nil
+          end
+        }
+      end
+
+      rule trailing_inline_module
+        space inline_module {
+          def inline_modules
+            [inline_module]
+          end
+
+          def inline_module_name
+            inline_module.module_name
+          end
+        }
+        /
+        '' {
+          def inline_modules
+            []
+          end
+
+          def inline_module
+            nil
+          end
+
+          def inline_module_name
+            nil
+          end
+        }
+      end
+
+      rule predicate_block
+        '' inline_module <PredicateBlock>
+      end
+
+      rule inline_module
+        '{' (inline_module / ![{}] .)* '}' <InlineModule>
+      end
+
+      rule keyword_inside_grammar
+        ('rule' / 'end') !non_space_char
+      end
+
+      rule non_space_char
+        !space .
+      end
+
+      rule alpha_char
+        [A-Za-z_]
+      end
+
+      rule alphanumeric_char
+        alpha_char / [0-9]
+      end
+
+      rule space
+        (white / comment_to_eol)+
+      end
+
+      rule comment_to_eol
+        '#' (!"\n" .)*
+      end
+
+      rule white
+        [ \t\n\r]
+      end
+    end
+  end
+end