def __init__(self, **options):
self.builtinshighlighting = get_bool_opt(options, "builtinshighlighting", True)
self.requiredelimiters = get_bool_opt(options, "requiredelimiters", False)
self._builtins = set()
self._members = set()
if self.builtinshighlighting:
from pygments.lexers._lasso_builtins import BUILTINS, MEMBERS
for key, value in iteritems(BUILTINS):
self._builtins.update(value)
for key, value in iteritems(MEMBERS):
self._members.update(value)
RegexLexer.__init__(self, **options)
def get_style_defs(self, arg=''):
"""
Return the command sequences needed to define the commands
used to format text in the verbatim environment. ``arg`` is ignored.
"""
cp = self.commandprefix
styles = []
for name, definition in iteritems(self.cmd2def):
styles.append(r'\expandafter\def\csname %s@tok@%s\endcsname{%s}' %
(cp, name, definition))
return STYLE_TEMPLATE % {'cp': self.commandprefix,
'styles': '\n'.join(styles)}
def get_tokendefs(cls):
"""
Merge tokens from superclasses in MRO order, returning a single tokendef
dictionary.
Any state that is not defined by a subclass will be inherited
automatically. States that *are* defined by subclasses will, by
default, override that state in the superclass. If a subclass wishes to
inherit definitions from a superclass, it can use the special value
"inherit", which will cause the superclass' state definition to be
included at that point in the state.
"""
tokens = {}
inheritable = {}
for c in cls.__mro__:
toks = c.__dict__.get('tokens', {})
for state, items in iteritems(toks):
curitems = tokens.get(state)
if curitems is None:
# N.b. because this is assigned by reference, sufficiently
# deep hierarchies are processed incrementally (e.g. for
# A(B), B(C), C(RegexLexer), B will be premodified so X(B)
# will not see any inherits in B).
tokens[state] = items
try:
inherit_ndx = items.index(inherit)
except ValueError:
continue
inheritable[state] = inherit_ndx
continue
inherit_ndx = inheritable.pop(state, None)
if inherit_ndx is None:
continue
# Replace the "inherit" value with the items
curitems[inherit_ndx:inherit_ndx+1] = items
try:
# N.b. this is the index in items (that is, the superclass
# copy), so offset required when storing below.
new_inh_ndx = items.index(inherit)
except ValueError:
pass
else:
inheritable[state] = inherit_ndx + new_inh_ndx
return tokens
def __init__(self, **options):
self.funcnamehighlighting = get_bool_opt(
options, 'funcnamehighlighting', True)
self.disabledmodules = get_list_opt(
options, 'disabledmodules', ['unknown'])
self.startinline = get_bool_opt(options, 'startinline', False)
# private option argument for the lexer itself
if '_startinline' in options:
self.startinline = options.pop('_startinline')
# collect activated functions in a set
self._functions = set()
if self.funcnamehighlighting:
from pygments.lexers._php_builtins import MODULES
for key, value in iteritems(MODULES):
if key not in self.disabledmodules:
self._functions.update(value)
RegexLexer.__init__(self, **options)
def get_tokendefs(cls):
"""
Merge tokens from superclasses in MRO order, returning a single tokendef
dictionary.
Any state that is not defined by a subclass will be inherited
automatically. States that *are* defined by subclasses will, by
default, override that state in the superclass. If a subclass wishes to
inherit definitions from a superclass, it can use the special value
"inherit", which will cause the superclass' state definition to be
included at that point in the state.
"""
tokens = {}
inheritable = {}
for c in itertools.chain((cls,), cls.__mro__):
toks = c.__dict__.get('tokens', {})
for state, items in iteritems(toks):
curitems = tokens.get(state)
if curitems is None:
tokens[state] = items
try:
inherit_ndx = items.index(inherit)
except ValueError:
continue
inheritable[state] = inherit_ndx
continue
inherit_ndx = inheritable.pop(state, None)
if inherit_ndx is None:
continue
# Replace the "inherit" value with the items
curitems[inherit_ndx:inherit_ndx+1] = items
try:
new_inh_ndx = items.index(inherit)
except ValueError:
pass
else:
inheritable[state] = inherit_ndx + new_inh_ndx
return tokens
def get_style_defs(self, arg=None):
"""
Return CSS style definitions for the classes produced by the current
highlighting style. ``arg`` can be a string or list of selectors to
insert before the token type classes.
"""
if arg is None:
arg = ('cssclass' in self.options and '.'+self.cssclass or '')
if isinstance(arg, string_types):
args = [arg]
else:
args = list(arg)
def prefix(cls):
if cls:
cls = '.' + cls
tmp = []
for arg in args:
tmp.append((arg and arg + ' ' or '') + cls)
return ', '.join(tmp)
styles = [(level, ttype, cls, style)
for cls, (style, ttype, level) in iteritems(self.class2style)
if cls and style]
styles.sort()
lines = ['%s { %s } /* %s */' % (prefix(cls), style, repr(ttype)[6:])
for (level, ttype, cls, style) in styles]
if arg and not self.nobackground and \
self.style.background_color is not None:
text_style = ''
if Text in self.ttype2class:
text_style = ' ' + self.class2style[self.ttype2class[Text]][0]
lines.insert(0, '%s { background: %s;%s }' %
(prefix(''), self.style.background_color, text_style))
if self.style.highlight_color is not None:
lines.insert(0, '%s.hll { background-color: %s }' %
(prefix(''), self.style.highlight_color))
return '\n'.join(lines)
def get_style_defs(self, arg=None):
"""
Return CSS style definitions for the classes produced by the current
highlighting style. ``arg`` can be a string or list of selectors to
insert before the token type classes.
"""
if arg is None:
arg = ('cssclass' in self.options and '.'+self.cssclass or '')
if isinstance(arg, string_types):
args = [arg]
else:
args = list(arg)
def prefix(cls):
if cls:
cls = '.' + cls
tmp = []
for arg in args:
tmp.append((arg and arg + ' ' or '') + cls)
return ', '.join(tmp)
styles = [(level, ttype, cls, style)
for cls, (style, ttype, level) in iteritems(self.class2style)
if cls and style]
styles.sort()
lines = ['%s { %s } /* %s */' % (prefix(cls), style, repr(ttype)[6:])
for (level, ttype, cls, style) in styles]
if arg and not self.nobackground and \
self.style.background_color is not None:
text_style = ''
if Text in self.ttype2class:
text_style = ' ' + self.class2style[self.ttype2class[Text]][0]
lines.insert(0, '%s { background: %s;%s }' %
(prefix(''), self.style.background_color, text_style))
if self.style.highlight_color is not None:
lines.insert(0, '%s.hll { background-color: %s }' %
(prefix(''), self.style.highlight_color))
return '\n'.join(lines)
class PsqlRegexLexer(PostgresBase, RegexLexer):
"""
Extend the PostgresLexer adding support specific for psql commands.
This is not a complete psql lexer yet as it lacks prompt support
and output rendering.
"""
name = 'PostgreSQL console - regexp based lexer'
aliases = [] # not public
flags = re.IGNORECASE
tokens = dict((k, l[:]) for (k, l) in iteritems(PostgresLexer.tokens))
tokens['root'].append((r'\\[^\s]+', Keyword.Pseudo, 'psql-command'))
tokens['psql-command'] = [
(r'\n', Text, 'root'),
(r'\s+', Text),
(r'\\[^\s]+', Keyword.Pseudo),
(r""":(['"]?)[a-z]\w*\b\1""", Name.Variable),
(r"'(''|[^'])*'", String.Single),
(r"`([^`])*`", String.Backtick),
(r"[^\s]+", String.Symbol),
]
class SassLexer(ExtendedRegexLexer):
"""
For Sass stylesheets.
.. versionadded:: 1.3
"""
name = 'Sass'
aliases = ['sass']
filenames = ['*.sass']
mimetypes = ['text/x-sass']
flags = re.IGNORECASE
tokens = {
'root': [
(r'[ \t]*\n', Text),
(r'[ \t]*', _indentation),
],
'content': [
(r'//[^\n]*', _starts_block(Comment.Single,
'single-comment'), 'root'),
(r'/\*[^\n]*', _starts_block(Comment.Multiline,
'multi-comment'), 'root'),
(r'@import', Keyword, 'import'),
(r'@for', Keyword, 'for'),
(r'@(debug|warn|if|while)', Keyword, 'value'),
(r'(@mixin)( [\w-]+)', bygroups(Keyword, Name.Function), 'value'),
(r'(@include)( [\w-]+)', bygroups(Keyword,
Name.Decorator), 'value'),
(r'@extend', Keyword, 'selector'),
(r'@[\w-]+', Keyword, 'selector'),
(r'=[\w-]+', Name.Function, 'value'),
(r'\+[\w-]+', Name.Decorator, 'value'),
(r'([!$][\w-]\w*)([ \t]*(?:(?:\|\|)?=|:))',
bygroups(Name.Variable, Operator), 'value'),
(r':', Name.Attribute, 'old-style-attr'),
(r'(?=.+?[=:]([^a-z]|$))', Name.Attribute, 'new-style-attr'),
default('selector'),
],
'single-comment': [
(r'.+', Comment.Single),
(r'\n', Text, 'root'),
],
'multi-comment': [
(r'.+', Comment.Multiline),
(r'\n', Text, 'root'),
],
'import': [
(r'[ \t]+', Text),
(r'\S+', String),
(r'\n', Text, 'root'),
],
'old-style-attr': [
(r'[^\s:="\[]+', Name.Attribute),
(r'#\{', String.Interpol, 'interpolation'),
(r'[ \t]*=', Operator, 'value'),
default('value'),
],
'new-style-attr': [
(r'[^\s:="\[]+', Name.Attribute),
(r'#\{', String.Interpol, 'interpolation'),
(r'[ \t]*[=:]', Operator, 'value'),
],
'inline-comment': [
(r"(\\#|#(?=[^\n{])|\*(?=[^\n/])|[^\n#*])+", Comment.Multiline),
(r'#\{', String.Interpol, 'interpolation'),
(r"\*/", Comment, '#pop'),
],
}
for group, common in iteritems(common_sass_tokens):
tokens[group] = copy.copy(common)
tokens['value'].append((r'\n', Text, 'root'))
tokens['selector'].append((r'\n', Text, 'root'))
class CSharpLexer(RegexLexer):
"""
For `C# <http://msdn2.microsoft.com/en-us/vcsharp/default.aspx>`_
source code.
Additional options accepted:
`unicodelevel`
Determines which Unicode characters this lexer allows for identifiers.
The possible values are:
* ``none`` -- only the ASCII letters and numbers are allowed. This
is the fastest selection.
* ``basic`` -- all Unicode characters from the specification except
category ``Lo`` are allowed.
* ``full`` -- all Unicode characters as specified in the C# specs
are allowed. Note that this means a considerable slowdown since the
``Lo`` category has more than 40,000 characters in it!
The default value is ``basic``.
.. versionadded:: 0.8
"""
name = 'C#'
aliases = ['csharp', 'c#']
filenames = ['*.cs']
mimetypes = ['text/x-csharp'] # inferred
flags = re.MULTILINE | re.DOTALL | re.UNICODE
# for the range of allowed unicode characters in identifiers, see
# http://www.ecma-international.org/publications/files/ECMA-ST/Ecma-334.pdf
levels = {
'none':
'@?[_a-zA-Z]\w*',
'basic': ('@?[_' + uni.combine('Lu', 'Ll', 'Lt', 'Lm', 'Nl') + ']' +
'[' + uni.combine('Lu', 'Ll', 'Lt', 'Lm', 'Nl', 'Nd', 'Pc',
'Cf', 'Mn', 'Mc') + ']*'),
'full':
('@?(?:_|[^' + uni.allexcept('Lu', 'Ll', 'Lt', 'Lm', 'Lo', 'Nl') +
'])' + '[^' + uni.allexcept('Lu', 'Ll', 'Lt', 'Lm', 'Lo', 'Nl', 'Nd',
'Pc', 'Cf', 'Mn', 'Mc') + ']*'),
}
tokens = {}
token_variants = True
for levelname, cs_ident in iteritems(levels):
tokens[levelname] = {
'root': [
# method names
(
r'^([ \t]*(?:' + cs_ident +
r'(?:\[\])?\s+)+?)' # return type
r'(' + cs_ident + ')' # method name
r'(\s*)(\()', # signature start
bygroups(using(this), Name.Function, Text, Punctuation)),
(r'^\s*\[.*?\]', Name.Attribute),
(r'[^\S\n]+', Text),
(r'\\\n', Text), # line continuation
(r'//.*?\n', Comment.Single),
(r'/[*].*?[*]/', Comment.Multiline),
(r'\n', Text),
(r'[~!%^&*()+=|\[\]:;,.<>/?-]', Punctuation),
(r'[{}]', Punctuation),
(r'@"(""|[^"])*"', String),
(r'"(\\\\|\\"|[^"\n])*["\n]', String),
(r"'\\.'|'[^\\]'", String.Char),
(r"[0-9](\.[0-9]*)?([eE][+-][0-9]+)?"
r"[flFLdD]?|0[xX][0-9a-fA-F]+[Ll]?", Number),
(r'#[ \t]*(if|endif|else|elif|define|undef|'
r'line|error|warning|region|endregion|pragma)\b.*?\n',
Comment.Preproc),
(r'\b(extern)(\s+)(alias)\b', bygroups(Keyword, Text,
Keyword)),
(r'(abstract|as|async|await|base|break|by|case|catch|'
r'checked|const|continue|default|delegate|'
r'do|else|enum|event|explicit|extern|false|finally|'
r'fixed|for|foreach|goto|if|implicit|in|interface|'
r'internal|is|let|lock|new|null|on|operator|'
r'out|override|params|private|protected|public|readonly|'
r'ref|return|sealed|sizeof|stackalloc|static|'
r'switch|this|throw|true|try|typeof|'
r'unchecked|unsafe|virtual|void|while|'
r'get|set|new|partial|yield|add|remove|value|alias|ascending|'
r'descending|from|group|into|orderby|select|thenby|where|'
r'join|equals)\b', Keyword),
(r'(global)(::)', bygroups(Keyword, Punctuation)),
(r'(bool|byte|char|decimal|double|dynamic|float|int|long|object|'
r'sbyte|short|string|uint|ulong|ushort|var)\b\??',
Keyword.Type),
(r'(class|struct)(\s+)', bygroups(Keyword, Text), 'class'),
(r'(namespace|using)(\s+)', bygroups(Keyword,
Text), 'namespace'),
(cs_ident, Name),
],
'class': [
(cs_ident, Name.Class, '#pop'),
default('#pop'),
],
'namespace': [
(r'(?=\()', Text, '#pop'), # using (resource)
('(' + cs_ident + r'|\.)+', Name.Namespace, '#pop'),
]
}
def __init__(self, **options):
level = get_choice_opt(options, 'unicodelevel', list(self.tokens),
'basic')
if level not in self._all_tokens:
# compile the regexes now
self._tokens = self.__class__.process_tokendef(level)
else:
self._tokens = self._all_tokens[level]
RegexLexer.__init__(self, **options)
class NemerleLexer(RegexLexer):
"""
For `Nemerle <http://nemerle.org>`_ source code.
Additional options accepted:
`unicodelevel`
Determines which Unicode characters this lexer allows for identifiers.
The possible values are:
* ``none`` -- only the ASCII letters and numbers are allowed. This
is the fastest selection.
* ``basic`` -- all Unicode characters from the specification except
category ``Lo`` are allowed.
* ``full`` -- all Unicode characters as specified in the C# specs
are allowed. Note that this means a considerable slowdown since the
``Lo`` category has more than 40,000 characters in it!
The default value is ``basic``.
.. versionadded:: 1.5
"""
name = 'Nemerle'
aliases = ['nemerle']
filenames = ['*.n']
mimetypes = ['text/x-nemerle'] # inferred
flags = re.MULTILINE | re.DOTALL | re.UNICODE
# for the range of allowed unicode characters in identifiers, see
# http://www.ecma-international.org/publications/files/ECMA-ST/Ecma-334.pdf
levels = {
'none':
'@?[_a-zA-Z]\w*',
'basic': ('@?[_' + uni.combine('Lu', 'Ll', 'Lt', 'Lm', 'Nl') + ']' +
'[' + uni.combine('Lu', 'Ll', 'Lt', 'Lm', 'Nl', 'Nd', 'Pc',
'Cf', 'Mn', 'Mc') + ']*'),
'full':
('@?(?:_|[^' + uni.allexcept('Lu', 'Ll', 'Lt', 'Lm', 'Lo', 'Nl') +
'])' + '[^' + uni.allexcept('Lu', 'Ll', 'Lt', 'Lm', 'Lo', 'Nl', 'Nd',
'Pc', 'Cf', 'Mn', 'Mc') + ']*'),
}
tokens = {}
token_variants = True
for levelname, cs_ident in iteritems(levels):
tokens[levelname] = {
'root': [
# method names
(
r'^([ \t]*(?:' + cs_ident +
r'(?:\[\])?\s+)+?)' # return type
r'(' + cs_ident + ')' # method name
r'(\s*)(\()', # signature start
bygroups(using(this), Name.Function, Text, Punctuation)),
(r'^\s*\[.*?\]', Name.Attribute),
(r'[^\S\n]+', Text),
(r'\\\n', Text), # line continuation
(r'//.*?\n', Comment.Single),
(r'/[*].*?[*]/', Comment.Multiline),
(r'\n', Text),
(r'\$\s*"', String, 'splice-string'),
(r'\$\s*<#', String, 'splice-string2'),
(r'<#', String, 'recursive-string'),
(r'(<\[)\s*(' + cs_ident + ':)?', Keyword),
(r'\]\>', Keyword),
# quasiquotation only
(r'\$' + cs_ident, Name),
(r'(\$)(\()', bygroups(Name,
Punctuation), 'splice-string-content'),
(r'[~!%^&*()+=|\[\]:;,.<>/?-]', Punctuation),
(r'[{}]', Punctuation),
(r'@"(""|[^"])*"', String),
(r'"(\\\\|\\"|[^"\n])*["\n]', String),
(r"'\\.'|'[^\\]'", String.Char),
(r"0[xX][0-9a-fA-F]+[Ll]?", Number),
(r"[0-9](\.[0-9]*)?([eE][+-][0-9]+)?[flFLdD]?", Number),
(r'#[ \t]*(if|endif|else|elif|define|undef|'
r'line|error|warning|region|endregion|pragma)\b.*?\n',
Comment.Preproc),
(r'\b(extern)(\s+)(alias)\b', bygroups(Keyword, Text,
Keyword)),
(r'(abstract|and|as|base|catch|def|delegate|'
r'enum|event|extern|false|finally|'
r'fun|implements|interface|internal|'
r'is|macro|match|matches|module|mutable|new|'
r'null|out|override|params|partial|private|'
r'protected|public|ref|sealed|static|'
r'syntax|this|throw|true|try|type|typeof|'
r'virtual|volatile|when|where|with|'
r'assert|assert2|async|break|checked|continue|do|else|'
r'ensures|for|foreach|if|late|lock|new|nolate|'
r'otherwise|regexp|repeat|requires|return|surroundwith|'
r'unchecked|unless|using|while|yield)\b', Keyword),
(r'(global)(::)', bygroups(Keyword, Punctuation)),
(r'(bool|byte|char|decimal|double|float|int|long|object|sbyte|'
r'short|string|uint|ulong|ushort|void|array|list)\b\??',
Keyword.Type),
(r'(:>?)\s*(' + cs_ident + r'\??)',
bygroups(Punctuation, Keyword.Type)),
(r'(class|struct|variant|module)(\s+)',
bygroups(Keyword, Text), 'class'),
(r'(namespace|using)(\s+)', bygroups(Keyword,
Text), 'namespace'),
(cs_ident, Name),
],
'class': [(cs_ident, Name.Class, '#pop')],
'namespace': [
(r'(?=\()', Text, '#pop'), # using (resource)
('(' + cs_ident + r'|\.)+', Name.Namespace, '#pop')
],
'splice-string':
[(r'[^"$]', String), (r'\$' + cs_ident, Name),
(r'(\$)(\()', bygroups(Name,
Punctuation), 'splice-string-content'),
(r'\\"', String), (r'"', String, '#pop')],
'splice-string2':
[(r'[^#<>$]', String), (r'\$' + cs_ident, Name),
(r'(\$)(\()', bygroups(Name,
Punctuation), 'splice-string-content'),
(r'<#', String, '#push'), (r'#>', String, '#pop')],
'recursive-string': [(r'[^#<>]', String), (r'<#', String, '#push'),
(r'#>', String, '#pop')],
'splice-string-content':
[(r'if|match', Keyword),
(r'[~!%^&*+=|\[\]:;,.<>/?-\\"$ ]', Punctuation), (cs_ident, Name),
(r'\d+', Number), (r'\(', Punctuation, '#push'),
(r'\)', Punctuation, '#pop')]
}
def __init__(self, **options):
level = get_choice_opt(options, 'unicodelevel', list(self.tokens),
'basic')
if level not in self._all_tokens:
# compile the regexes now
self._tokens = self.__class__.process_tokendef(level)
else:
self._tokens = self._all_tokens[level]
RegexLexer.__init__(self, **options)
